- Email: [email protected]
Outcome of Clinical Mastitis in Dairy Heifers Assessed by Reexamination of Cases One Month After Treatment
Outcome of Clinical Mastitis in Dairy Heifers Assessed by Reexamination of Cases One Month After Treatment S. Waage,*,1 H. R. Skei,† J. Rise,‡ T. Rogdo,§ S. Sviland,* and S. A. Ødegaard|| *Cattle Health Section, National Veterinary Institute, PO Box 8156 Dep, 0033 Oslo, Norway †Veterinary Center in Sparbu, 7710 Sparbu, Norway ‡Veterinary Surgeon, 2550 Os, Norway §Veterinary Center in Sandnes, 4300 Sandnes, Norway ||Department of Reproduction and Forensic Medicine, The Norwegian School of Veterinary Science, 0033 Oslo, Norway
ABSTRACT Heifers that were treated for clinical mastitis prior to parturition or within 14 d postpartum were reexamined approximately 1 mo after treatment. Clinical examination of the heifers and microbiological examination of quarter milk samples were carried out on both occasions. Of the 1000 heifers included in the study, 10.9% were culled within 28 d after treatment. Udder damage caused by mastitis was the only or main reason for culling in 96% of those heifers. In comparison, 4.5% of nonmastitic heifers from the same herds were culled within 30 d postpartum. Twenty-five percent of those heifers that were not culled at d 28 after treatment had at least one nonfunctional quarter at that time. One thousand one hundred twenty-two quarters that were clinically affected at the time of treatment were reexamined; 22% were nonfunctional, 14% were still affected by clinical mastitis, 12% had subclinical mastitis, 5% had a latent infection with coagulase-positive staphylococci or Streptococcus dysgalactiae, and 46% were bacteriologically negative and had a normal cell count at the time of reexamination. High percentages of nonfunctional quarters were observed among those quarters that were infected with Arcanobacterium pyogenes or with coagulasepositive staphylococci at treatment. When all quarters that were clinically affected at treatment were considered, 40% of quarters were cured and were still in lactation at reexamination. Quarters infected with coagulasenegative staphylococci had a higher cure rate than quarters infected with other organisms. At reexamination, clinical signs of thelitis were observed in many of those quarters that were nonfunctional following the episode of clinical mastitis and also in 25% of lactating quarters in which clinical mastitis persisted. (Key words: mastitis, heifer, outcome, culling)
Received March 17, 1999. Accepted August 26, 1999. 1 Corresponding author. 2000 J Dairy Sci 83:70–76
70
Abbreviation key: CPS = coagulase-positive staphylococci. INTRODUCTION Frequent occurrence of bovine mastitis and the immediate expenses and pathoanatomic complications associated with the disease lead to great financial losses for the farmers. Even cases that receive an appropriate antibacterial treatment at an early stage of a clinical IMI are frequently left with permanent tissue damage of parts of the affected quarter. The consequences are decreased milk production, reduced milk quality, and occasionally a completely nonfunctional quarter. Cows that suffer such complications following an episode of clinical mastitis are at a greater risk of being culled before their optimal age for disposal than are nonmastitic cows of similar age (4). Subsequent to a clinical episode, a chronic subclinical or clinical infection may be established, particularly when Staphylococcus aureus is the causal organism (14). Thus a reservoir of mammary pathogens is formed which may infect quarters of other cows in the herd. Many dairy heifers are affected by clinical mastitis around parturition. According to records of disease data in the Norwegian Health Card System for Cattle, 13.3% of dairy heifers that calved in Norway in 1995 were treated for clinical mastitis within 14 d postpartum (S. Waage, 1999, unpublished data). In general, the prospects for the complete recovery of affected secretory and ductular tissue in quarters of heifers that experience an episode of clinical mastitis are considered to be relatively poor (12). However, no comprehensive studies seem to have been conducted to assess the outcome of clinical mastitis in heifers. In the present study, dairy heifers that were treated for clinical mastitis prepartum or within 14 d postpartum were reexamined approximately 1 mo after treatment to determine the status of the animals and their udder quarters.
71
OUTCOME OF CLINICAL MASTITIS IN HEIFERS
MATERIALS AND METHODS Definitions Cases of mastitis were classified as clinical or subclinical in accordance with the definitions given by the International Dairy Federation (6). Udder quarters in which lactation ceased because of chronic clinical mastitis were classified separately as nonfunctional quarters. Infectious (clinical or subclinical) mastitis was present when pathogens and inflammatory changes were detected in the secretion (6). Nonspecific (clinical or subclinical) mastitis was present when there were inflammatory changes but no pathogens in the secretion (6). A latent infection was present when the secretion contained pathogens but had a normal cell count (6). The distinction between a normal and a high cell count of the secretion, as estimated by the California Mastitis Test, was based on Scandinavian recommendations for interpretation of results of the California Mastitis Test (8). The reaction of the secretion of the particular quarter as well as the reactions of the secretions of the remaining quarters of the cow were taken into account. Thelitis was considered to be present when the wall of the teat cistern was thickened and hardened, as felt by clinical examination of the teat (10). Study Design A field study was carried out in 24 veterinary districts in Norway. Approximately 100 veterinary surgeons were involved. The general arrangement of the study and the location of the veterinary districts in which the study took place have been described previously (15). The study included heifers that, during a particular year (1997), were treated for clinical mastitis prepartum or within 14 d postpartum. When a heifer experienced more than one episode of clinical mastitis, only the first episode was considered. The veterinary practitioner supplied information, on a standard form, from the heifer and from the herd to which she belonged and collected secretion samples from all quarters. Microbiological examination of secretion was carried out at the Mastitis Laboratory of the National Veterinary Institute in Oslo, Norway. According to the agreement with the veterinary practitioners, reexamination and resampling of cases were performed approximately 4 wk after the initial examination. Routinely, 3 wk after treatment the Mastitis Laboratory mailed to the veterinary practitioners a reminding letter, in which the name of the farmer, identity number of the heifer, and optimal date of reexamination (28 d after treatment) were given. An exception was made for heifers that were treated earlier than 4 wk prepartum; for those heifers reexamination was delayed until just before or after parturition. For heifers that had been
culled before the time of reexamination, the date of and reason for culling were reported. For those heifers that were not culled 4 wk after treatment, quarter milk samples were collected, and data on the clinical condition of the heifer were recorded on a standard form. The culling rate for heifers that were treated for clinical mastitis was compared with the culling rate for nonmastitic heifers in the same herds. Clinical Cases and Control Animals The number of heifers that were treated for clinical mastitis was 1040, and a total of 1361 quarters was clinically affected. Bacteriological findings in quarter secretion samples collected from these heifers at the time of treatment have been published previously (15). No information was received on the course of the disease for 40 heifers, for which 56 quarters were clinically affected at treatment. Thus, the final material used for assessment of the outcome of clinical mastitis on individual animals included 1000 heifers. For these heifers, a total of 1305 quarters were clinically affected at treatment. However, heifers that were culled within 28 d after treatment were not resampled before culling, and the assessment of the outcome of quarters was based on the bacteriological reexamination of secretions from those heifers that were still alive 4 wk after treatment. Of the quarters that were reexamined, 1122 had been clinically affected at the time of treatment. Initially, antibacterial therapy was administered to 925 of the 1000 heifers; 827 received both systemic and intramammary treatment, 88 received systemic treatment alone, and 10 received intramammary treatment alone. Of those heifers that received systemic treatment, an intramuscular injection of penicillin G was given to 892 heifers, a combination of penicillin G and dihydrostreptomycin was given to 11 heifers, oxytetracycline was given to 6 heifers, and various other antibacterial drugs were given to 6 heifers. For intramammary treatment, 581 heifers received a combination of penicillin G and dihydrostreptomycin, 251 heifers received penicillin G alone, and 5 heifers received various other drugs. Antibacterial therapy was not administered when such therapy was considered useless, for example when the clinically affected quarter was nonfunctional due to chronic inflammatory changes. Dates of parturition and treatment or resampling were reported by the veterinary practitioners on a standard form. Date of parturition was missing for some heifers and was later obtained from the database of the National Production Recording Scheme or from the farmer. Of the 1000 heifers included in the study, 14% were treated at least 1 d prior to parturition, 17% were treated on the day of parturition, and 69% were treated between d 1 Journal of Dairy Science Vol. 83, No. 1, 2000
72
WAAGE ET AL.
Table 1. Distribution of the 709 dairy herds from which the 1000 clinical cases in the study originated, according to the number of heifers they supplied. Heifers per herd
Herds
(n) 1 2 3 4 5 7
(n) 506 141 44 12 5 1
and 14 postpartum. Information on the clinical condition of the heifer was given on the form. Quarter signs that were recorded were presence or absence of acute or chronic signs of inflammation, of a nonfunctional quarter, of a recent or chronic teat lesion, and of visually abnormal secretion. A California Mastitis Test was performed on quarter secretions collected postpartum, and the result for each quarter was reported on the form. Of the 709 herds from which the heifers with clinical mastitis originated (Table 1), 669 were enrolled in the National Production Recording Scheme in 1997 (657 in 1996). In these herds, the Norwegian Health Card System for Cattle (16) was in operation. Disease data records from this system made it possible to identify heifers that experienced episodes of clinical mastitis and to date the occurrence of episodes in relation to time of parturition. For all culled heifers and cows in herds enrolled in the National Production Recording Scheme, date of and reason for culling are routinely recorded. To compare culling rates among heifers with and without clinical mastitis, records were studied for heifers that calved in 1996 and for heifers that calved in 1997 in herds from which the clinical cases originated. For each of those years, the proportion of heifers that were culled within 30 d postpartum was determined for those heifers that had not been treated for clinical mastitis prior to d 30 postpartum (1996, n = 3325 heifers; 1997, n = 3700 heifers). For 1996, the proportion of heifers that were culled within 30 d postpartum was also determined for those heifers that had been treated for clinical mastitis within 14 d postpartum (n = 633). Ninety-nine percent of the heifers that were clinically affected and 97% of the control heifers were of the Norwegian Red Cattle breed. The majority of the remaining heifers were cross-breeds of Norwegian Red Cattle and various beef cattle breeds.
titioners. In each envelope was a box containing four sterile, 10-ml plastic tubes and a standard form. The tubes were numbered to identify the particular quarters. Samples were collected from all quarters, except for those from which no secretion was obtained. As soon as possible after sampling the envelope was mailed to the Mastitis Laboratory. The veterinary practitioner received a fee for each case that was sampled or resampled. Laboratory Methods Methods and procedures used for the examination of quarter secretion samples were as described by the State Veterinary Laboratories of Norway (13) and in agreement with the recommendations of the International Dairy Federation (5). Secretions were characterized by appearance. The cell content in postpartum samples was estimated using the California Mastitis Test. Quarter secretions (0.01 ml) were plated on blood agar containing 5% washed bovine erythrocytes. Cultures were examined following incubation under aerobic conditions at 37°C for 24 and 48 h. When growth was not detected after incubation of the primary culture for 24 h, the original secretion sample was preincubated for 4 h at 37°C. Then an increased inoculum (0.05 ml) was plated on each of three blood agar plates and incubated at 37°C for 24 h under aerobic conditions, in 5% CO2 atmosphere, and under anaerobic conditions, respectively. Recommended methods (5) were used to identify bacteria. All secretion samples were examined for the presence of inhibitory substances using agar diffusion on MuellerHinton agar, which contained Micrococcus luteus ATCC 9341 as the test bacteria. Secretions in which inhibitory substances were detected were subjected to further examinations to determine whether a β-lactam (using penicillinase), sulfonamide (using para-aminobenzoic acid), or some unspecific heat-sensitive, inhibitory substance was present (warming to 80°C before retesting for inhibitory effect). Statistics Standard deviation of relative risk was estimated according to Rothman (11), and a confidence interval for relative risk was obtained by taking antilogarithms of confidence limits calculated for the logarithmic transformed value of relative risk (11). RESULTS AND DISCUSSION
Sample Collection Envelopes that were suitable for mailing quarter secretion samples were distributed to the veterinary pracJournal of Dairy Science Vol. 83, No. 1, 2000
Of the 1000 heifers with clinical mastitis, 109 (10.9%) were culled within 4 wk after the initial treatment. Four heifers were culled for reasons other than mastitis. For
OUTCOME OF CLINICAL MASTITIS IN HEIFERS
the remaining 105 heifers, mastitis was the only reason or the main reason for culling. Four of these heifers died or were euthanatized because of gangrenous mastitis accompanied by a severe systemic reaction, and 101 heifers were slaughtered. The heifers that were slaughtered suffered from chronic mastitis, frequently combined with marked chronic thelitis. According to the reports from the veterinary practitioners, 36 of the heifers that were slaughtered had at least one nonfunctional quarter. The 1000 heifers were treated between 208 d prepartum and 14 d postpartum, and median time of treatment was 1 d postpartum. Determination of culling rate was based on a follow-up period that lasted from the time of treatment until d 28 after treatment with the exception of those heifers that were treated earlier than 4 wk prepartum. For those heifers, the follow-up period was prolonged until parturition. For heifers that calved in 1997 and were in the same herds as the clinical cases in the current study, but were not treated for clinical mastitis within 30 d postpartum, the proportion that was culled within 30 d postpartum was 4.54% (168 out of 3700 heifers). Thus, the ratio between the proportion of heifers included in the present investigation that were culled and the proportion of nonmastitic heifers that were culled in 1997 was 2.40 (95% confidence interval, 2.01 to 2.87). In 1996 the proportion of corresponding nonmastitic heifers that were culled from these herds within 30 d postpartum was 4.00% (133 out of 3325 heifers), and the proportion of heifers that were culled during the same time interval of those that were treated for clinical mastitis within 14 d postpartum was 9.64% (61 out of 633 heifers). Thus, in 1996 the relative risk of being culled within 1 mo after treatment for those groups of heifers was 2.41 (95% confidence interval, 1.90 to 3.06) (i.e., almost identical to the ratio observed for the subsequent year). The follow-up period used for determination of culling rate of heifers included in the field study, as related to the time of parturition, varied among individual animals. However, the median final day of the follow-up period was d 30 postpartum, and for 97.7% of the heifers, the follow-up period was terminated between d 18 and 42 postpartum. For heifers with mastitis and for heifers without mastitis, the number of animals that were culled per day showed a very small and approximately linear decrease between d 18 and 42 postpartum (data not shown). Thus, the ratio that was calculated for 1997 should be considered as a good approximation of the relative risk of being culled because of mastitis during the first 30 d postpartum. If heifers that had been culled for reasons other than mastitis were excluded, the proportion of culled heifers was 10.54%. Thus, the relative risk of being culled for those heifers for which mastitis was the only cause or
73
main cause for culling was 2.32 (95% confidence interval, 1.94 to 2.79). Several studies (1, 2, 3, 4, 9) have reported that cows that have suffered an episode of clinical mastitis have a greater probability of being culled during some specified subsequent period than do nonmastitic cows of similar age. Erb et al. (3) reported that the odds of being culled before second calving was 5.2 times higher for primiparous cows that were treated for clinical mastitis than for primiparous cows that did not suffer from clinical mastitis. Within a path analysis framework, those researchers used a multivariate logistic regression model, and the effect of mastitis on culling was adjusted for the effects of dystocia, milk yield, and reproductive perfor˚ strand (1) found that the risk of mance. Bendixen and A being culled within 1 mo postpartum was 4.0 times higher in cows that experienced a clinical episode of mastitis during the first month of lactation than in cows without clinical mastitis in the current lactation period. Dohoo and Martin (2), who compared cows that received systemic treatment for clinical mastitis and cows that were not treated for any of several diseases recorded, reported a relative risk of being culled within 150 d postpartum of 30.8 for the former versus the latter group. The corresponding risk for cows with mastitis that received intramammary treatment only, relative to the disease-free group, was 3.6. Gro¨ hn et al. (4), who used survival analysis and included time-dependent covariates, found a relative risk of 1.9 for cows to be culled within 1 mo postpartum following an episode of mastitis compared with cows without mastitis during the first 2 mo postpartum. The estimated relative risk was adjusted for the effects of parity, current milk yield, and current conception status. Oltenacu et al. (9) found association between the occurrence of clinical mastitis in primiparous cows and increased risk of being culled before second calving (odds ratio of 1.8). Possible explanations for differences in the relative risk of being culled that were reported by different researchers are variations in the duration of the followup period and differences among the groups of cases or reference animals included in the various studies. For example, the average severity of the condition might have varied among the clinical cases of mastitis assigned to the different studies. Previous studies in which the relative risk was estimated included all animals that were culled following the mastitis episode in question. In the present field study, we also considered spearately those culled heifers for which mastitis was verified to be an important reason for culling. Reference groups that have been used were either cows or heifers without ˚ strand (1) and in the mastitis, as in Bendixen and A present study, or cows without mastitis and additional risk factors for being culled (2). Adjustment of analysis Journal of Dairy Science Vol. 83, No. 1, 2000
74
Journal of Dairy Science Vol. 83, No. 1, 2000
Only quarters (n = 1122) that were clinically affected at the time of treatment are included. Coagulase-positive staphylococci. 3 One quarter was nonfunctional at treatment. 4 Growth from secretion of CPS and Strep. dysgalactiae, CPS and A. pyogenes, or Strep. dysgalactiae and A. pyogenes. 5 Coagulase-negative staphylococci. 6 Streptococcus uberis (n = 24), Streptococcus sp. (n = 1), Enterococcus spp. (n = 3), coliforms other than E. coli (n = 3), Pasteurella haemolytica (n = 1), and Bacillus cereus (n = 1). 7 Bacteriologically negative samples and samples containing a mixed culture of environmental organisms. 8 Three quarters were nonfunctional at treatment. 2
1
(%) 15.5 3.9 0.8 3.1 4.7 5.4 66.7 (%) 15.2 3.0 9.1 3.0 6.1 0 63.6 (%) 21.1 4.2 7.0 4.2 11.3 5.6 46.5 (%) 3.4 2.7 2.7 6.1 6.8 6.8 71.4 (%) 40.0 0 6.7 13.3 0 0 40.0 (%) 15.4 8.5 11.4 5.5 7.0 4.0 48.3 (%) 30.4 10.3 6.5 8.3 5.5 5.7 33.4
(%) 53.1 18.8 3.1 0 0 0 25.0
No growth or mixed culture7 (n = 129)8 Various pathogens6 (n = 33) Escherichia coli (n = 71)
Nonfunctional quarter Infectious clinical mastitis Nonspecific clinical mastitis Infectious subclinical mastitis Nonspecific subclinical mastitis Latent infection Normal
for effects of other potential risk factors for being culled (3, 4, 9) also may influence the estimate of the relative risk of culling associated with mastitis. Economic factors that affect culling strategies, such as prices for milk, culled cows, and replacement heifers, may vary over time and among different countries or regions and may affect the risk estimates. Quarter secretion samples from 891 heifers were reexamined. Distribution of the heifers according to length of time between treatment and reexamination is shown in Table 2. In those heifers, 1122 quarters had been treated for clinical mastitis; Table 3 shows the distribution of diagnoses for these quarters at reexamination in relation to the bacteriological findings at treatment. Two hundred forty-nine quarters (22.2%) were nonfunctional, and the proportion of heifers in which at least one quarter was nonfunctional was 25.1%. Secretions from 521 of the quarters that were clinically affected at treatment were normal at reexamination. When the heifers that were culled because of mastitis before the stipulated time of reexamination and those heifers that were reexamined were considered together, 39.9% (521 out of 1305) of the quarters that were clinically affected at treatment were still in lactation and had normal secretion at reexamination. This cure rate was lower than were overall cure rates (between 47 and 58%) previously observed (7, 14) for clinically affected quarters of cows that were treated for mastitis during lactation. A 4- or 5-d antibacterial regimen that is commonly used in Norway was used in those studies, and quarters were reexamined approximately 1 mo after treatment. One factor that might have contributed to the lower cure rate for the heifers in the current study compared with cure rates observed in previous studies for lactating cows with clinical mastitis was that several heifers were treated prepartum or at the day of parturition. In many of those cases, the condition most likely had persisted for some time without being noticed by the farmer. In
CNS5 (n = 147)
One hundred nine heifers were culled before reexamination.
Combinations4 (n = 15)
1
CPS2 (n = 494)3
(%) 2.1 13.2 42.2 20.3 8.3 3.3 2.8 1.8 1.3 4.6
Diagnosis at reexamination
(n) 19 118 376 181 74 29 25 16 12 41
Bacteriological findings at treatment
(d) <24 24–26 27–29 30–32 33–35 36–38 39–41 42–44 45–47 >47
Heifers
Arcanobacterium pyogenes (n = 32)3
Treatment to reexamination
Streptococcus dysgalactiae (n = 201)3
Table 2. Distribution of 891 heifers with clinical mastitis1 according to the time interval between treatment and reexamination.
Table 3. Mammary quarter diagnoses approximately 1 mo after treatment for clinical mastitis in relation to organisms detected at treatment of heifers (n = 891) that suffered from clinical mastitis prepartum or within 14 d postpartum.1
WAAGE ET AL.
75
OUTCOME OF CLINICAL MASTITIS IN HEIFERS
such cases, the delayed initiation of antibacterial therapy would lead to increased risk of quarter damage and, thus, decreased probability of being cured. The percentage of nonfunctional quarters was highest and the percentage of completely cured quarters was lowest for those quarters that were infected with Arcanobacterium pyogenes at treatment (Table 3). Furthermore, cure rates were relatively low for quarters infected with coagulase-positive staphylococci (CPS) and were relatively high for quarters infected with coagulase-negative staphylococci. Of the 57 quarters that had a latent infection at reexamination, 56 quarters harbored CPS, and 1 quarter harbored Streptococcus dysgalactiae. Previous studies (7) have revealed differences in cure rates for quarters infected with different pathogens. Thus, the cure rate recorded in a particular study is likely to be influenced by the distribution of the different causal organisms within the group of heifers or cows under study. Of the quarters that were clinically affected at treatment, secretion was not obtained from 183 of those that were nonfunctional at reexamination. Table 4 shows the distribution of organisms found at reexamination in secretions from the remaining nonfunctional quarters and from the lactating quarters in which clinical mastitis persisted or subclinical mastitis was present in relation to the bacteriological findings at treatment. In secretions from 21% of the nonfunctional quarters and from almost 50% of the remaining mastitic quarters, no organism was detected or a mixed culture of environmental organisms was present. Both at treatment and at reexamination, CPS were the most prevalent pathogens in the
infected quarters. At reexamination, approximately 30% of those quarters that were originally infected with A. pyogenes or CPS were still infected with the same species. Of those quarters that were infected with organisms other than CPS and A. pyogenes at treatment, only a small proportion was infected with the original organism at reexamination. Inhibitory substances were detected in secretions from 21 of the quarters that had been clinically affected at treatment. In 17 cases the substance was a β-lactam; in 1 case sulfonamide was detected, and in 3 cases the substance, which was heat-stable, was not identified by the methods employed in the study. Eleven of the 21 quarters in which inhibitory substances were present were normal, and the secretions from 17 of the 21 quarters were bacteriologically negative. In addition to persistent IMI and chronic inflammation of quarters, the presence of chronic thelitis was a serious complication in many of the clinically affected quarters of the heifers in this study. Thelitis frequently was observed in the quarters that were nonfunctional. Furthermore, palpable abnormalities of the tissue lining the teat sinus or the streak canal, which indicated chronic thelitis, were recorded in 25 (28.7%) of the quarters with infectious clinical mastitis, in 15 (21.4%) of the quarters with nonspecific clinical mastitis, in 7 (5.1%) of the quarters with infectious or nonspecific subclinical mastitis, and in 5 (8.8%) of the quarters with a latent infection at the time of reexamination. Also, obvious clinical signs of thelitis were present in 18 (3.5%) of the quarters in which the secretion was normal at reexami-
Table 4. Bacteriological findings in secretions from mammary quarters of heifers (n = 716) sampled at the time of treatment for clinical mastitis (prepartum or within 14 d postpartum) and approximately 1 mo later.1 Bacteriological findings at treatment Bacteriological findings at reexamination CPS2 Strep. dysgalactiae A. pyogenes Combinations3 CNS4 E. coli Various pathogens7 No growth or mixed culture6
CPS2 (n = 355)
Streptococcus dysgalactiae (n = 174)
Arcanobacterium pyogenes (n = 17)
Combinations3 (n = 11)
CNS4 (n = 135)
Escherichia coli (n = 56)
Various pathogens5 (n = 28)
No growth or mixed culture6 (n = 106)
(%) 29.1 0.8 0.3 0.3 3.4 0 0.8
(%) 5.2 6.3 0.6 0 7.5 1.7 1.1
(%) 11.8 0 35.3 0 0 0 0
(%) 27.3 0 0 0 0 9.1 0
(%) 5.9 0 0 1.5 3.0 0 0
(%) 5.4 1.8 1.8 0 3.6 3.6 0
(%) 3.6 0 3.6 0 0 0 0
(%) 5.7 0.9 0.9 0 0.9 0 1.9
65.4
77.6
52.9
63.6
89.6
83.9
92.9
89.6
1
Only quarters (n = 882) with clinical mastitis at treatment and clinical or subclinical mastitis at reexamination are included. Coagulase-positive staphylococci. 3 Growth from secretion of CPS and Strep. dysgalactiae, CPS and A. pyogenes, or Strep. dysgalactiae and A. pyogenes. 4 Coagulase-negative staphylococci. 5 Streptococcus uberis (n = 21), Streptococcus sp. (n = 1), Enterococcus spp. (n = 2), coliforms other than E. coli (n = 3), and Pasteurella haemolytica (n = 1). 6 Bacteriologically negative samples and samples containing a mixed culture of environmental organisms. 7 Streptococcus agalactiae (n = 1), Strep. uberis (n = 1), Streptococcus spp. (n = 2), Enterococcus spp. (n = 2), and Proteus sp. (n = 1). 2
Journal of Dairy Science Vol. 83, No. 1, 2000
76
WAAGE ET AL.
nation. The recorded cases of thelitis were mainly those in which the changes were marked and, to some extent, interferred with the rate of milk removal from the particular quarter. In the present study, the consequences of clinical mastitis in heifers were assessed by the proportion of heifers that were culled within 1 mo after treatment and by the status of the udder quarters 1 mo after treatment. Most likely, additional effects also were present. The large proportion of quarters that was nonfunctional or exhibited clinical or subclinical signs of inflammation at reexamination indicated that many of the heifers that were still alive 1 mo after treatment were candidates for being culled later in the current lactation. In such heifers the production capacity and, under optimal feeding conditions, the observed milk yield most likely would have been reduced compared with nonmastitic primiparous cows. CONCLUSIONS In this study, 10.9% of heifers that were treated for clinical mastitis before parturition or within 14 d postpartum were culled within 1 mo after treatment, and mastitis was the only cause or the main cause for culling for 96% of those heifers. The risk for being culled within 1 mo postpartum was 2.4 times greater for heifers that experienced a clinical episode of mastitis than for nonmastitic heifers. At reexamination 1 mo after treatment, a large proportion (25%) of the unculled heifers had at least one quarter that was nonfunctional because of chronic inflammatory changes. Of the mammary quarters that were clinically affected at treatment, 22% were nonfunctional at the time of reexamination, and 26% were affected by clinical or subclinical mastitis. In approximately 25% of the lactating quarters that were still affected by clinical mastitis at reexamination, thelitis also was present. Forty percent of the quarters that were clinically affected at treatment had a normal secretion at reexamination, but chronic thelitis was observed in 3.5% of those cases. These findings indicate that clinical mastitis in heifers could lead to great economic losses for dairy farmers. The losses would mainly be caused by increased replacement costs following the involuntary culling of severely affected heifers and decreased production capacity for heifers in which chronic mastitis has led to partial or complete damage of affected quarters. In addition, clinical
Journal of Dairy Science Vol. 83, No. 1, 2000
cases left with chronic IMI caused by contagious pathogens (e.g., Staph. aureus) represent reservoirs from which pathogens may be spread to quarters of other heifers or cows in the herd. ACKNOWLEDGMENTS The authors thank the veterinary practitioners that participated in the study. The study was supported by the Research Council of Norway (Oslo, Norway). REFERENCES ˚ strand. 1989. Removal risks in Swed1 Bendixen, P. H., and D. B. A ish Friesian dairy cows according to parity, stage of lactation, and occurrence of clinical mastitis. Acta Vet. Scand. 30:37–42. 2 Dohoo I. R., and S. W. Martin. 1984. Disease, production and culling in Holstein-Friesian cows. V. Survivorship. Prev. Vet. Med. 2:771–784. 3 Erb, H. N., R. D. Smith, P. A. Oltenacu, C. L. Guard, R. B. Hillman, P. A. Powers, M. C. Smith, and M. E. White. 1985. Path model of reproductive disorders and performance, milk fever, mastitis, milk yield, and culling in Holstein cows. J. Dairy Sci. 68:3337– 3349. 4 Gro¨ hn, Y. T., S. W. Eicker, V. Ducrocq, and J. A. Hertl. 1998. Effect of diseases on the culling of Holstein dairy cows in New York State. J. Dairy Sci. 81:966–978. 5 International Dairy Federation. 1981. Laboratory Methods for Use in Mastitis Work. Document 132. Int. Dairy Fed., Brussels, Belgium. 6 International Dairy Federation. 1987. Bovine Mastitis. Definition and Guidelines for Diagnosis. Bull. 211. Int. Dairy Fed., Brussels, Belgium. 7 Jarp, J., H. P. Bugge, and S. Larsen. 1989. Clinical trial of three therapeutic regimens for bovine mastitis. Vet. Rec. 124:630–634. 8 Klastrup, O. 1975. Scandinavian recommendations on examination of quarter milk samples. Pages 49–52 in Proc. of Sem. on Mastitis Control, Reading, United Kingdom. Int. Dairy Fed., Brussels, Belgium. 9 Oltenacu, P. A., A. Frick, and B. Lindhe´ . 1990. Epidemiological study of several clinical diseases, reproductive performance and culling in primiparous Swedish cattle. Prev. Vet. Med. 9:59–74. 10 Radostits, O. M., D. C. Blood, and C. C. Gay. 1994. Page 614 in Veterinary Medicine. 8th ed. W. B. Saunders Co., London, United Kingdom. 11 Rothman, K. J. 1986. Pages 171–173 in Modern Epidemiology. Little, Brown and Co., Boston, MA. 12 Shearer, J. K., and R. J. Harmon. 1993. Mastitis in heifers. Vet. Clin. North Am. Food Anim. Pract. 9:583–595. 13 State Veterinary Laboratories of Norway. 1993. Routines for Mastitis Diagnostics. State Vet. Lab., Oslo, Norway. 14 Waage, S. 1997. Comparison of two regimens for the treatment of clinical bovine mastitis caused by bacteria sensitive to penicillin. Vet. Rec. 141:616–620. 15 Waage, S., T. Mørk, A. Røros, D. Aasland, A. Hunshamar, and S. A. Ødegaard. 1999. Bacteria associated with clinical mastitis in dairy heifers. J. Dairy Sci. 82:712–719. 16 Waage, S., S. Sviland, and S. A. Ødegaard. 1998. Identification of risk factors for clinical mastitis in dairy heifers. J. Dairy Sci. 81:1275–1284.
ABSTRACT Heifers that were treated for clinical mastitis prior to parturition or within 14 d postpartum were reexamined approximately 1 mo after treatment. Clinical examination of the heifers and microbiological examination of quarter milk samples were carried out on both occasions. Of the 1000 heifers included in the study, 10.9% were culled within 28 d after treatment. Udder damage caused by mastitis was the only or main reason for culling in 96% of those heifers. In comparison, 4.5% of nonmastitic heifers from the same herds were culled within 30 d postpartum. Twenty-five percent of those heifers that were not culled at d 28 after treatment had at least one nonfunctional quarter at that time. One thousand one hundred twenty-two quarters that were clinically affected at the time of treatment were reexamined; 22% were nonfunctional, 14% were still affected by clinical mastitis, 12% had subclinical mastitis, 5% had a latent infection with coagulase-positive staphylococci or Streptococcus dysgalactiae, and 46% were bacteriologically negative and had a normal cell count at the time of reexamination. High percentages of nonfunctional quarters were observed among those quarters that were infected with Arcanobacterium pyogenes or with coagulasepositive staphylococci at treatment. When all quarters that were clinically affected at treatment were considered, 40% of quarters were cured and were still in lactation at reexamination. Quarters infected with coagulasenegative staphylococci had a higher cure rate than quarters infected with other organisms. At reexamination, clinical signs of thelitis were observed in many of those quarters that were nonfunctional following the episode of clinical mastitis and also in 25% of lactating quarters in which clinical mastitis persisted. (Key words: mastitis, heifer, outcome, culling)
Received March 17, 1999. Accepted August 26, 1999. 1 Corresponding author. 2000 J Dairy Sci 83:70–76
70
Abbreviation key: CPS = coagulase-positive staphylococci. INTRODUCTION Frequent occurrence of bovine mastitis and the immediate expenses and pathoanatomic complications associated with the disease lead to great financial losses for the farmers. Even cases that receive an appropriate antibacterial treatment at an early stage of a clinical IMI are frequently left with permanent tissue damage of parts of the affected quarter. The consequences are decreased milk production, reduced milk quality, and occasionally a completely nonfunctional quarter. Cows that suffer such complications following an episode of clinical mastitis are at a greater risk of being culled before their optimal age for disposal than are nonmastitic cows of similar age (4). Subsequent to a clinical episode, a chronic subclinical or clinical infection may be established, particularly when Staphylococcus aureus is the causal organism (14). Thus a reservoir of mammary pathogens is formed which may infect quarters of other cows in the herd. Many dairy heifers are affected by clinical mastitis around parturition. According to records of disease data in the Norwegian Health Card System for Cattle, 13.3% of dairy heifers that calved in Norway in 1995 were treated for clinical mastitis within 14 d postpartum (S. Waage, 1999, unpublished data). In general, the prospects for the complete recovery of affected secretory and ductular tissue in quarters of heifers that experience an episode of clinical mastitis are considered to be relatively poor (12). However, no comprehensive studies seem to have been conducted to assess the outcome of clinical mastitis in heifers. In the present study, dairy heifers that were treated for clinical mastitis prepartum or within 14 d postpartum were reexamined approximately 1 mo after treatment to determine the status of the animals and their udder quarters.
71
OUTCOME OF CLINICAL MASTITIS IN HEIFERS
MATERIALS AND METHODS Definitions Cases of mastitis were classified as clinical or subclinical in accordance with the definitions given by the International Dairy Federation (6). Udder quarters in which lactation ceased because of chronic clinical mastitis were classified separately as nonfunctional quarters. Infectious (clinical or subclinical) mastitis was present when pathogens and inflammatory changes were detected in the secretion (6). Nonspecific (clinical or subclinical) mastitis was present when there were inflammatory changes but no pathogens in the secretion (6). A latent infection was present when the secretion contained pathogens but had a normal cell count (6). The distinction between a normal and a high cell count of the secretion, as estimated by the California Mastitis Test, was based on Scandinavian recommendations for interpretation of results of the California Mastitis Test (8). The reaction of the secretion of the particular quarter as well as the reactions of the secretions of the remaining quarters of the cow were taken into account. Thelitis was considered to be present when the wall of the teat cistern was thickened and hardened, as felt by clinical examination of the teat (10). Study Design A field study was carried out in 24 veterinary districts in Norway. Approximately 100 veterinary surgeons were involved. The general arrangement of the study and the location of the veterinary districts in which the study took place have been described previously (15). The study included heifers that, during a particular year (1997), were treated for clinical mastitis prepartum or within 14 d postpartum. When a heifer experienced more than one episode of clinical mastitis, only the first episode was considered. The veterinary practitioner supplied information, on a standard form, from the heifer and from the herd to which she belonged and collected secretion samples from all quarters. Microbiological examination of secretion was carried out at the Mastitis Laboratory of the National Veterinary Institute in Oslo, Norway. According to the agreement with the veterinary practitioners, reexamination and resampling of cases were performed approximately 4 wk after the initial examination. Routinely, 3 wk after treatment the Mastitis Laboratory mailed to the veterinary practitioners a reminding letter, in which the name of the farmer, identity number of the heifer, and optimal date of reexamination (28 d after treatment) were given. An exception was made for heifers that were treated earlier than 4 wk prepartum; for those heifers reexamination was delayed until just before or after parturition. For heifers that had been
culled before the time of reexamination, the date of and reason for culling were reported. For those heifers that were not culled 4 wk after treatment, quarter milk samples were collected, and data on the clinical condition of the heifer were recorded on a standard form. The culling rate for heifers that were treated for clinical mastitis was compared with the culling rate for nonmastitic heifers in the same herds. Clinical Cases and Control Animals The number of heifers that were treated for clinical mastitis was 1040, and a total of 1361 quarters was clinically affected. Bacteriological findings in quarter secretion samples collected from these heifers at the time of treatment have been published previously (15). No information was received on the course of the disease for 40 heifers, for which 56 quarters were clinically affected at treatment. Thus, the final material used for assessment of the outcome of clinical mastitis on individual animals included 1000 heifers. For these heifers, a total of 1305 quarters were clinically affected at treatment. However, heifers that were culled within 28 d after treatment were not resampled before culling, and the assessment of the outcome of quarters was based on the bacteriological reexamination of secretions from those heifers that were still alive 4 wk after treatment. Of the quarters that were reexamined, 1122 had been clinically affected at the time of treatment. Initially, antibacterial therapy was administered to 925 of the 1000 heifers; 827 received both systemic and intramammary treatment, 88 received systemic treatment alone, and 10 received intramammary treatment alone. Of those heifers that received systemic treatment, an intramuscular injection of penicillin G was given to 892 heifers, a combination of penicillin G and dihydrostreptomycin was given to 11 heifers, oxytetracycline was given to 6 heifers, and various other antibacterial drugs were given to 6 heifers. For intramammary treatment, 581 heifers received a combination of penicillin G and dihydrostreptomycin, 251 heifers received penicillin G alone, and 5 heifers received various other drugs. Antibacterial therapy was not administered when such therapy was considered useless, for example when the clinically affected quarter was nonfunctional due to chronic inflammatory changes. Dates of parturition and treatment or resampling were reported by the veterinary practitioners on a standard form. Date of parturition was missing for some heifers and was later obtained from the database of the National Production Recording Scheme or from the farmer. Of the 1000 heifers included in the study, 14% were treated at least 1 d prior to parturition, 17% were treated on the day of parturition, and 69% were treated between d 1 Journal of Dairy Science Vol. 83, No. 1, 2000
72
WAAGE ET AL.
Table 1. Distribution of the 709 dairy herds from which the 1000 clinical cases in the study originated, according to the number of heifers they supplied. Heifers per herd
Herds
(n) 1 2 3 4 5 7
(n) 506 141 44 12 5 1
and 14 postpartum. Information on the clinical condition of the heifer was given on the form. Quarter signs that were recorded were presence or absence of acute or chronic signs of inflammation, of a nonfunctional quarter, of a recent or chronic teat lesion, and of visually abnormal secretion. A California Mastitis Test was performed on quarter secretions collected postpartum, and the result for each quarter was reported on the form. Of the 709 herds from which the heifers with clinical mastitis originated (Table 1), 669 were enrolled in the National Production Recording Scheme in 1997 (657 in 1996). In these herds, the Norwegian Health Card System for Cattle (16) was in operation. Disease data records from this system made it possible to identify heifers that experienced episodes of clinical mastitis and to date the occurrence of episodes in relation to time of parturition. For all culled heifers and cows in herds enrolled in the National Production Recording Scheme, date of and reason for culling are routinely recorded. To compare culling rates among heifers with and without clinical mastitis, records were studied for heifers that calved in 1996 and for heifers that calved in 1997 in herds from which the clinical cases originated. For each of those years, the proportion of heifers that were culled within 30 d postpartum was determined for those heifers that had not been treated for clinical mastitis prior to d 30 postpartum (1996, n = 3325 heifers; 1997, n = 3700 heifers). For 1996, the proportion of heifers that were culled within 30 d postpartum was also determined for those heifers that had been treated for clinical mastitis within 14 d postpartum (n = 633). Ninety-nine percent of the heifers that were clinically affected and 97% of the control heifers were of the Norwegian Red Cattle breed. The majority of the remaining heifers were cross-breeds of Norwegian Red Cattle and various beef cattle breeds.
titioners. In each envelope was a box containing four sterile, 10-ml plastic tubes and a standard form. The tubes were numbered to identify the particular quarters. Samples were collected from all quarters, except for those from which no secretion was obtained. As soon as possible after sampling the envelope was mailed to the Mastitis Laboratory. The veterinary practitioner received a fee for each case that was sampled or resampled. Laboratory Methods Methods and procedures used for the examination of quarter secretion samples were as described by the State Veterinary Laboratories of Norway (13) and in agreement with the recommendations of the International Dairy Federation (5). Secretions were characterized by appearance. The cell content in postpartum samples was estimated using the California Mastitis Test. Quarter secretions (0.01 ml) were plated on blood agar containing 5% washed bovine erythrocytes. Cultures were examined following incubation under aerobic conditions at 37°C for 24 and 48 h. When growth was not detected after incubation of the primary culture for 24 h, the original secretion sample was preincubated for 4 h at 37°C. Then an increased inoculum (0.05 ml) was plated on each of three blood agar plates and incubated at 37°C for 24 h under aerobic conditions, in 5% CO2 atmosphere, and under anaerobic conditions, respectively. Recommended methods (5) were used to identify bacteria. All secretion samples were examined for the presence of inhibitory substances using agar diffusion on MuellerHinton agar, which contained Micrococcus luteus ATCC 9341 as the test bacteria. Secretions in which inhibitory substances were detected were subjected to further examinations to determine whether a β-lactam (using penicillinase), sulfonamide (using para-aminobenzoic acid), or some unspecific heat-sensitive, inhibitory substance was present (warming to 80°C before retesting for inhibitory effect). Statistics Standard deviation of relative risk was estimated according to Rothman (11), and a confidence interval for relative risk was obtained by taking antilogarithms of confidence limits calculated for the logarithmic transformed value of relative risk (11). RESULTS AND DISCUSSION
Sample Collection Envelopes that were suitable for mailing quarter secretion samples were distributed to the veterinary pracJournal of Dairy Science Vol. 83, No. 1, 2000
Of the 1000 heifers with clinical mastitis, 109 (10.9%) were culled within 4 wk after the initial treatment. Four heifers were culled for reasons other than mastitis. For
OUTCOME OF CLINICAL MASTITIS IN HEIFERS
the remaining 105 heifers, mastitis was the only reason or the main reason for culling. Four of these heifers died or were euthanatized because of gangrenous mastitis accompanied by a severe systemic reaction, and 101 heifers were slaughtered. The heifers that were slaughtered suffered from chronic mastitis, frequently combined with marked chronic thelitis. According to the reports from the veterinary practitioners, 36 of the heifers that were slaughtered had at least one nonfunctional quarter. The 1000 heifers were treated between 208 d prepartum and 14 d postpartum, and median time of treatment was 1 d postpartum. Determination of culling rate was based on a follow-up period that lasted from the time of treatment until d 28 after treatment with the exception of those heifers that were treated earlier than 4 wk prepartum. For those heifers, the follow-up period was prolonged until parturition. For heifers that calved in 1997 and were in the same herds as the clinical cases in the current study, but were not treated for clinical mastitis within 30 d postpartum, the proportion that was culled within 30 d postpartum was 4.54% (168 out of 3700 heifers). Thus, the ratio between the proportion of heifers included in the present investigation that were culled and the proportion of nonmastitic heifers that were culled in 1997 was 2.40 (95% confidence interval, 2.01 to 2.87). In 1996 the proportion of corresponding nonmastitic heifers that were culled from these herds within 30 d postpartum was 4.00% (133 out of 3325 heifers), and the proportion of heifers that were culled during the same time interval of those that were treated for clinical mastitis within 14 d postpartum was 9.64% (61 out of 633 heifers). Thus, in 1996 the relative risk of being culled within 1 mo after treatment for those groups of heifers was 2.41 (95% confidence interval, 1.90 to 3.06) (i.e., almost identical to the ratio observed for the subsequent year). The follow-up period used for determination of culling rate of heifers included in the field study, as related to the time of parturition, varied among individual animals. However, the median final day of the follow-up period was d 30 postpartum, and for 97.7% of the heifers, the follow-up period was terminated between d 18 and 42 postpartum. For heifers with mastitis and for heifers without mastitis, the number of animals that were culled per day showed a very small and approximately linear decrease between d 18 and 42 postpartum (data not shown). Thus, the ratio that was calculated for 1997 should be considered as a good approximation of the relative risk of being culled because of mastitis during the first 30 d postpartum. If heifers that had been culled for reasons other than mastitis were excluded, the proportion of culled heifers was 10.54%. Thus, the relative risk of being culled for those heifers for which mastitis was the only cause or
73
main cause for culling was 2.32 (95% confidence interval, 1.94 to 2.79). Several studies (1, 2, 3, 4, 9) have reported that cows that have suffered an episode of clinical mastitis have a greater probability of being culled during some specified subsequent period than do nonmastitic cows of similar age. Erb et al. (3) reported that the odds of being culled before second calving was 5.2 times higher for primiparous cows that were treated for clinical mastitis than for primiparous cows that did not suffer from clinical mastitis. Within a path analysis framework, those researchers used a multivariate logistic regression model, and the effect of mastitis on culling was adjusted for the effects of dystocia, milk yield, and reproductive perfor˚ strand (1) found that the risk of mance. Bendixen and A being culled within 1 mo postpartum was 4.0 times higher in cows that experienced a clinical episode of mastitis during the first month of lactation than in cows without clinical mastitis in the current lactation period. Dohoo and Martin (2), who compared cows that received systemic treatment for clinical mastitis and cows that were not treated for any of several diseases recorded, reported a relative risk of being culled within 150 d postpartum of 30.8 for the former versus the latter group. The corresponding risk for cows with mastitis that received intramammary treatment only, relative to the disease-free group, was 3.6. Gro¨ hn et al. (4), who used survival analysis and included time-dependent covariates, found a relative risk of 1.9 for cows to be culled within 1 mo postpartum following an episode of mastitis compared with cows without mastitis during the first 2 mo postpartum. The estimated relative risk was adjusted for the effects of parity, current milk yield, and current conception status. Oltenacu et al. (9) found association between the occurrence of clinical mastitis in primiparous cows and increased risk of being culled before second calving (odds ratio of 1.8). Possible explanations for differences in the relative risk of being culled that were reported by different researchers are variations in the duration of the followup period and differences among the groups of cases or reference animals included in the various studies. For example, the average severity of the condition might have varied among the clinical cases of mastitis assigned to the different studies. Previous studies in which the relative risk was estimated included all animals that were culled following the mastitis episode in question. In the present field study, we also considered spearately those culled heifers for which mastitis was verified to be an important reason for culling. Reference groups that have been used were either cows or heifers without ˚ strand (1) and in the mastitis, as in Bendixen and A present study, or cows without mastitis and additional risk factors for being culled (2). Adjustment of analysis Journal of Dairy Science Vol. 83, No. 1, 2000
74
Journal of Dairy Science Vol. 83, No. 1, 2000
Only quarters (n = 1122) that were clinically affected at the time of treatment are included. Coagulase-positive staphylococci. 3 One quarter was nonfunctional at treatment. 4 Growth from secretion of CPS and Strep. dysgalactiae, CPS and A. pyogenes, or Strep. dysgalactiae and A. pyogenes. 5 Coagulase-negative staphylococci. 6 Streptococcus uberis (n = 24), Streptococcus sp. (n = 1), Enterococcus spp. (n = 3), coliforms other than E. coli (n = 3), Pasteurella haemolytica (n = 1), and Bacillus cereus (n = 1). 7 Bacteriologically negative samples and samples containing a mixed culture of environmental organisms. 8 Three quarters were nonfunctional at treatment. 2
1
(%) 15.5 3.9 0.8 3.1 4.7 5.4 66.7 (%) 15.2 3.0 9.1 3.0 6.1 0 63.6 (%) 21.1 4.2 7.0 4.2 11.3 5.6 46.5 (%) 3.4 2.7 2.7 6.1 6.8 6.8 71.4 (%) 40.0 0 6.7 13.3 0 0 40.0 (%) 15.4 8.5 11.4 5.5 7.0 4.0 48.3 (%) 30.4 10.3 6.5 8.3 5.5 5.7 33.4
(%) 53.1 18.8 3.1 0 0 0 25.0
No growth or mixed culture7 (n = 129)8 Various pathogens6 (n = 33) Escherichia coli (n = 71)
Nonfunctional quarter Infectious clinical mastitis Nonspecific clinical mastitis Infectious subclinical mastitis Nonspecific subclinical mastitis Latent infection Normal
for effects of other potential risk factors for being culled (3, 4, 9) also may influence the estimate of the relative risk of culling associated with mastitis. Economic factors that affect culling strategies, such as prices for milk, culled cows, and replacement heifers, may vary over time and among different countries or regions and may affect the risk estimates. Quarter secretion samples from 891 heifers were reexamined. Distribution of the heifers according to length of time between treatment and reexamination is shown in Table 2. In those heifers, 1122 quarters had been treated for clinical mastitis; Table 3 shows the distribution of diagnoses for these quarters at reexamination in relation to the bacteriological findings at treatment. Two hundred forty-nine quarters (22.2%) were nonfunctional, and the proportion of heifers in which at least one quarter was nonfunctional was 25.1%. Secretions from 521 of the quarters that were clinically affected at treatment were normal at reexamination. When the heifers that were culled because of mastitis before the stipulated time of reexamination and those heifers that were reexamined were considered together, 39.9% (521 out of 1305) of the quarters that were clinically affected at treatment were still in lactation and had normal secretion at reexamination. This cure rate was lower than were overall cure rates (between 47 and 58%) previously observed (7, 14) for clinically affected quarters of cows that were treated for mastitis during lactation. A 4- or 5-d antibacterial regimen that is commonly used in Norway was used in those studies, and quarters were reexamined approximately 1 mo after treatment. One factor that might have contributed to the lower cure rate for the heifers in the current study compared with cure rates observed in previous studies for lactating cows with clinical mastitis was that several heifers were treated prepartum or at the day of parturition. In many of those cases, the condition most likely had persisted for some time without being noticed by the farmer. In
CNS5 (n = 147)
One hundred nine heifers were culled before reexamination.
Combinations4 (n = 15)
1
CPS2 (n = 494)3
(%) 2.1 13.2 42.2 20.3 8.3 3.3 2.8 1.8 1.3 4.6
Diagnosis at reexamination
(n) 19 118 376 181 74 29 25 16 12 41
Bacteriological findings at treatment
(d) <24 24–26 27–29 30–32 33–35 36–38 39–41 42–44 45–47 >47
Heifers
Arcanobacterium pyogenes (n = 32)3
Treatment to reexamination
Streptococcus dysgalactiae (n = 201)3
Table 2. Distribution of 891 heifers with clinical mastitis1 according to the time interval between treatment and reexamination.
Table 3. Mammary quarter diagnoses approximately 1 mo after treatment for clinical mastitis in relation to organisms detected at treatment of heifers (n = 891) that suffered from clinical mastitis prepartum or within 14 d postpartum.1
WAAGE ET AL.
75
OUTCOME OF CLINICAL MASTITIS IN HEIFERS
such cases, the delayed initiation of antibacterial therapy would lead to increased risk of quarter damage and, thus, decreased probability of being cured. The percentage of nonfunctional quarters was highest and the percentage of completely cured quarters was lowest for those quarters that were infected with Arcanobacterium pyogenes at treatment (Table 3). Furthermore, cure rates were relatively low for quarters infected with coagulase-positive staphylococci (CPS) and were relatively high for quarters infected with coagulase-negative staphylococci. Of the 57 quarters that had a latent infection at reexamination, 56 quarters harbored CPS, and 1 quarter harbored Streptococcus dysgalactiae. Previous studies (7) have revealed differences in cure rates for quarters infected with different pathogens. Thus, the cure rate recorded in a particular study is likely to be influenced by the distribution of the different causal organisms within the group of heifers or cows under study. Of the quarters that were clinically affected at treatment, secretion was not obtained from 183 of those that were nonfunctional at reexamination. Table 4 shows the distribution of organisms found at reexamination in secretions from the remaining nonfunctional quarters and from the lactating quarters in which clinical mastitis persisted or subclinical mastitis was present in relation to the bacteriological findings at treatment. In secretions from 21% of the nonfunctional quarters and from almost 50% of the remaining mastitic quarters, no organism was detected or a mixed culture of environmental organisms was present. Both at treatment and at reexamination, CPS were the most prevalent pathogens in the
infected quarters. At reexamination, approximately 30% of those quarters that were originally infected with A. pyogenes or CPS were still infected with the same species. Of those quarters that were infected with organisms other than CPS and A. pyogenes at treatment, only a small proportion was infected with the original organism at reexamination. Inhibitory substances were detected in secretions from 21 of the quarters that had been clinically affected at treatment. In 17 cases the substance was a β-lactam; in 1 case sulfonamide was detected, and in 3 cases the substance, which was heat-stable, was not identified by the methods employed in the study. Eleven of the 21 quarters in which inhibitory substances were present were normal, and the secretions from 17 of the 21 quarters were bacteriologically negative. In addition to persistent IMI and chronic inflammation of quarters, the presence of chronic thelitis was a serious complication in many of the clinically affected quarters of the heifers in this study. Thelitis frequently was observed in the quarters that were nonfunctional. Furthermore, palpable abnormalities of the tissue lining the teat sinus or the streak canal, which indicated chronic thelitis, were recorded in 25 (28.7%) of the quarters with infectious clinical mastitis, in 15 (21.4%) of the quarters with nonspecific clinical mastitis, in 7 (5.1%) of the quarters with infectious or nonspecific subclinical mastitis, and in 5 (8.8%) of the quarters with a latent infection at the time of reexamination. Also, obvious clinical signs of thelitis were present in 18 (3.5%) of the quarters in which the secretion was normal at reexami-
Table 4. Bacteriological findings in secretions from mammary quarters of heifers (n = 716) sampled at the time of treatment for clinical mastitis (prepartum or within 14 d postpartum) and approximately 1 mo later.1 Bacteriological findings at treatment Bacteriological findings at reexamination CPS2 Strep. dysgalactiae A. pyogenes Combinations3 CNS4 E. coli Various pathogens7 No growth or mixed culture6
CPS2 (n = 355)
Streptococcus dysgalactiae (n = 174)
Arcanobacterium pyogenes (n = 17)
Combinations3 (n = 11)
CNS4 (n = 135)
Escherichia coli (n = 56)
Various pathogens5 (n = 28)
No growth or mixed culture6 (n = 106)
(%) 29.1 0.8 0.3 0.3 3.4 0 0.8
(%) 5.2 6.3 0.6 0 7.5 1.7 1.1
(%) 11.8 0 35.3 0 0 0 0
(%) 27.3 0 0 0 0 9.1 0
(%) 5.9 0 0 1.5 3.0 0 0
(%) 5.4 1.8 1.8 0 3.6 3.6 0
(%) 3.6 0 3.6 0 0 0 0
(%) 5.7 0.9 0.9 0 0.9 0 1.9
65.4
77.6
52.9
63.6
89.6
83.9
92.9
89.6
1
Only quarters (n = 882) with clinical mastitis at treatment and clinical or subclinical mastitis at reexamination are included. Coagulase-positive staphylococci. 3 Growth from secretion of CPS and Strep. dysgalactiae, CPS and A. pyogenes, or Strep. dysgalactiae and A. pyogenes. 4 Coagulase-negative staphylococci. 5 Streptococcus uberis (n = 21), Streptococcus sp. (n = 1), Enterococcus spp. (n = 2), coliforms other than E. coli (n = 3), and Pasteurella haemolytica (n = 1). 6 Bacteriologically negative samples and samples containing a mixed culture of environmental organisms. 7 Streptococcus agalactiae (n = 1), Strep. uberis (n = 1), Streptococcus spp. (n = 2), Enterococcus spp. (n = 2), and Proteus sp. (n = 1). 2
Journal of Dairy Science Vol. 83, No. 1, 2000
76
WAAGE ET AL.
nation. The recorded cases of thelitis were mainly those in which the changes were marked and, to some extent, interferred with the rate of milk removal from the particular quarter. In the present study, the consequences of clinical mastitis in heifers were assessed by the proportion of heifers that were culled within 1 mo after treatment and by the status of the udder quarters 1 mo after treatment. Most likely, additional effects also were present. The large proportion of quarters that was nonfunctional or exhibited clinical or subclinical signs of inflammation at reexamination indicated that many of the heifers that were still alive 1 mo after treatment were candidates for being culled later in the current lactation. In such heifers the production capacity and, under optimal feeding conditions, the observed milk yield most likely would have been reduced compared with nonmastitic primiparous cows. CONCLUSIONS In this study, 10.9% of heifers that were treated for clinical mastitis before parturition or within 14 d postpartum were culled within 1 mo after treatment, and mastitis was the only cause or the main cause for culling for 96% of those heifers. The risk for being culled within 1 mo postpartum was 2.4 times greater for heifers that experienced a clinical episode of mastitis than for nonmastitic heifers. At reexamination 1 mo after treatment, a large proportion (25%) of the unculled heifers had at least one quarter that was nonfunctional because of chronic inflammatory changes. Of the mammary quarters that were clinically affected at treatment, 22% were nonfunctional at the time of reexamination, and 26% were affected by clinical or subclinical mastitis. In approximately 25% of the lactating quarters that were still affected by clinical mastitis at reexamination, thelitis also was present. Forty percent of the quarters that were clinically affected at treatment had a normal secretion at reexamination, but chronic thelitis was observed in 3.5% of those cases. These findings indicate that clinical mastitis in heifers could lead to great economic losses for dairy farmers. The losses would mainly be caused by increased replacement costs following the involuntary culling of severely affected heifers and decreased production capacity for heifers in which chronic mastitis has led to partial or complete damage of affected quarters. In addition, clinical
Journal of Dairy Science Vol. 83, No. 1, 2000
cases left with chronic IMI caused by contagious pathogens (e.g., Staph. aureus) represent reservoirs from which pathogens may be spread to quarters of other heifers or cows in the herd. ACKNOWLEDGMENTS The authors thank the veterinary practitioners that participated in the study. The study was supported by the Research Council of Norway (Oslo, Norway). REFERENCES ˚ strand. 1989. Removal risks in Swed1 Bendixen, P. H., and D. B. A ish Friesian dairy cows according to parity, stage of lactation, and occurrence of clinical mastitis. Acta Vet. Scand. 30:37–42. 2 Dohoo I. R., and S. W. Martin. 1984. Disease, production and culling in Holstein-Friesian cows. V. Survivorship. Prev. Vet. Med. 2:771–784. 3 Erb, H. N., R. D. Smith, P. A. Oltenacu, C. L. Guard, R. B. Hillman, P. A. Powers, M. C. Smith, and M. E. White. 1985. Path model of reproductive disorders and performance, milk fever, mastitis, milk yield, and culling in Holstein cows. J. Dairy Sci. 68:3337– 3349. 4 Gro¨ hn, Y. T., S. W. Eicker, V. Ducrocq, and J. A. Hertl. 1998. Effect of diseases on the culling of Holstein dairy cows in New York State. J. Dairy Sci. 81:966–978. 5 International Dairy Federation. 1981. Laboratory Methods for Use in Mastitis Work. Document 132. Int. Dairy Fed., Brussels, Belgium. 6 International Dairy Federation. 1987. Bovine Mastitis. Definition and Guidelines for Diagnosis. Bull. 211. Int. Dairy Fed., Brussels, Belgium. 7 Jarp, J., H. P. Bugge, and S. Larsen. 1989. Clinical trial of three therapeutic regimens for bovine mastitis. Vet. Rec. 124:630–634. 8 Klastrup, O. 1975. Scandinavian recommendations on examination of quarter milk samples. Pages 49–52 in Proc. of Sem. on Mastitis Control, Reading, United Kingdom. Int. Dairy Fed., Brussels, Belgium. 9 Oltenacu, P. A., A. Frick, and B. Lindhe´ . 1990. Epidemiological study of several clinical diseases, reproductive performance and culling in primiparous Swedish cattle. Prev. Vet. Med. 9:59–74. 10 Radostits, O. M., D. C. Blood, and C. C. Gay. 1994. Page 614 in Veterinary Medicine. 8th ed. W. B. Saunders Co., London, United Kingdom. 11 Rothman, K. J. 1986. Pages 171–173 in Modern Epidemiology. Little, Brown and Co., Boston, MA. 12 Shearer, J. K., and R. J. Harmon. 1993. Mastitis in heifers. Vet. Clin. North Am. Food Anim. Pract. 9:583–595. 13 State Veterinary Laboratories of Norway. 1993. Routines for Mastitis Diagnostics. State Vet. Lab., Oslo, Norway. 14 Waage, S. 1997. Comparison of two regimens for the treatment of clinical bovine mastitis caused by bacteria sensitive to penicillin. Vet. Rec. 141:616–620. 15 Waage, S., T. Mørk, A. Røros, D. Aasland, A. Hunshamar, and S. A. Ødegaard. 1999. Bacteria associated with clinical mastitis in dairy heifers. J. Dairy Sci. 82:712–719. 16 Waage, S., S. Sviland, and S. A. Ødegaard. 1998. Identification of risk factors for clinical mastitis in dairy heifers. J. Dairy Sci. 81:1275–1284.