Serratia Species Isolated from Bovine lntramammary Infections’ D. A. TODHUNTER, K. L. SMITH, and J. S. HOGAN
Depammt of Dalry Science
Ohio Agricultural Research and Development Center wooster 44691 ABSTRACT
INTRODUCTION
Jntramammary infections from which Serratia spp. were isolated were studied over a 32-mo period in a research dairy herd consisting of approximately 120 lactating cows. A total of 29 Serratia spp. intramammary infections were detected and accounted for 9% of all Gram-negative bacterial intramamnmy infections. Serratia marcescens was the most common Serratia spp. isolated. origin of intramammary infections was 48.3% during the frrst half of the dry period, 31% during the last half of the dry period, and 20.7% during lactation. A total Of 64% of -ni infwtions that were first detected during the first half of the dry period persisted to calving. Geometric mean number of lactation days infected for all infections was 55. Intramarmnary infections that originated during the first half of the dry period were present in lactation for a significantly greater number of days compared with intramammary infections new during the last half of the dry period or lactation. A total of 48% of infections were clinical. Serratia spp. intramammary infections tended to be of long duration compared with other Gramnegative bacterial intramammary infections and were highly associated with the dry period (Key words: mastitis, Serratia spp., dry period)
Serratia marcescens and Serratia liquefaciens have been reported as causative agents of bovine mastitis (1, 2, 6, 13). Many of the investigations on Serratia spp. -ni infections (MI) were described as outbreaks or episodic (1, 2, 13) and linked to contaminated water or teat dips (4, 5, 12). The purposes of this study were to identify Serratia spp. isolated from naturally occurring IMI in a total confinement herd and to determine various infection parameters. such as origin of IMI and lactation days infected. MATERIALS AND METHODS Animals
me
Abbreviation key: IMI = intramammary infection
Received October 29, 1990. Accepted January 15, 1991. ‘Salaries and mearch support provided by state and federal fands appmpriated to tbe Ohio Agricultural R e Search and DeveloptIIUIt C~ntcr,Thc Ohio State Ulrivcrsi@. Mamucript Npmba 282-90. 1991 J Dairy Sci 74:1860-1865
Experimental animals were Holstein (80%) and Jersey (20%) dairy cattle from the research herd of the Ohio Agn’lcultural Research and Development Center, Wooster. Herd size was approximately 120 lactating cows. Cows were milked in a double 3, side opening Surge (Babson Bros. Co.,Oak Brook, IL) parlor. AU teats were predipped and postdipped with 4% sodium hypochlorite Clorox Co., Oakland, CA). Housing was total confinement, with appmximately 75% of the lactating herd in free stalls and 25% in tie stalls. Bedding material in free stalls was ~ n treated, noncomposted recycled manure Solids Separator, Babson Bros. Co., Oak Brodr, IL). Wood shavings were used as bedding material in tie stalls. Dry cows were housed in free stalls and bedded with recycled manure. Cows were dried off appmxjmately 60 d prior to anticipated calving date by abrupt cessation of milking. AU quarters of all cows were infused with a commercially available antibiotic preparaxion (300 mg cephapirin benzathine; Tomorrow, Agricultural products, Division of Bristol-Myers Co.,DeWitt, NY). Dry cow therapy was administered by mastitis laboratory personnel using aseptic techniques.
1860
-
Cows were moved to individual box stalls 3 to 7 d prior to parturition. Bedding material in box stalls was pelleted corn cobs (The Anderson co.,Marnee, OH). cows remained in box stalls for 3 d after calving and then were moved to either free or tie stalls. Management of cows was as &scribed by Smith et aL (8). Quarter Foremilk Sampling Schedule
Quarter foremilk samples for microbiological culture were obtained from all cows at the following time points: 14 and 7 d prior to drying off; the day of drying off and prior to antibiotic infusion; 30 d of involution; calving; 3, 7, 14, and 30 d postcalving; and at 3O-d intervals for the remainder of lactation. Single quarter milk samples were taken at all time points, except at 30 d of involution when samples were taken in duplicate. Sampling procedm was as previously described (8). Samples were also obtained from laming COWS reported with clinical -titis. A mmary quarter was classified as clinical when the secretion displayed signs of abnormality (flakes, clots, etc.) at the time of sampling. Cows with clinical mastitis were sampled for 2 consecutive d and 7 and 14 d after initial sampling. Cows with clinical mastitis that received antibiotic therapy were sampled 1 d before treatment, the day of treatment prior to antibiotic infusion, and 7 and 14 d posttreatment. Antibiotic treatment consisted of two in@infusions of 200 mg cephapirin sodium (TODAY, Agricultural P d u c t s , Division of Sristol-Myers Co., Syracuse, NY) in the clinical quarter at a 12-h interval. Additional samples were obtained from cows that were on other experimental trials or to confirm infection status of quarters. Data summarization was from all cows sampled January 1986 through August 1989.
streaked with .I ml of milk Plates were incubated at 37’C and examined for bacterial p w t h at 24 and 48 h of incubation. colonies tentatively identified as Gram-negative bacteria were Gram stained. Gram-negative bacteria were streaked for isolation on blood agar and identified with the API2OE System (Analytab products, plainview, NY). The API2OE system identification was performed according to manufacturer’s instructions for an 18- to 24-h incubation. Dlagnosla, Origln by Stage of Lactatton, and Duratlon of Infections
A Serratia spp. IMI was diagnosed under one of the following conditions: 1) single isolation of a Serratia spp. fmm a clinical quarter,
2) isolation of a Serratia spp. from both of duplicate quarter samples, or 3) isolation of a Serratia spp. in two out of three consecutive samples with less than 31 d between isolations. An exception to the 30 d between isolations was made when the same Serratia spp. was
isolated at 30 d of involution and again at calving and greater than 30 d had elapsed between calving and the 3 0 4 dry period sample. A Serratia spp. IMI was classified as a mixed infection when another pathogen was simultaneously isolated fmm the quarter and met the guidelines described above for IMI diagnosis. Duration of an IMI was the interval of time in days between the first and last isolation plus 1 d A single isolation of Serratia spp. from a clinical quarter had a duration of 1 d. Lactation days infected were defined as the number of days during lactation that an IMI was present. The day of calving was considered as d 1 of lactation. Lactation days infected were transformed to 1ogl@ An M I was diagnosed as clinical if at least one of the samples was associated with clinical signs during the duration of the IML. PercentMlcroblologlcai Procedures age of samples fmm an IMI displaying clinical AU milk samples were examined microbio- signs was also calculated by dividing the total logically (8) by streaking .01 ml milk onto one number of clinical samples by the total number quadrant of a trypticase soy agar plate (BBL of samples obtained from the IMI. Percentage Microbiology Systems, Becton Diclcinson Co., of samples positive for Serratia spp. during the Cockeysville, MD) containing 5 % bovine duration of an IMI was calculated by dividing blood and .l%esculin (Sigma Chemical Co., number of samples from which a Serrafiu spp. St. Louis, MO). One-half of a Maccoakey was isolated by the total number of samples agar (BBL Microbiology Systems) plate was obtained from the IMI. Jomnal of Dairy Science Vol. 74. No. 6, 1991
1862
35 30
TODHUNTER ET AL.
1
7
9
-
5
U
2
25-
8
4
u
0
L 3 0
P $ 2
z
1 0
JAN FEE YAR AF'R MAY JUN JUL AUG SEP OCT NOV DEC
Month
1
2
4
Lactation Number
25
Figure 1. Origin Of Serrati0 ~ p pintnrmammary . infections 0 by lactation number. Values are expressed as percentage of total IMI.Numbers above bars are number of IMI for indicated lactation number.
Figure 2.Distribution of first detection of Smutiu spp. n it-y infections by month. values are expressed as number of IMI per month for indicated year.
RESULTS
Serratia spp. were isolated from 29 IMI in 25 cows (6 Jersey and 19 Holstein) over the 32-mo study period. Distribution of lactation number of io& with Serratia spp. IMI is in A Serratia spp. IMI first detected at 30 d of Figure 1. Approximately 31% of IMI were involution and not present in the three samples detected in cows in fifth lactation or greater. A obtained prior to drying off was considered to total of four or 13.8% of Serratia spp. IMI have originated during the first half of the dry were detected in heifers with two IMI first period. Intramammary infections that were de- detected at calving. Percentage distribution of tected in the first half of the dry period and a l l cows in the herd by lactation number was again at calving were classified as having per- first lactation 34.2%; second lactation, 25.6%; sisted to calving. Infections first detected at third lactation, 16.8%; fourth lactation, 11.7%; calving or within 3 d after calving were classi- and fifth lactation or greater, 11.7%. Mammary fied as having originated during the last half of quarter distribution of IMI was 34.5% in front the dry period. Serratia spp. IMI first detected quarters and 65.5% in rear quarters. between 4 d of lactation and drying offwere Number of new Serratia spp. IMI detected considered to have originated during lactation. for each year of the study was 6 during 1986, The most commonly isolated species of 11 during 1987, 6 during 1988, and 6 from Serratia, as identified by the APEOE system January 1989 through August 1989. First defor a particular IMI, was considered the causa- tection of Serratia spp. I M I were distributed tive organism. An IMI was classified as Serra- throughout several months of the year for each tia spp. when no one species predominated or year of the study (Figure 2). Serratia marcescens was the species most when isolations were i d e n a d only at the fresuently identified. Species identified and genus level. their frequency are in Table 1. Isolates from 5 or 17.2% of IMI were identified only at the Statlstlcal Analysis genus level. Three IMI identified as the genera Comparison of mean lactation days infect- Serratia were classified as Serratia spp. by the ed,loglo, was by least squares ANOVA. Mul- APEOE system. Two IMI were classified as tiple comparisons of means of unequal sample Serratia spp. because several different species sizes was by the Tukey-Kramer Method (10). were identified by API20E. The mean number Journal of Dairy Science Vol. 74, No. 6, 1991
SERRATIA Y-
1863
INFECTIONS
for Serratia spp. ranged from 22.2 to 100%. Fourteen or 48.3% of Serratia spp. IMI were clinical for at least one sample throughout the IMl duration of the IMI. None of the cows was diagnosed with peracute coliform mastitis. SDecies Number Percentaae Mean percentage of total samples from an IMI Serratia marcescens 19 65.5 that were clinical was 17.3% (Table 2). PerSerratia odoHfera 2 6.9 Serratia Iiqrrefaciens 1 35 centage of samples from an IMI that were Serratia plymuthica 1 3.5 clinical ranged from 0 to 50%. Serratia rubidae 1 3.5 Origin of IMI by stage of lactation is in Serratia spp. 5 17.2 Figure 3. Approximately 48.3% of Serraria Total 29 100.0 spp. IMI originated in the first half of the dry period, 31% in the last half of the dry period, and 20.7% during lactation. Nine or 64.3% of of samples obtained per IMI was 15.7 (range 1 MI new in the first half of the dry period to 79). Serratia spp. were not isolated from persisted to lactation. One IMI that originated every sample taken during the duration of an in lactation was from a cow that had a Serratia IMI within a quarter. Mean number of samples spp. MI in another quarter. Mean lactation days infected, loglo of all obtained per I MI from which a Serratia spp. Serratia ssp. IMI, was 1.74 or a geometric was isolated and identified by AFVOE was of 55 d. Serratia spp. IMI that originated mean 11.1 (range 1 to 63). The API2OE system identified the same species for each isolation in the first half of the dry period were present in lactation for a greater (P < .05) number of from a particular IMI in 69% of the MI. Serratia spp. were isolated from appmxi- days than IMI new during the last half of the mately 9% of all Gram-negative bacterial IMI. dry period or lactation (Table 3). Breed and Three of the Serratia spp. IMI were mixed age of cow had no influence on lactation days with other pathogens. Mixed IMI included one infected. Reasons for elimination of Serratia spp. esculin-positive Streptococcus spp. and two Klebsiella pneumoniue. A mean of 32% of the IMI were (number of IMI,percentage of IMI): samples from such IMI were mixed with the spontaneous, 18, 62.1%; culled or died, 4, 13.8%;lactation antibiotics, 2, 6.9%;dried off, other pathogens. Serratia ssp. were isolated from a mean of 3, 10.3%; and stdl present at end of data 78.2% of samples taken throughout the dura- summarization, 2, 6.9%. A total of 14 IMI tion of an IMI.Percentage of samples positive were treated with lactation antibiotics, with a TABLE! 1. Serratia spp. isolated from naturally Occuning bovine inframmmq infections 0 in a dairy research herd over 32 mo.
TABLE 2. Percentage of samples displaying clinical signs and samples positive for Serratia spp. during the duration of an inhlunarmnary infection @MI). Percentage of samples
Serratia marcescens Serratia spp. Serratia odorifera Serratia fiquefaciens Serratia plymfhica Serratia rubidae AllIMI
IhiI
clinical sirms' Tz SE
19
112 35.2 50.0 0 11.1 0 17.3
5
2 1 1 1 29
3.8 18.7 50.0
52
Positive for Serratia mp2
X
SE
76.6
4.2 12.9 0
83.2 100.0 100.0 22.2 73.3 782
' ( ~ u t w x of clinical samples/tota.lnumber of samples from IM) x 100. 2(Nomber of samples from which a Serratia spp. was isolatcd/total number of samples from I?vQ
4.2 X
100.
Journal of Dairy Science Vol. 74. No. 6. 1991
1864 TABLE 3. Wed of origin of Serratia ipp.
TODHUNTER ET AL. '
.
my infections on lactation days infected, loglo. Geometric
Lactation days infected, loglo
origin (n)' First half of dry period Last half of dry period Lactation
9 9 6
-
X 222a 1.61b lab
mean
SE
(4
21 .30 .31
165.9 40.7 17.4
followed by d i f f m t superscripts differ (P < .05). ary infections. 'Number of . .
%leans
study. Treatment of IMI with an approved lactation antibiotic was relatively ineffective. Serratia spp. accounted for 9% of the Gram-negative bacterial IMI in the herd (11). outbreaks of Serratia spp. mastitis have been described in lactating cows, with the source DISCUSSION traced to the use of contaminated teat dips (4, Serratia spp. IMI were highly associated 12) or teat dip cups (12). New Serratia spp. with the dry period and tended to be of long IMI were detected in each year of this study duration by comparison with duration of other and distributed over several months of each Gram-negative bacteria IMI (8). Approxi- year. Additionally, the majority of Serratia mately 80% of Serratia spp. IMI were first spp. IMI were first detected during the dry detected during the dry period. The dry period period, when teat dipping was not practiced. has been shown to be a critical time in the dynamics of Gram-negative bacterial IMI (7, Although outbreaks of Serratia mastitis occur, 9) and would appear to be extremely critical the current study demonstrated that Serratia for Serratia spp. IMI. Current data also demonstrated that those Serratia spp. that were established in the f i s t half of the dry period 50 and persisted to lactation femained in the lactating gland for a greater time period than 45 those IMI that were first detected at calving or 40 during lactation. Reasons for the persistence of IMT that are established in the first half of the 2 35 dry period are unknown; however, this is not w unique to Serratia spp. IML Klebsiella spp. 0 30 LMI that originated in the first half of the dry c, 25 period also persisted in the lactating gland for greater time periods compared with IMI that 0 20 were new at calving or during lactation (9). a 15 Other investigators have reported the persist10 ent, chronic nature of s. murcescens (1,4, 13) and S. liquefaciens (2) IMI. 5 Serratia spp. IMI have proven difficult to 0 treat with antibiotics (2, 12); however, success First Half Last Half Lactation with neomycin intramammary therapy has _ _ - Dry Period-- been reported (1, 6). Spontaneous elimination i p 3. origin of Serratia spp.-m infecof the organism in the absence of antibiotic tionsP 0. Values are expressed as percentage of total therapy was the most frequent reason for ter- IMI.Numbers above bars arc number of IMI for indicated mination of a Serratia spp. IMI in the current stage of lactation. mean number of treatments per IMI equal to 2.1 (range 1 to 7). Of the 14 treated quarters, Serratia spp. were no longer isolated fmm 14.3% of IMI following antibiotic infusion.
c (
z
Journal of Dairy Science Vol. 74, No. 6, 1991
L1
SERRATLA Y-
spp. IMI can represent a portion of Gramnegative bacterial mastitis in a herd and yet not be associated with an “outbreak”. The source of Serratia spp. was not determined in this study. A contaminated dry cow product seems unlikely, given the sporadic nature of the occurrence of IMI and the fact that heifers were infected at calving. Recycled manure has been shown to support large numbers of Gramnegative bacteria (8), and teat end exposure to recycled manure bedding may have been a possible source of Serratiu spp. during the dry period and was common to both mature cows and heifers. Approximately 48% of Serratia spp. IMI were clinical at some point throughout the duration of infection, and none of the COWS displayed systemic signs of peracute coliform mastitis. The frequency of clinical mastitis by Serratia spp. is lower than that reported for other Gram-negative bacterial IMI (8). Although Serratia spp. were isolated from 78% of samples obtained from an JMI, only 17%of the samples were clinical. The subclinical form of infection appears to be more characteristic of Serratia spp. IMI than for other Gramnegative bacterial IMI (3, 8). Diagnosis of a Serratia spp. LMI can be complicated by the fact that the organism may not be isolated from all milk samples and by the subclinical nature of the infection. Serratia spp. are occasionally isolated from milk samples and cannot be diagnosed as IMI based on criteria listed here. Clearly, Serrazia spp. are present in the environment and could be in milk samples as a result of contamination. This would further complicate diagnoses of IMI. Serratia murcescells (1, 5 , 12, 13) and S. liquefaciens (2, 6) have been isolated from bovine IMI.Serratia murcescens was the most commonly isolated Serratia spp. in this study. Several other species of Serratia were also identified, demonstrating the diversity of Serratia spp. that can infect the bovine mammary gland. Serratia spp. IMI may also be mixed with other pathogens. The rate of coliform IMI has been shown to increase with parity (8). The greatest percentage of Serratia spp. IMI were isolated from cows in fifth or greater lactation, indicating that older cows are more susceptible to infection with Serratia spp. However, Serratia spp.
INFECIlONS
1865
IMI were detected in cows of a l l ages, including heifers. Serratia spp. I MI were highly associated with the dry period and tended to be chronic, subclinical infections. Current data indicates that ccmtrol methods for Serratia spp. mastitis, in the absence of an obvious contaminating source, should be directed at dry cows. ACKNOWLEDGMENTS
The authors thank Pamela Schoenberger, Sue Romig, Heidi Rennecker, and Lucinda Shock for technical assistance. REFERENCES
lBamum, D. A., E. L. Thackeray, and N. A. Fish. 1958. An outbreak of mastitis by Serraria marcescens. Cae J. Comp. Med 2 3 9 2 . 2Bowman, G. L., W. D. Hueston, G. I. Boner, J. J. Hurley, and J. E. Andreas. 1986. Serratia lquefociens mastitis in a dahy herd. J. Am. Vet. Med. Assoc. 189: 913. 3 Eberhart, R. J., and J. M. Buckalew. 1972. Intramammary infections in a dairy herd with a low incidence of Sfreptococcus agalactiae and Staphylococcus aureus infections. J. Am. Vet. Med. Assoc. 171:630. 4 Eberbiut, R. J.. and R J. Erskine. 1986. Herd mastitis problems caused by unusual pathogens. Roc. 19th Amm. Mtg. Am. Assoc. Bovine Ract. 19:84. 5 Howell, D. 1972. Survey on mastitis caused by environmental bacteria. Vet. Rec. 90:654. 6Nicholls, T. I., M. G. Barton. and B. P. Anderson. 1981. Serratia liqrcefociens as a cause of mastitk in dairy cows. Vet. Rec. 109:288. 70liver. S. P., and B. A. Mitchell. 1983. Susceptibility of bovine mammary gland to infections during the dry period. J. Dairy Sci. 66:1162. 8 Smith,K. L.. D. A. Todhunter, and P. S. Schoenkga. 1985. Envhnambd mastitis: cause, prevalence, prevention. I. Dairy Sci. 68:1531. 9 Smith, K. L., D. A. Todhunter, and P. S. Schoenberga.1985. Environmental pathogens and inbramammary infection during the dry period. J. Dairy Sci 68:402. 10 Sokal, R.R, and F. I. Rohlf,ed. 1969. Biometry. The priaciples and practice of statistics in biological research W. H. Freaaanand Company, San Francisco,
CA. llTodhunter, D., K. L. Smith, and J. S. Hogan. 1990. Growth of Gram-negative bacteria in dry cow secre tioa J. Dairy Sci. 73363. 12 Van Damme, D. M. 1982. Mastitis caused by contaminated teat dip and dipping cup. Vet. Med. Small Anim. Clia 77541. 13WiLson, D. J., J. H.Kirk, R D. Wallrer, and Q. W. Boswcntb. 1990. Serratia marcescens mastitis in a dairy herd. J. Am. Vet. Med. Assoc. 1%11U2.
Journal of Dairy Science Vol. 74, No. 6, 1991
Depammt of Dalry Science
Ohio Agricultural Research and Development Center wooster 44691 ABSTRACT
INTRODUCTION
Jntramammary infections from which Serratia spp. were isolated were studied over a 32-mo period in a research dairy herd consisting of approximately 120 lactating cows. A total of 29 Serratia spp. intramammary infections were detected and accounted for 9% of all Gram-negative bacterial intramamnmy infections. Serratia marcescens was the most common Serratia spp. isolated. origin of intramammary infections was 48.3% during the frrst half of the dry period, 31% during the last half of the dry period, and 20.7% during lactation. A total Of 64% of -ni infwtions that were first detected during the first half of the dry period persisted to calving. Geometric mean number of lactation days infected for all infections was 55. Intramarmnary infections that originated during the first half of the dry period were present in lactation for a significantly greater number of days compared with intramammary infections new during the last half of the dry period or lactation. A total of 48% of infections were clinical. Serratia spp. intramammary infections tended to be of long duration compared with other Gramnegative bacterial intramammary infections and were highly associated with the dry period (Key words: mastitis, Serratia spp., dry period)
Serratia marcescens and Serratia liquefaciens have been reported as causative agents of bovine mastitis (1, 2, 6, 13). Many of the investigations on Serratia spp. -ni infections (MI) were described as outbreaks or episodic (1, 2, 13) and linked to contaminated water or teat dips (4, 5, 12). The purposes of this study were to identify Serratia spp. isolated from naturally occurring IMI in a total confinement herd and to determine various infection parameters. such as origin of IMI and lactation days infected. MATERIALS AND METHODS Animals
me
Abbreviation key: IMI = intramammary infection
Received October 29, 1990. Accepted January 15, 1991. ‘Salaries and mearch support provided by state and federal fands appmpriated to tbe Ohio Agricultural R e Search and DeveloptIIUIt C~ntcr,Thc Ohio State Ulrivcrsi@. Mamucript Npmba 282-90. 1991 J Dairy Sci 74:1860-1865
Experimental animals were Holstein (80%) and Jersey (20%) dairy cattle from the research herd of the Ohio Agn’lcultural Research and Development Center, Wooster. Herd size was approximately 120 lactating cows. Cows were milked in a double 3, side opening Surge (Babson Bros. Co.,Oak Brook, IL) parlor. AU teats were predipped and postdipped with 4% sodium hypochlorite Clorox Co., Oakland, CA). Housing was total confinement, with appmximately 75% of the lactating herd in free stalls and 25% in tie stalls. Bedding material in free stalls was ~ n treated, noncomposted recycled manure Solids Separator, Babson Bros. Co., Oak Brodr, IL). Wood shavings were used as bedding material in tie stalls. Dry cows were housed in free stalls and bedded with recycled manure. Cows were dried off appmxjmately 60 d prior to anticipated calving date by abrupt cessation of milking. AU quarters of all cows were infused with a commercially available antibiotic preparaxion (300 mg cephapirin benzathine; Tomorrow, Agricultural products, Division of Bristol-Myers Co.,DeWitt, NY). Dry cow therapy was administered by mastitis laboratory personnel using aseptic techniques.
1860
-
Cows were moved to individual box stalls 3 to 7 d prior to parturition. Bedding material in box stalls was pelleted corn cobs (The Anderson co.,Marnee, OH). cows remained in box stalls for 3 d after calving and then were moved to either free or tie stalls. Management of cows was as &scribed by Smith et aL (8). Quarter Foremilk Sampling Schedule
Quarter foremilk samples for microbiological culture were obtained from all cows at the following time points: 14 and 7 d prior to drying off; the day of drying off and prior to antibiotic infusion; 30 d of involution; calving; 3, 7, 14, and 30 d postcalving; and at 3O-d intervals for the remainder of lactation. Single quarter milk samples were taken at all time points, except at 30 d of involution when samples were taken in duplicate. Sampling procedm was as previously described (8). Samples were also obtained from laming COWS reported with clinical -titis. A mmary quarter was classified as clinical when the secretion displayed signs of abnormality (flakes, clots, etc.) at the time of sampling. Cows with clinical mastitis were sampled for 2 consecutive d and 7 and 14 d after initial sampling. Cows with clinical mastitis that received antibiotic therapy were sampled 1 d before treatment, the day of treatment prior to antibiotic infusion, and 7 and 14 d posttreatment. Antibiotic treatment consisted of two in@infusions of 200 mg cephapirin sodium (TODAY, Agricultural P d u c t s , Division of Sristol-Myers Co., Syracuse, NY) in the clinical quarter at a 12-h interval. Additional samples were obtained from cows that were on other experimental trials or to confirm infection status of quarters. Data summarization was from all cows sampled January 1986 through August 1989.
streaked with .I ml of milk Plates were incubated at 37’C and examined for bacterial p w t h at 24 and 48 h of incubation. colonies tentatively identified as Gram-negative bacteria were Gram stained. Gram-negative bacteria were streaked for isolation on blood agar and identified with the API2OE System (Analytab products, plainview, NY). The API2OE system identification was performed according to manufacturer’s instructions for an 18- to 24-h incubation. Dlagnosla, Origln by Stage of Lactatton, and Duratlon of Infections
A Serratia spp. IMI was diagnosed under one of the following conditions: 1) single isolation of a Serratia spp. fmm a clinical quarter,
2) isolation of a Serratia spp. from both of duplicate quarter samples, or 3) isolation of a Serratia spp. in two out of three consecutive samples with less than 31 d between isolations. An exception to the 30 d between isolations was made when the same Serratia spp. was
isolated at 30 d of involution and again at calving and greater than 30 d had elapsed between calving and the 3 0 4 dry period sample. A Serratia spp. IMI was classified as a mixed infection when another pathogen was simultaneously isolated fmm the quarter and met the guidelines described above for IMI diagnosis. Duration of an IMI was the interval of time in days between the first and last isolation plus 1 d A single isolation of Serratia spp. from a clinical quarter had a duration of 1 d. Lactation days infected were defined as the number of days during lactation that an IMI was present. The day of calving was considered as d 1 of lactation. Lactation days infected were transformed to 1ogl@ An M I was diagnosed as clinical if at least one of the samples was associated with clinical signs during the duration of the IML. PercentMlcroblologlcai Procedures age of samples fmm an IMI displaying clinical AU milk samples were examined microbio- signs was also calculated by dividing the total logically (8) by streaking .01 ml milk onto one number of clinical samples by the total number quadrant of a trypticase soy agar plate (BBL of samples obtained from the IMI. Percentage Microbiology Systems, Becton Diclcinson Co., of samples positive for Serratia spp. during the Cockeysville, MD) containing 5 % bovine duration of an IMI was calculated by dividing blood and .l%esculin (Sigma Chemical Co., number of samples from which a Serrafiu spp. St. Louis, MO). One-half of a Maccoakey was isolated by the total number of samples agar (BBL Microbiology Systems) plate was obtained from the IMI. Jomnal of Dairy Science Vol. 74. No. 6, 1991
1862
35 30
TODHUNTER ET AL.
1
7
9
-
5
U
2
25-
8
4
u
0
L 3 0
P $ 2
z
1 0
JAN FEE YAR AF'R MAY JUN JUL AUG SEP OCT NOV DEC
Month
1
2
4
Lactation Number
25
Figure 1. Origin Of Serrati0 ~ p pintnrmammary . infections 0 by lactation number. Values are expressed as percentage of total IMI.Numbers above bars are number of IMI for indicated lactation number.
Figure 2.Distribution of first detection of Smutiu spp. n it-y infections by month. values are expressed as number of IMI per month for indicated year.
RESULTS
Serratia spp. were isolated from 29 IMI in 25 cows (6 Jersey and 19 Holstein) over the 32-mo study period. Distribution of lactation number of io& with Serratia spp. IMI is in A Serratia spp. IMI first detected at 30 d of Figure 1. Approximately 31% of IMI were involution and not present in the three samples detected in cows in fifth lactation or greater. A obtained prior to drying off was considered to total of four or 13.8% of Serratia spp. IMI have originated during the first half of the dry were detected in heifers with two IMI first period. Intramammary infections that were de- detected at calving. Percentage distribution of tected in the first half of the dry period and a l l cows in the herd by lactation number was again at calving were classified as having per- first lactation 34.2%; second lactation, 25.6%; sisted to calving. Infections first detected at third lactation, 16.8%; fourth lactation, 11.7%; calving or within 3 d after calving were classi- and fifth lactation or greater, 11.7%. Mammary fied as having originated during the last half of quarter distribution of IMI was 34.5% in front the dry period. Serratia spp. IMI first detected quarters and 65.5% in rear quarters. between 4 d of lactation and drying offwere Number of new Serratia spp. IMI detected considered to have originated during lactation. for each year of the study was 6 during 1986, The most commonly isolated species of 11 during 1987, 6 during 1988, and 6 from Serratia, as identified by the APEOE system January 1989 through August 1989. First defor a particular IMI, was considered the causa- tection of Serratia spp. I M I were distributed tive organism. An IMI was classified as Serra- throughout several months of the year for each tia spp. when no one species predominated or year of the study (Figure 2). Serratia marcescens was the species most when isolations were i d e n a d only at the fresuently identified. Species identified and genus level. their frequency are in Table 1. Isolates from 5 or 17.2% of IMI were identified only at the Statlstlcal Analysis genus level. Three IMI identified as the genera Comparison of mean lactation days infect- Serratia were classified as Serratia spp. by the ed,loglo, was by least squares ANOVA. Mul- APEOE system. Two IMI were classified as tiple comparisons of means of unequal sample Serratia spp. because several different species sizes was by the Tukey-Kramer Method (10). were identified by API20E. The mean number Journal of Dairy Science Vol. 74, No. 6, 1991
SERRATIA Y-
1863
INFECTIONS
for Serratia spp. ranged from 22.2 to 100%. Fourteen or 48.3% of Serratia spp. IMI were clinical for at least one sample throughout the IMl duration of the IMI. None of the cows was diagnosed with peracute coliform mastitis. SDecies Number Percentaae Mean percentage of total samples from an IMI Serratia marcescens 19 65.5 that were clinical was 17.3% (Table 2). PerSerratia odoHfera 2 6.9 Serratia Iiqrrefaciens 1 35 centage of samples from an IMI that were Serratia plymuthica 1 3.5 clinical ranged from 0 to 50%. Serratia rubidae 1 3.5 Origin of IMI by stage of lactation is in Serratia spp. 5 17.2 Figure 3. Approximately 48.3% of Serraria Total 29 100.0 spp. IMI originated in the first half of the dry period, 31% in the last half of the dry period, and 20.7% during lactation. Nine or 64.3% of of samples obtained per IMI was 15.7 (range 1 MI new in the first half of the dry period to 79). Serratia spp. were not isolated from persisted to lactation. One IMI that originated every sample taken during the duration of an in lactation was from a cow that had a Serratia IMI within a quarter. Mean number of samples spp. MI in another quarter. Mean lactation days infected, loglo of all obtained per I MI from which a Serratia spp. Serratia ssp. IMI, was 1.74 or a geometric was isolated and identified by AFVOE was of 55 d. Serratia spp. IMI that originated mean 11.1 (range 1 to 63). The API2OE system identified the same species for each isolation in the first half of the dry period were present in lactation for a greater (P < .05) number of from a particular IMI in 69% of the MI. Serratia spp. were isolated from appmxi- days than IMI new during the last half of the mately 9% of all Gram-negative bacterial IMI. dry period or lactation (Table 3). Breed and Three of the Serratia spp. IMI were mixed age of cow had no influence on lactation days with other pathogens. Mixed IMI included one infected. Reasons for elimination of Serratia spp. esculin-positive Streptococcus spp. and two Klebsiella pneumoniue. A mean of 32% of the IMI were (number of IMI,percentage of IMI): samples from such IMI were mixed with the spontaneous, 18, 62.1%; culled or died, 4, 13.8%;lactation antibiotics, 2, 6.9%;dried off, other pathogens. Serratia ssp. were isolated from a mean of 3, 10.3%; and stdl present at end of data 78.2% of samples taken throughout the dura- summarization, 2, 6.9%. A total of 14 IMI tion of an IMI.Percentage of samples positive were treated with lactation antibiotics, with a TABLE! 1. Serratia spp. isolated from naturally Occuning bovine inframmmq infections 0 in a dairy research herd over 32 mo.
TABLE 2. Percentage of samples displaying clinical signs and samples positive for Serratia spp. during the duration of an inhlunarmnary infection @MI). Percentage of samples
Serratia marcescens Serratia spp. Serratia odorifera Serratia fiquefaciens Serratia plymfhica Serratia rubidae AllIMI
IhiI
clinical sirms' Tz SE
19
112 35.2 50.0 0 11.1 0 17.3
5
2 1 1 1 29
3.8 18.7 50.0
52
Positive for Serratia mp2
X
SE
76.6
4.2 12.9 0
83.2 100.0 100.0 22.2 73.3 782
' ( ~ u t w x of clinical samples/tota.lnumber of samples from IM) x 100. 2(Nomber of samples from which a Serratia spp. was isolatcd/total number of samples from I?vQ
4.2 X
100.
Journal of Dairy Science Vol. 74. No. 6. 1991
1864 TABLE 3. Wed of origin of Serratia ipp.
TODHUNTER ET AL. '
.
my infections on lactation days infected, loglo. Geometric
Lactation days infected, loglo
origin (n)' First half of dry period Last half of dry period Lactation
9 9 6
-
X 222a 1.61b lab
mean
SE
(4
21 .30 .31
165.9 40.7 17.4
followed by d i f f m t superscripts differ (P < .05). ary infections. 'Number of . .
%leans
study. Treatment of IMI with an approved lactation antibiotic was relatively ineffective. Serratia spp. accounted for 9% of the Gram-negative bacterial IMI in the herd (11). outbreaks of Serratia spp. mastitis have been described in lactating cows, with the source DISCUSSION traced to the use of contaminated teat dips (4, Serratia spp. IMI were highly associated 12) or teat dip cups (12). New Serratia spp. with the dry period and tended to be of long IMI were detected in each year of this study duration by comparison with duration of other and distributed over several months of each Gram-negative bacteria IMI (8). Approxi- year. Additionally, the majority of Serratia mately 80% of Serratia spp. IMI were first spp. IMI were first detected during the dry detected during the dry period. The dry period period, when teat dipping was not practiced. has been shown to be a critical time in the dynamics of Gram-negative bacterial IMI (7, Although outbreaks of Serratia mastitis occur, 9) and would appear to be extremely critical the current study demonstrated that Serratia for Serratia spp. IMI. Current data also demonstrated that those Serratia spp. that were established in the f i s t half of the dry period 50 and persisted to lactation femained in the lactating gland for a greater time period than 45 those IMI that were first detected at calving or 40 during lactation. Reasons for the persistence of IMT that are established in the first half of the 2 35 dry period are unknown; however, this is not w unique to Serratia spp. IML Klebsiella spp. 0 30 LMI that originated in the first half of the dry c, 25 period also persisted in the lactating gland for greater time periods compared with IMI that 0 20 were new at calving or during lactation (9). a 15 Other investigators have reported the persist10 ent, chronic nature of s. murcescens (1,4, 13) and S. liquefaciens (2) IMI. 5 Serratia spp. IMI have proven difficult to 0 treat with antibiotics (2, 12); however, success First Half Last Half Lactation with neomycin intramammary therapy has _ _ - Dry Period-- been reported (1, 6). Spontaneous elimination i p 3. origin of Serratia spp.-m infecof the organism in the absence of antibiotic tionsP 0. Values are expressed as percentage of total therapy was the most frequent reason for ter- IMI.Numbers above bars arc number of IMI for indicated mination of a Serratia spp. IMI in the current stage of lactation. mean number of treatments per IMI equal to 2.1 (range 1 to 7). Of the 14 treated quarters, Serratia spp. were no longer isolated fmm 14.3% of IMI following antibiotic infusion.
c (
z
Journal of Dairy Science Vol. 74, No. 6, 1991
L1
SERRATLA Y-
spp. IMI can represent a portion of Gramnegative bacterial mastitis in a herd and yet not be associated with an “outbreak”. The source of Serratia spp. was not determined in this study. A contaminated dry cow product seems unlikely, given the sporadic nature of the occurrence of IMI and the fact that heifers were infected at calving. Recycled manure has been shown to support large numbers of Gramnegative bacteria (8), and teat end exposure to recycled manure bedding may have been a possible source of Serratiu spp. during the dry period and was common to both mature cows and heifers. Approximately 48% of Serratia spp. IMI were clinical at some point throughout the duration of infection, and none of the COWS displayed systemic signs of peracute coliform mastitis. The frequency of clinical mastitis by Serratia spp. is lower than that reported for other Gram-negative bacterial IMI (8). Although Serratia spp. were isolated from 78% of samples obtained from an JMI, only 17%of the samples were clinical. The subclinical form of infection appears to be more characteristic of Serratia spp. IMI than for other Gramnegative bacterial IMI (3, 8). Diagnosis of a Serratia spp. LMI can be complicated by the fact that the organism may not be isolated from all milk samples and by the subclinical nature of the infection. Serratia spp. are occasionally isolated from milk samples and cannot be diagnosed as IMI based on criteria listed here. Clearly, Serrazia spp. are present in the environment and could be in milk samples as a result of contamination. This would further complicate diagnoses of IMI. Serratia murcescells (1, 5 , 12, 13) and S. liquefaciens (2, 6) have been isolated from bovine IMI.Serratia murcescens was the most commonly isolated Serratia spp. in this study. Several other species of Serratia were also identified, demonstrating the diversity of Serratia spp. that can infect the bovine mammary gland. Serratia spp. IMI may also be mixed with other pathogens. The rate of coliform IMI has been shown to increase with parity (8). The greatest percentage of Serratia spp. IMI were isolated from cows in fifth or greater lactation, indicating that older cows are more susceptible to infection with Serratia spp. However, Serratia spp.
INFECIlONS
1865
IMI were detected in cows of a l l ages, including heifers. Serratia spp. I MI were highly associated with the dry period and tended to be chronic, subclinical infections. Current data indicates that ccmtrol methods for Serratia spp. mastitis, in the absence of an obvious contaminating source, should be directed at dry cows. ACKNOWLEDGMENTS
The authors thank Pamela Schoenberger, Sue Romig, Heidi Rennecker, and Lucinda Shock for technical assistance. REFERENCES
lBamum, D. A., E. L. Thackeray, and N. A. Fish. 1958. An outbreak of mastitis by Serraria marcescens. Cae J. Comp. Med 2 3 9 2 . 2Bowman, G. L., W. D. Hueston, G. I. Boner, J. J. Hurley, and J. E. Andreas. 1986. Serratia lquefociens mastitis in a dahy herd. J. Am. Vet. Med. Assoc. 189: 913. 3 Eberhart, R. J., and J. M. Buckalew. 1972. Intramammary infections in a dairy herd with a low incidence of Sfreptococcus agalactiae and Staphylococcus aureus infections. J. Am. Vet. Med. Assoc. 171:630. 4 Eberbiut, R. J.. and R J. Erskine. 1986. Herd mastitis problems caused by unusual pathogens. Roc. 19th Amm. Mtg. Am. Assoc. Bovine Ract. 19:84. 5 Howell, D. 1972. Survey on mastitis caused by environmental bacteria. Vet. Rec. 90:654. 6Nicholls, T. I., M. G. Barton. and B. P. Anderson. 1981. Serratia liqrcefociens as a cause of mastitk in dairy cows. Vet. Rec. 109:288. 70liver. S. P., and B. A. Mitchell. 1983. Susceptibility of bovine mammary gland to infections during the dry period. J. Dairy Sci. 66:1162. 8 Smith,K. L.. D. A. Todhunter, and P. S. Schoenkga. 1985. Envhnambd mastitis: cause, prevalence, prevention. I. Dairy Sci. 68:1531. 9 Smith, K. L., D. A. Todhunter, and P. S. Schoenberga.1985. Environmental pathogens and inbramammary infection during the dry period. J. Dairy Sci 68:402. 10 Sokal, R.R, and F. I. Rohlf,ed. 1969. Biometry. The priaciples and practice of statistics in biological research W. H. Freaaanand Company, San Francisco,
CA. llTodhunter, D., K. L. Smith, and J. S. Hogan. 1990. Growth of Gram-negative bacteria in dry cow secre tioa J. Dairy Sci. 73363. 12 Van Damme, D. M. 1982. Mastitis caused by contaminated teat dip and dipping cup. Vet. Med. Small Anim. Clia 77541. 13WiLson, D. J., J. H.Kirk, R D. Wallrer, and Q. W. Boswcntb. 1990. Serratia marcescens mastitis in a dairy herd. J. Am. Vet. Med. Assoc. 1%11U2.
Journal of Dairy Science Vol. 74, No. 6, 1991
