- Email: [email protected]
Characteristics of neospora caninum-associated abortion storms in dairy herds in The Netherlands (1995 to1997)
ELSEVIER
CHARACTERISTICS OF Neosnora caninum-ASSOCIATED ABORTION IN DAIRY HERDS IN THE NETHERLANDS (1995 to1997) W. W~uda,‘~C. J. M. Bartels,‘and
STORMS
A. R. Moen’
‘Animal Health Service Drachten, The Netherlands *Department of Herd Health and Reproduction Utrecht University, The Netherlands Received for publication: Accepted:
23 July 9 April
1998 1999
ABSTRACT Abortion storms in 50 dairy herds in The Netherlands were reported in which there was a strong association with Neosnora caninum-infection. The duration of the abortion storms ranged from 6 to 65 d (mean 41.5 d). The cumulative proportion of aborting cows ranged from 0.11 to 0.57 (mean 0.26) of the animals at risk. An apparent seasonal infhtence was noted as most abortion storms occurred during the summer and early autumn. The prevalence of antibodies to N. caninum in 50 herds which had had an abortion storm was compared with that of 100 control herds which had no history of an abortion storm. Seroprevalence was estimated by testing a 20% cross sectional herd sample using a tachyzoite lysate-based ELISA method. Seroprevalence in case herds (range 17 to 87%, mean 5 1.5%) was significantly higher than that in control herds (range 0 to 53%, mean 13.9%). For most herds the seroprevalence levels were equal across all age groups, which suggests that the infection had been perpetuated by vertical transmission. In these herds, the abortion storms appeared to be induced by factors causing recrudescence of a N. caninuminfection in chronically infected animals rather than being the result of a recent introduction. In 6 case herds the seroprevalence in the dairy cows was significantly higher than in the young stock, which may have been attributable to superimposed postnatal infection. 0 1999 by Elawier Science Inc.
Key words: Neosnora caninum, bovine, abortion, seroprevalence, transmission, epidemiology
Acknowledgments The authors thank all the dairy farmers and their veterinary practitioners for participation; G. H. A. Borst and J. H. Vos for referral of cases; J. Brinkhof, A. Damsma, G. Koning, A. van der Meulen, M. Swinkels and H. Westra for technical laboratory assistance; and P. G. M. Versluys for preparing Figure 1. We are indebted to Y. H. Schukken for advise and help with the study design and to R. N. Zadoks for valuable comments on the manuscript. ‘Correspondence and reprint requests: Animal Health Service, P. 0. Box, 9200 AJ Drachten. The Netherlands. Phone: 3 1 5 12 570 700 Fax: 3 1 512 520 013 E-mail: Theriogenology 52:233-245. 0 1999 by Elsevier Science
1999 Inc.
0093~691X&9/$-see front matter PII SOO93-691X(99)001259
234 INTRODUCTION During the last decade Neosnora caninum has been recognized as an important cause of abortion in cattle worldwide (for review see: 9). The infection has been associated with endemic and epidemic abortions in dairy herds (1, 4, 14, 15, 21, 22, 24, 28, 29). The importance of neosporosis for the Dutch dairy industry is reflected in the proportion of submitted fetuses in which the infection is diagnosed. After the first cases of N. caninum abortion in cattle had been recognized in the early nineties (26), yearly 15 to 20% of aborted calves submitted to the Animal Health Service laboratories were found to have histological lesions compatible with N_. caninum infection (28). This makes N. caninum the most recognized cause of abortion in cattle in The Netherlands. Vertical transmission from cow to calf has been well established (2, 5, 18, 27) and may contribute significantly to the persistence of the infection in the herd (7, 18). Recently it has been demonstrated that dogs can act as definitive hosts of N. caninum (15). Three of 4 dogs that were fed mouse tissues containing N. caninum tissue cysts shed fecal oocysts. Mice that received feces from these dogs were shown to be infected by N. caninum. The implications of these experimental findings for the epidemiology bovine neosporosis have yet to be investigated. Case reports of N. caninum associated abortion storms have contributed to the knowledge of the epidemiology of the infection by providing circumstantial evidence for a point source infection (14, 16,22,29). Recently, Thurmond et al. (24) reported a study of 20 herds which had a N. caninum-associated abortion epidemic. They found evidence for postnatal transmission of N. caninum based on lack of association between the seropositivity of the aborting cows and that of their dams or daughters. In contrast, a similar analysis of 2 herds with endemic abortion showed that aborting cows were more likely to have had a previous seropositive daughter than seronegative nonaborting cows, suggesting that infection had been acquired before the conception of the aborted fetus (24). In the present study, we describe 50 abortion storms in Dutch dairy herds between 1995 and 1997, for which N. caninum was considered the major cause. The herd characteristics and the N. caninum seroprevalence data of the 50 case herds are compared with those of 100 control herds with no history of an abortion storm. The main objective was to get a better understanding of the transmission of the infection. In a companion paper (6), herd level risk factors for the abortion storms are analyzed using conditional logistic regression in a matched 2 to 1 case control study design. MATERIALS
AND METHODS
Data Collection The herds that had experienced an abortion storm were selected from the data-base of the Animal Health Service (AI-IS). During a period of 3 yr (1995 to 1997), data were screened for postmortem results on aborted fetuses. Dairy farms that had had multiple N. caninum abortions were contacted and asked for further information on the number of abortions and the time span over which these had occurred. Dairy managers who indicated that
Theriogenology
235
approximately 12.5% (1 out of 8) or more of the animals at risk had aborted within a period of 2 mo were invited to participate in the study. For each case farm, 2 matched control farms were included which had no history of an abortion storm during the last 2 yr. Control farms were located within the same veterinary practice area. All participating case and control farms were visited by one of the authors (CJMB) between January and November 1997. A questionnaire regarding the state of affairs at the farm during the last 2 yr prior to the abortion storm or prior to the visit (control herds) was administered on site. Details on selection criteria of case and control herds and the questionnaire are given in a companion paper (6). Blood Sampling Eight herds in which an abortion storm had occurred in 1995 had had all blood samples collected by one of the authors (ARM), but for this study a random selection of 20% of these samples was used for analysis. In all other case herds and in the control herds blood samples were taken from 20% of the animals by the local veterinary practitioner. For case herds the time interval between the abortion storm and the herd blood sampling varied considerably. In 50% of the herds blood samples were taken within 6 mo after the abortion storm (mean 1.9 mo.), in the other 50% of the herds blood sampling was done with a delay of up to 2 yr (mean 12.9 mo.). Blood samples were collected from animals in 5 age groups: calves 0 to1 yr (Group 1), heifers 1 to 2 yr (Group 2), Parity 1 cows (Group 3), Parity 2 cows (Group 4), and Parity 3 or older cows (Group 5). Although animals for blood sampling were selected randomly, some preference was given to cows that were in the last 3 mo of gestation or the first 3 mo post partum, as during this period the highest levels of antibodies to N. caninum could be expected (8, 12, 25). In addition, blood samples were collected from all cows that had aborted during an abortion storm. Diagnostic Procedures Fetuses submitted to the AH’S laboratories were examined using a standard protocol as described previously (16, 28). Briefly, this protocol included bacteriological culture of organs and abomasal content (with special emphasis on Brucella abortus), histological examination of brain, heart, liver and placenta (if available), and examination of a pooled sample of lung and spleen for the presence of bovine viral diarrhea virus (BVDV) by virus culture or an antigen ELISA. In addition, immunohistochemistry for bovine herpesvirus 1 (BHVl) was performed on fetal liver if multifocal hepatic necrosis was found on histological examination. Immunofluorescence for Chlamvdia a. was done if grossly suspect lesions were present in the placenta. Immunohistochemical identification of N. caninum in suspect tissues was only attempted in a minority of cases, but this method was not performed routinely as it is laborious and relatively insensitive. Diagnosis of N. caninum infection was mostly based on the presence of characteristic histological lesions in the brain, heart and liver
(28). Sera were stored at -70” C before being assayed for antibodies to N. caninum by means of a tachyzoite lysate-based ELISA as described previously (25). This ELISA was shown to have a sensitivity of 0.93 to 1.00 and a specificity of 0.85 to 0.98. Results were expressed as sample to positive ratio (S/P), using a cut-off value of 0.7 (25). The assays were run within a period of 3 wk after the collection of samples had been completed.
Theriogenology
236 Definitions
Abortion was defined as termination of pregnancy by the observed expulsion of a fetus between 100 and 260 d post insemination. An abortion storm was defined as a series of abortions affecting at least 12.5% (1 out of 8) of the animals at risk within a 2-mo period. Animals at risk were defined as all heifers and cows pregnant between 100 to 260 d (10). An association of an abortion storm with N -. -caninum was presumed when N. caninum infection was diagnosed in at least 2 fetuses (28) and no other cause of abortion was found in more than 1 fetus, and when antibodies to N were present in at least 80% of the sera from -. caninum aborting cows. In 4 herds the diagnosis of N. caninum infection was only made in 1 fetus, but 100% of the aborting cows were seropositive to N. canimun. Statistical Analysis Averages of case and control herd parameters were compared with the use of an independent t-test or Chi-square test. For comparison of age groups we used an one way ANOVA. Statistical significance was defined at PI 0.05. RESULTS Geographical
Distribution
of Dairy Herds with Abortion
Storms
The 50 herds which had N. caninum associated abortion storms were located all over the country but mostly in the provinces of Friesland, Overiissel and Gelderland, which have a relatively high dairy farm density (Figure 1).
I
THE NETHERLANDS
Figure 1. Distribution of 50 farms which had occurrences of abortion storms between 1995 and 1997 in relation to regional dairy farm density (light: 0 to 0.5 dairy/kn?; light gray: 0.5 to 1 dairy/km$ dark gray: 1 to 1.5 dairies/kn?; dark: 1.5 to 2 dairies/ km’)
237 Seasonal Distribution
of Abortion
Storms
The number of abortion storms recorded during the years of 1995,1996 and 1997 was 13, 20 and 17, respectively. Figure 2 shows the seasonal distribution of the abortion storms. The onset of 38 of the 50 abortion storms occurred in the period of June to September (summer and early autumn). In 12 of the 50 herds the dairy cattle had a seasonal calving pattern, but this was not significantly associated with the time of the abortion storms (P=O.778).
12 10 8 6
JAN
MAY
MAR FEB
APR
JUL JUN
SEP AUG
NOV OCT
DEC
Month Figure 2 Seasonal distribution of the onset of Neosnora caninum associated abortion storms in 50 dairy herds in The Netherlands.
Characteristics of Herds with Abortion
Storms
Herd size, production characteristics and abortion history of the case and control herds are presented in Table 1. Five herds had experienced N. caninum abortions for 2 to 3 yr before the abortion storm.
Table 1. Main characteristics of dairy herds which had experienced Neosnora caninum associated abortion storms (n=SO) compared with control herds (n=lOO) Characteristic Case herds Control herds P-value Range Mean (SD) Range (t-test) Mean (SD) No. of dairy cows 69.9(25.9) 29-189 65.8 (23.2) 24-140 0.28 No. of heifers 29.1 (15.4) 1o-94 29.4(13.5) 6-68 0.93 Herd Standard Cow 39.3 (3.7) 33-48 39.3 (3.6) 32-50 0.99 (kg milk per day) Calving interval (days) 391.5 (12.8) 370-425 393.7(14.8) 365-438 0.39 No. of abortions during the vast yeara 3.6 (5.2) o-19 1.5 (1.3) o-5 0.027 aYear prior to the year of the abortion storm in case herds and year prior to the year of the interview in control herds.
Theriogenology
238 Numerical Characteristics
of the Abortion
Storms
The duration of the abortion storms ranged from 6 to 65 d, with a median of 41.5 d. The cumulative incidence of abortions during the storms ranged from 0.11 to 0.57 (mean 0.26) of the animals at risk. Three herds with a cumulative incidence lower than 12.5% were excluded for the risk factor analysis presented in a separate paper (6). A total of 433 abortions was recorded during the defined storm periods. Numerical data are summarized in Table 2.
Table 2. Numerical characteristics of Neosnora caninum associated abortion storms in 50 dairy herds in The Netherlands between 1995 and 1997 Characteristic of abortion storm Duration of storm (days) No. of abortions during storm Cumulative incidence Submitted fetuses with evidence of N. caninum
Mean per herd 41.5 8.6 0.26
Range
Total
6-65 5-19 0.1 l-0.57
433
l/l-l
l/15
1751226
Gestational Age of Aborted Fetuses Dates of insemination were available for 379 of the 433 (87.5%) aborting cows. The average gestational age at the time of the abortion was 180.4 d (SD 39.0 d), while the median was 179.0 d. Eight cows that had each aborted a mummified fetus were excluded from the calculation because the time interval between fetal death and abortion varied considerably. The mummified fetuses had died at approximately 3 to 4 mo gestation, based on crown-rump length estimates. Of the 433 fetuses aborted during the abortion storms, 226 fetuses (52%) were submitted to the Animal Health Service laboratories. There was no difference in gestational age between fetuses that were submitted and those that were not submitted (P = 0.22). Diagnosis for Submitted Fetuses A diagnosis of N. caninum infection was made in 175 of the 226 (77.4%) fetuses submitted during the storms. In 3 of these fetuses a bacterial infection was diagnosed: Actinomvces uvogenes, Pasteurella haemolvtica and Chlamvdia Q. In 1 fetus a fimgal infection was found. In 47 submitted fetuses, including 6 mummies, no diagnosis was reached. A total of 57 abortions was recorded in the 50 case herds during a period of 12 mo before the onset of the abortion storms. Only 4 of these 57 fetuses had been submitted, and in 1 fetus a N. caninum infection was diagnosed. During the first year after the abortion storms 175 abortions were recorded. In 45 of the 75 (60%) submitted fetuses evidence for infection with N. caninum was found.
239
Theriogenology Parity of Aborting
Cows
The average proportion of abortions increased with parity, based on 380 (88%) aborting cows for which parity was known. Heifers (age Group 2) were relatively underrepresented in the abortion storms as they comprised approximately 30% of the total population at risk. An increase of abortions with parity was also seen in the control herds, but at a much lower level than in the case herds. Seroprevalence
of N. caninum Antibodies
in Cows of Case and Control Herds
The seroprevalence of antibodies to N. caninum varied from 17 to 87% (mean 5 1.5%) for case herds, and from 0 to 53% (mean 13.9%) for control herds. All figures were based on a sample of 20 % of each herd, excluding the aborting cows. The distribution of the case and control herds by seroprevalence is presented in Figure 3. In the case herds 2 15 of the 240 (89.6%) cows aborting during the abortion storms were seropositive for N. caninum. whereas 51 of the 64 (79.7%) cows aborting within 12 mo after the abortion storm were seropositive. In the control herds 18 of the 74 (24.3%) cows with a history of abortion were seropositive for N. caninum.
0 Control herd I
Case herd
Seroprevalence (%)
Figure 3. Distribution of case herds (n = 50) and control herds (n = 100) by seroprevalence of Neosnora caninum antibodies based on a 20% cross sectional sample (excluding aborting cows).
Theriogenology
240
The overall seroprevalences per age group for case herds and control herds are depicted in Figure 4. There were no significant differences in seroprevalence among the age groups in either the case herds (P = 0.0667) or control herds (P = 0.749). Only after analyzing seroprevalence data in the individual herds were significant differences per age group found in some herds. In 6 case herds, the young animals (age Groups 1, 2) had a significantly (P < 0.05) lower seroprevalence to N. caninum than the adult cows (age Groups 3,4, and 5). An even more extreme situation was observed in 1 control herd where none of the 14 young animals proved seropositive, whereas the proportion of seropositive adult cows was 16 out of 20 (80%). In another control herd 9 of the 11 animals in age Group 2 were seropositive versus 2 of 27 animals in all the other combined age groups. After comparing 25 case herds from which blood samples were collected within 6 mo after the abortion storm (mean 1.9 mo) with 25 herds that were blood sampled later than 6 months after the storm (mean 12.9 mo) no significant differences in seroprevalence were found between the early and the late sampled herds for each of the 5 age groups (P = 0.49).
Figure 4. Seroprevalence to Neosnora caninum in different age groups in 50 case herds and 100 control herds (Group 1 = calves 0 to 1 yr; Group 2 = heifers 1 to 2 yr; Group 3 = Parity 1 cows; Group 4 = Parity 2 cows; Group 5 = Parity 3 and older cows).
N. caninum Antibody
Levels in Aborting and Nonaborting
Cows in Case Herds
Aborting cows had significantly higher antibody levels (S/P ratios) than nonaborting cows in all age groups (Figure 5). Antibody values in calves ranged to higher levels than in nonaborting animals in the older age groups.
73eriogenology
241
:g 4I 2
2
z
oa
0
8 *
Nomborting
-2. N=
rAborting
305.
m-31 1
m-72
2
174
3
a7 4
267
182 5
Figure 5. N. caninum antibody levels (sample to positive ratio) in aborting and nonaborting cows in different age groups of 50 dairy herds which had experienced abortion storms (Group 1 = calves; Group 2 = heifers 1 to 2 yr; Group 3 = Parity 1 cows; Group 4 = Parity 2 cows; Group 5 = Parity 3 and older cows).
DISCUSSION Abortion storms in 50 Dutch dairy herds are reported in which there was a strong association with N. caninum-infection. In all herds taken together, a diagnosis of infection with N. caninum was made in 175 of the 226 (77.4%) submitted fetuses. In 46 herds N. caninum was the only identified cause of abortion, while in 4 herds N. caninum-infection was the main diagnosis, with a bacterial or fungal infection being demonstrated in an additional fetus. Moreover, 215 of the 240 (89.6%) cows aborting during the outbreaks were seropositive for & caninum, and the antibody levels in aborting cows were significantly higher than in nonaborting cows. There was an apparent seasonal influence on the occurrence of N. caninum-associated abortion storms in the 50 dairy herds of this study. Most abortion storms occurred during the summer and early autumn. Possible seasonal risk factors are discussed in a companion paper (6). The average gestational age of the aborted fetuses was 180 d (n = 371). This finding is similar to that of earlier reports on N. canimun-associated abortion storms (16,21,22). The average gestational age of 95 N. caninum-infected fetuses from California dairies was 5.4 mo (164 d) with a range of 3.5 to 8 mo (1). In a cohort study of congenitally infected cows Thurmond and Bietala (23) concluded from fetal survival curves that the period at risk of s. caninum abortion was predominantly between 90 and 180 d of gestation. They found that fetal age at abortion increased and that the length of the period of high risk decreased with
242
Theriogenology
each subsequent pregnancy. The relatively old age of the fetuses in our study compared with those in the Californian study may be related to the underrepresentation of heifers in the abortion storms. A possible explanation for this underrepresentation of heifers is the fact that these nonlactating animals are less frequently under the observation of the farmer. The feeding regimen may also be important. Heifers usually are pastured without additional feeding during the summer months. It has been reported that the prevalence of -.N -caninum abortions is much lower in pastured beef cattle than in drylot dairy cattle (4). In general,the incidence of abortion in the cows in the present study increased with age. This could be related to the level of production and the associated metabolic activity, which increase with age. After analyzing the N_.caninum seroprevalence data, the most noteworthy results were 1) seroprevalence was significantly higher in case herds than in control herds, 2) except for 6 case herds and 2 control herds, there were no significant differences in seroprevalence among the different age groups. In other words, generally speaking, seropositive animals were equally distributed in all age groups, but at a much higher level in the case herds than in the control herds. The high seroprevalence of N. caninum in case herds compared with control herds indicates a strong relationship between the prevalence of N. caninum-infection in the herds and the occurrence of an abortion storm. However, it is impossible to draw conclusions as to the cause and effect of this relationship because the seroprevalence was not known before the abortion storms. Theoretically there are 2 major possibilities: 1) A high prevalence was the effect of a recent introduction of the infection into the herd, which then caused the abortion storm. 2) A high prevalence already existed before the onset of the abortion storm and then formed the basis for recrudescence of the infection in a number of cows by factors causing immune suppression. In most herds similar proportions of all age groups were infected. It is hard to imagine that a recent introduction caused simultaneous infection of all age groups. Particularly the fact that young animals and adult cows are usually kept separately and are fed different rations argues against such simultaneous infection. A more plausible explanation for the similarity of seroprevalence in all age groups is vertical transmission of the infection, which has been established as the major route of infection in dairy herds (2, 18, 19,24,27). Assuming that an equal distribution of seropositives over all ages was already present at the time of the abortion storms, the infection must have been perpetuated by vertical transmission for some time before the onset of the abortion storms. This would favor the hypothesis that the abortion storms were due to a reactivation of neosporosis in a number of chronically infected cows. Further evidence for this hypothesis might lay in higher abortion incidence in the case herds before the abortion storms as compared with the control herds. This might indicate that the infection was already endemic in @art of) the case herds before the onset of the abortion storms. Indeed, some case herds were known to have had N. caninum abortions for several years. On the other hand, the high levels of prevalence of N. caninum found strongly suggest that postnatal transmission had occurred at some time point. Through vertical transmission alone N. caninum-infection could not reach and maintain such high prevalence levels, as has been demonstrated by modeling (11). The significant overrepresentation of seropositives in
Theriogenology
243
adult cows in 7 herds (6 case and 1 control herd) and in l- to 2-yr-old heifers in 1 control herd also favors a preferential exposure of these animals to the infection, which can only be explained by assuming postnatal infection. The 6 case herds with a disproportionately high seroprevalence in adult cows all had had an explosive abortion storm within a period of 4 wk (range 9 to 28 d, mean 15.3 d), suggesting a point source infection. The possibility that part of the herd (only the aborting cows) may have had a primary infection leading to an abortion storm, as suggested in a previous report (16) cannot be excluded. Pre-abortion storm serology is needed to clarify this. The N. caninum infection appears to be widespread in Dutch dairy cattle population as seropositive animals were found in 78 of 100 control herds. In addition, due to the sampling policy of 20% of animals per herd, some herds may have been misclassified into the zero prevalence group, although the chance for such misclassification with an average seroprevalence of 14% is low. Similar findings have been reported from Canada (17) and Denmark (13). In a Canadian study, 16 out of 22 (73%) herds with no known history of N. caninum abortions had at least 1 seropositive cow (17) whereas Danish researchers (13) found seroprevalence values ranging from 6 to 59% in 8 out of 16 herds with no history of abortions. Although case herds in our study had, on average, a much higher seroprevalence than control herds (51.5 vs 13.9%), there was a clear overlap in seroprevalence among cases and controls. The control herds with the 30% highest seroprevalence were within the range of the case herds. In our estimation this would make these herds at risk of an abortion storm. In 1 control herd with an estimated seroprevalence of 20%, a minor outbreak of abortions was recorded 4 mo after the cross-sectional herd blood sampling. Four cows aborted within 1 wk and all 4 had high levels of antibodies to N. caninum in post abortion blood samples. Two fetuses were submitted and both had evidence of N. -~ caninum infection. Unfortunately, the preabortion serostatus to N. caninum of the 4 aborting cows in this herd had not been determined. It appears from this study that the knowledge of the seroprevalence in herds before the onset of an abortion storm may contribute significantly to the understanding of the epidemiology of neosporosis. Therefore, serobanking should be encouraged. REFERENCES 1. Anderson ML, Blanchard PC, Barr BC, Dubey JP, Hoffman RL and Conrad PA. Neosnora-like protozoan infections as a major cause of abortion in California dairy cattle. J Am Vet Med Assoc 1991; 198: 241-244. 2. Anderson ML, Reynolds JP, Rowe JD, Sverlow KW, Packham AE, Barr BC, Conrad PA. Evidence of vertical transmission of Neosnora sp infection in dairy cattle. J Am Vet Med Assoc 1997;210:1169-1172. 3. Barr BC, Anderson ML, Blanchard PC, Daft BM, Kinde H, Conrad PA. Bovine fetal encephalitis and myocarditis associated with protozoa1 infections. Vet Path01 1990;27:354-361. 4. Barr BC, Bjerkas I, Buxton D, Conrad PA, Dubey JP, Ellis JT, Jenkins MC, Johnston SA, Lindsay DS, Sibley D, Trees AJ, Wouda W. Neosporosis - Report of the International Neospora Workshop. Compend Contin Educ Pratt Vet 1997; 19:5 1205126,5144.
244
Theriogenology
5. Barr BC, Conrad PA, Breitmeyer R, Sverlow IS, Anderson ML, Reynolds J, %hauvet AE, Dubey JP, Ardans AA. Congenital Neosnora infection in calves born from cows that had previously aborted Neospora-infected fetuses: four cases (1990-1992). J Am VetMed Assoc 1993;202:113-117. 6. Bartels CJM, Wouda W, Schukken YH. Risk factors for Neosnora caninum-associated abortion storms in dairy herds in The Netherlands (1995 to 1997). Theriogenology 1999;52:247-257. 7. Bjiirkman C, Johansson 0, Stenhmd S, Holmdahl JM, Uggla A. Neosnora species infection in a herd of dairy cattle. J Am Vet Med Assoc 1996;208:1441-1444. 8. Conrad PA, Sverlow K, Anderson M, Rowe J, BonDurant R, Tuter G, Breitmeyer R, Palmer C, Thurmond M, Ardans A, Dubey JP, Duhamel G, Barr B. Detection of serum antibody responses in cattle with natural or experimental Neosnora infections. J Vet Diagn Invest 1993;5:572-578. 9. Dubey JP, Lindsay DS. A review of Neosnora caninum and neosporosis. Vet Parasitol 1996;67:1-59. 10. Fetrow J, McClary D, Harman R, Butcher K, Weaver L, Studer E, Ehrlich J, Etherington W, Guterbock W, Klinborg D, Reneau J, Williamson N. Calculating selected reproductive indices: recommendations of the American Association of Bovine Practitioners. J Dairy Sci 1990;73:778-90. 11. French NP, Davison HC, Clancy D, Begon M, Trees AJ. Modelling of Neosnora species infection in dairy cattle: the importance of horizontal and vertical transmission and differential culling. Thrustield MV, Goodall EA (eds), Proc Sot Vet Epidemiol Prev Med, 1998;113-122. 12. Jenkins MC, Wouda W, Dubey JP. Serological response over time to recombinant Neosnora caninum antigens in cattle after a neosporosis-induced abortion. Clin Diagn Lab Immunol 1997;4:270-274. 13. Lind P, Kjeldsen AM, Bj?&man C, Uggla A, Willadsen C, Wedderkopp A, Jensen AM. Sero-epidemiological studies of Neospora infection in Danish cattle herds. VIIth International Coccidiosis Conference and European Union COST 820 Workshop, 1997:78 abstr. 14. McAllister MM, Huftinan EM, Hietala SK, Conrad PA, Anderson ML, Salman MD. Evidence suggesting a point source exposure in an outbreak of bovine abortion due to neosporosis. J Vet Diagn Invest 1996; 8:355-357. 15. McAllister MM, Dubey JP, Lindsay DS, Jolley WR, Wills RA, McGuire AM. Dogs are definitive hosts ofNeosoora caninum Int J Parasitol 1998;28:1473-1478. 16. Moen AR, Wouda W, Mul MGt EAM, Werven T van. Increased risk of abortion following Neosnora caninum abortion outbreaks: a retrospective and prospective cohort study in four dairy herds. Theriogenology 1998;49:1301-1309. 17. Pare J, Fecteau G, Fortin M, Marsolais G. Seroepidemiologic study of Neoswra caninum in dairy herds. J Am Vet Med Assoc 1998;213:1595-1598. 18. Pare J, Thurmond MC, Hietala SK. Congenital Neosnora canimtm infection in dairy cattle and associated calfhood mortality. Can J Vet Res 1996;60: 133-l 39. 19. Pare J, Thurmond MC, Hietala SK. Neosnora caninum antibodies in cows during pregnancy as a predictor of congenital infection and abortion. J Parasitol 1997;83:8287. 20. Schares G, Peters M, Wurm R, Tackmann K, Henning K, Conraths FJ. Neoswra caninum causes abortions in a cattle herd in North Rhine Westphalia. Dtsch Tieranzl Wochenschr 1997;104:208-212.
Theriogenology 21.
245
Thilsted JP, Dubey JP. Neosporosis-like abortions in a herd of dairy cattle. J Vet Diagn Invest 1989;1:205-209. 22. Thornton RN, Gajadhar A, Evans J. Neosnora abortion epidemic in a dairy herd. N Z Vet J 1994;42:190-191. 23. Thurmond M, Hietala S. Effect of congenitally acquired Neospora caninum infection on risk of abortion and subsequent abortions in dairy cattle. Am J Vet Res 1997;58: 13811385. 24. Thurmond M, Hietala S, Blanchard PC. Herd-based diagnosis of Neosnora caninuminduced endemic and epidemic abortion in cows and evidence for congenital and postnatal transmission. J Vet Diagn Invest 1997;9:44-49. 25. Wouda W, Brinkhof J, Maanen C van, Gee ALW de, Moen AR. Serodiagnosis of neosporosis in individual cows and dairy herds, a comparison of three enzyme linked immunosorbent assays. Clin Diagn Lab Immunol; 1998;5:711-716. 26. Wouda W, Ingh TSGAM van den, Knapen F van, Sluyter FJH, Koeman JP, Dubey JP. Neosnora abortus bij het rund in Nederland. Tijdschr Diergeneeskd 1992; 117599-602. 27. Wouda W, Moen AR, Schukken YH. Abortion risk in progeny of cows that experienced a Neosnora caninum abortion epidemic. Theriogenology 1998;49: 13 ll1316. 28. Wouda W, Moen AR, Visser IJR, Knapen F van. Bovine fetal neosporosis, a comparison of epizootic abortion cases and different age classes with regard to lesion severity and immunohistochemical identification of organisms in brain, heart and liver. J Vet Diagn Invest 1997;9:180-185. 29. Yaeger MJ, Shawd-Wessels SC, Leslie-Steen P. Neosnora abortion storm in a midwestem dairy. J Vet Diagn Invest 1994;6:506-508.
CHARACTERISTICS OF Neosnora caninum-ASSOCIATED ABORTION IN DAIRY HERDS IN THE NETHERLANDS (1995 to1997) W. W~uda,‘~C. J. M. Bartels,‘and
STORMS
A. R. Moen’
‘Animal Health Service Drachten, The Netherlands *Department of Herd Health and Reproduction Utrecht University, The Netherlands Received for publication: Accepted:
23 July 9 April
1998 1999
ABSTRACT Abortion storms in 50 dairy herds in The Netherlands were reported in which there was a strong association with Neosnora caninum-infection. The duration of the abortion storms ranged from 6 to 65 d (mean 41.5 d). The cumulative proportion of aborting cows ranged from 0.11 to 0.57 (mean 0.26) of the animals at risk. An apparent seasonal infhtence was noted as most abortion storms occurred during the summer and early autumn. The prevalence of antibodies to N. caninum in 50 herds which had had an abortion storm was compared with that of 100 control herds which had no history of an abortion storm. Seroprevalence was estimated by testing a 20% cross sectional herd sample using a tachyzoite lysate-based ELISA method. Seroprevalence in case herds (range 17 to 87%, mean 5 1.5%) was significantly higher than that in control herds (range 0 to 53%, mean 13.9%). For most herds the seroprevalence levels were equal across all age groups, which suggests that the infection had been perpetuated by vertical transmission. In these herds, the abortion storms appeared to be induced by factors causing recrudescence of a N. caninuminfection in chronically infected animals rather than being the result of a recent introduction. In 6 case herds the seroprevalence in the dairy cows was significantly higher than in the young stock, which may have been attributable to superimposed postnatal infection. 0 1999 by Elawier Science Inc.
Key words: Neosnora caninum, bovine, abortion, seroprevalence, transmission, epidemiology
Acknowledgments The authors thank all the dairy farmers and their veterinary practitioners for participation; G. H. A. Borst and J. H. Vos for referral of cases; J. Brinkhof, A. Damsma, G. Koning, A. van der Meulen, M. Swinkels and H. Westra for technical laboratory assistance; and P. G. M. Versluys for preparing Figure 1. We are indebted to Y. H. Schukken for advise and help with the study design and to R. N. Zadoks for valuable comments on the manuscript. ‘Correspondence and reprint requests: Animal Health Service, P. 0. Box, 9200 AJ Drachten. The Netherlands. Phone: 3 1 5 12 570 700 Fax: 3 1 512 520 013 E-mail: Theriogenology 52:233-245. 0 1999 by Elsevier Science
1999 Inc.
0093~691X&9/$-see front matter PII SOO93-691X(99)001259
234 INTRODUCTION During the last decade Neosnora caninum has been recognized as an important cause of abortion in cattle worldwide (for review see: 9). The infection has been associated with endemic and epidemic abortions in dairy herds (1, 4, 14, 15, 21, 22, 24, 28, 29). The importance of neosporosis for the Dutch dairy industry is reflected in the proportion of submitted fetuses in which the infection is diagnosed. After the first cases of N. caninum abortion in cattle had been recognized in the early nineties (26), yearly 15 to 20% of aborted calves submitted to the Animal Health Service laboratories were found to have histological lesions compatible with N_. caninum infection (28). This makes N. caninum the most recognized cause of abortion in cattle in The Netherlands. Vertical transmission from cow to calf has been well established (2, 5, 18, 27) and may contribute significantly to the persistence of the infection in the herd (7, 18). Recently it has been demonstrated that dogs can act as definitive hosts of N. caninum (15). Three of 4 dogs that were fed mouse tissues containing N. caninum tissue cysts shed fecal oocysts. Mice that received feces from these dogs were shown to be infected by N. caninum. The implications of these experimental findings for the epidemiology bovine neosporosis have yet to be investigated. Case reports of N. caninum associated abortion storms have contributed to the knowledge of the epidemiology of the infection by providing circumstantial evidence for a point source infection (14, 16,22,29). Recently, Thurmond et al. (24) reported a study of 20 herds which had a N. caninum-associated abortion epidemic. They found evidence for postnatal transmission of N. caninum based on lack of association between the seropositivity of the aborting cows and that of their dams or daughters. In contrast, a similar analysis of 2 herds with endemic abortion showed that aborting cows were more likely to have had a previous seropositive daughter than seronegative nonaborting cows, suggesting that infection had been acquired before the conception of the aborted fetus (24). In the present study, we describe 50 abortion storms in Dutch dairy herds between 1995 and 1997, for which N. caninum was considered the major cause. The herd characteristics and the N. caninum seroprevalence data of the 50 case herds are compared with those of 100 control herds with no history of an abortion storm. The main objective was to get a better understanding of the transmission of the infection. In a companion paper (6), herd level risk factors for the abortion storms are analyzed using conditional logistic regression in a matched 2 to 1 case control study design. MATERIALS
AND METHODS
Data Collection The herds that had experienced an abortion storm were selected from the data-base of the Animal Health Service (AI-IS). During a period of 3 yr (1995 to 1997), data were screened for postmortem results on aborted fetuses. Dairy farms that had had multiple N. caninum abortions were contacted and asked for further information on the number of abortions and the time span over which these had occurred. Dairy managers who indicated that
Theriogenology
235
approximately 12.5% (1 out of 8) or more of the animals at risk had aborted within a period of 2 mo were invited to participate in the study. For each case farm, 2 matched control farms were included which had no history of an abortion storm during the last 2 yr. Control farms were located within the same veterinary practice area. All participating case and control farms were visited by one of the authors (CJMB) between January and November 1997. A questionnaire regarding the state of affairs at the farm during the last 2 yr prior to the abortion storm or prior to the visit (control herds) was administered on site. Details on selection criteria of case and control herds and the questionnaire are given in a companion paper (6). Blood Sampling Eight herds in which an abortion storm had occurred in 1995 had had all blood samples collected by one of the authors (ARM), but for this study a random selection of 20% of these samples was used for analysis. In all other case herds and in the control herds blood samples were taken from 20% of the animals by the local veterinary practitioner. For case herds the time interval between the abortion storm and the herd blood sampling varied considerably. In 50% of the herds blood samples were taken within 6 mo after the abortion storm (mean 1.9 mo.), in the other 50% of the herds blood sampling was done with a delay of up to 2 yr (mean 12.9 mo.). Blood samples were collected from animals in 5 age groups: calves 0 to1 yr (Group 1), heifers 1 to 2 yr (Group 2), Parity 1 cows (Group 3), Parity 2 cows (Group 4), and Parity 3 or older cows (Group 5). Although animals for blood sampling were selected randomly, some preference was given to cows that were in the last 3 mo of gestation or the first 3 mo post partum, as during this period the highest levels of antibodies to N. caninum could be expected (8, 12, 25). In addition, blood samples were collected from all cows that had aborted during an abortion storm. Diagnostic Procedures Fetuses submitted to the AH’S laboratories were examined using a standard protocol as described previously (16, 28). Briefly, this protocol included bacteriological culture of organs and abomasal content (with special emphasis on Brucella abortus), histological examination of brain, heart, liver and placenta (if available), and examination of a pooled sample of lung and spleen for the presence of bovine viral diarrhea virus (BVDV) by virus culture or an antigen ELISA. In addition, immunohistochemistry for bovine herpesvirus 1 (BHVl) was performed on fetal liver if multifocal hepatic necrosis was found on histological examination. Immunofluorescence for Chlamvdia a. was done if grossly suspect lesions were present in the placenta. Immunohistochemical identification of N. caninum in suspect tissues was only attempted in a minority of cases, but this method was not performed routinely as it is laborious and relatively insensitive. Diagnosis of N. caninum infection was mostly based on the presence of characteristic histological lesions in the brain, heart and liver
(28). Sera were stored at -70” C before being assayed for antibodies to N. caninum by means of a tachyzoite lysate-based ELISA as described previously (25). This ELISA was shown to have a sensitivity of 0.93 to 1.00 and a specificity of 0.85 to 0.98. Results were expressed as sample to positive ratio (S/P), using a cut-off value of 0.7 (25). The assays were run within a period of 3 wk after the collection of samples had been completed.
Theriogenology
236 Definitions
Abortion was defined as termination of pregnancy by the observed expulsion of a fetus between 100 and 260 d post insemination. An abortion storm was defined as a series of abortions affecting at least 12.5% (1 out of 8) of the animals at risk within a 2-mo period. Animals at risk were defined as all heifers and cows pregnant between 100 to 260 d (10). An association of an abortion storm with N -. -caninum was presumed when N. caninum infection was diagnosed in at least 2 fetuses (28) and no other cause of abortion was found in more than 1 fetus, and when antibodies to N were present in at least 80% of the sera from -. caninum aborting cows. In 4 herds the diagnosis of N. caninum infection was only made in 1 fetus, but 100% of the aborting cows were seropositive to N. canimun. Statistical Analysis Averages of case and control herd parameters were compared with the use of an independent t-test or Chi-square test. For comparison of age groups we used an one way ANOVA. Statistical significance was defined at PI 0.05. RESULTS Geographical
Distribution
of Dairy Herds with Abortion
Storms
The 50 herds which had N. caninum associated abortion storms were located all over the country but mostly in the provinces of Friesland, Overiissel and Gelderland, which have a relatively high dairy farm density (Figure 1).
I
THE NETHERLANDS
Figure 1. Distribution of 50 farms which had occurrences of abortion storms between 1995 and 1997 in relation to regional dairy farm density (light: 0 to 0.5 dairy/kn?; light gray: 0.5 to 1 dairy/km$ dark gray: 1 to 1.5 dairies/kn?; dark: 1.5 to 2 dairies/ km’)
237 Seasonal Distribution
of Abortion
Storms
The number of abortion storms recorded during the years of 1995,1996 and 1997 was 13, 20 and 17, respectively. Figure 2 shows the seasonal distribution of the abortion storms. The onset of 38 of the 50 abortion storms occurred in the period of June to September (summer and early autumn). In 12 of the 50 herds the dairy cattle had a seasonal calving pattern, but this was not significantly associated with the time of the abortion storms (P=O.778).
12 10 8 6
JAN
MAY
MAR FEB
APR
JUL JUN
SEP AUG
NOV OCT
DEC
Month Figure 2 Seasonal distribution of the onset of Neosnora caninum associated abortion storms in 50 dairy herds in The Netherlands.
Characteristics of Herds with Abortion
Storms
Herd size, production characteristics and abortion history of the case and control herds are presented in Table 1. Five herds had experienced N. caninum abortions for 2 to 3 yr before the abortion storm.
Table 1. Main characteristics of dairy herds which had experienced Neosnora caninum associated abortion storms (n=SO) compared with control herds (n=lOO) Characteristic Case herds Control herds P-value Range Mean (SD) Range (t-test) Mean (SD) No. of dairy cows 69.9(25.9) 29-189 65.8 (23.2) 24-140 0.28 No. of heifers 29.1 (15.4) 1o-94 29.4(13.5) 6-68 0.93 Herd Standard Cow 39.3 (3.7) 33-48 39.3 (3.6) 32-50 0.99 (kg milk per day) Calving interval (days) 391.5 (12.8) 370-425 393.7(14.8) 365-438 0.39 No. of abortions during the vast yeara 3.6 (5.2) o-19 1.5 (1.3) o-5 0.027 aYear prior to the year of the abortion storm in case herds and year prior to the year of the interview in control herds.
Theriogenology
238 Numerical Characteristics
of the Abortion
Storms
The duration of the abortion storms ranged from 6 to 65 d, with a median of 41.5 d. The cumulative incidence of abortions during the storms ranged from 0.11 to 0.57 (mean 0.26) of the animals at risk. Three herds with a cumulative incidence lower than 12.5% were excluded for the risk factor analysis presented in a separate paper (6). A total of 433 abortions was recorded during the defined storm periods. Numerical data are summarized in Table 2.
Table 2. Numerical characteristics of Neosnora caninum associated abortion storms in 50 dairy herds in The Netherlands between 1995 and 1997 Characteristic of abortion storm Duration of storm (days) No. of abortions during storm Cumulative incidence Submitted fetuses with evidence of N. caninum
Mean per herd 41.5 8.6 0.26
Range
Total
6-65 5-19 0.1 l-0.57
433
l/l-l
l/15
1751226
Gestational Age of Aborted Fetuses Dates of insemination were available for 379 of the 433 (87.5%) aborting cows. The average gestational age at the time of the abortion was 180.4 d (SD 39.0 d), while the median was 179.0 d. Eight cows that had each aborted a mummified fetus were excluded from the calculation because the time interval between fetal death and abortion varied considerably. The mummified fetuses had died at approximately 3 to 4 mo gestation, based on crown-rump length estimates. Of the 433 fetuses aborted during the abortion storms, 226 fetuses (52%) were submitted to the Animal Health Service laboratories. There was no difference in gestational age between fetuses that were submitted and those that were not submitted (P = 0.22). Diagnosis for Submitted Fetuses A diagnosis of N. caninum infection was made in 175 of the 226 (77.4%) fetuses submitted during the storms. In 3 of these fetuses a bacterial infection was diagnosed: Actinomvces uvogenes, Pasteurella haemolvtica and Chlamvdia Q. In 1 fetus a fimgal infection was found. In 47 submitted fetuses, including 6 mummies, no diagnosis was reached. A total of 57 abortions was recorded in the 50 case herds during a period of 12 mo before the onset of the abortion storms. Only 4 of these 57 fetuses had been submitted, and in 1 fetus a N. caninum infection was diagnosed. During the first year after the abortion storms 175 abortions were recorded. In 45 of the 75 (60%) submitted fetuses evidence for infection with N. caninum was found.
239
Theriogenology Parity of Aborting
Cows
The average proportion of abortions increased with parity, based on 380 (88%) aborting cows for which parity was known. Heifers (age Group 2) were relatively underrepresented in the abortion storms as they comprised approximately 30% of the total population at risk. An increase of abortions with parity was also seen in the control herds, but at a much lower level than in the case herds. Seroprevalence
of N. caninum Antibodies
in Cows of Case and Control Herds
The seroprevalence of antibodies to N. caninum varied from 17 to 87% (mean 5 1.5%) for case herds, and from 0 to 53% (mean 13.9%) for control herds. All figures were based on a sample of 20 % of each herd, excluding the aborting cows. The distribution of the case and control herds by seroprevalence is presented in Figure 3. In the case herds 2 15 of the 240 (89.6%) cows aborting during the abortion storms were seropositive for N. caninum. whereas 51 of the 64 (79.7%) cows aborting within 12 mo after the abortion storm were seropositive. In the control herds 18 of the 74 (24.3%) cows with a history of abortion were seropositive for N. caninum.
0 Control herd I
Case herd
Seroprevalence (%)
Figure 3. Distribution of case herds (n = 50) and control herds (n = 100) by seroprevalence of Neosnora caninum antibodies based on a 20% cross sectional sample (excluding aborting cows).
Theriogenology
240
The overall seroprevalences per age group for case herds and control herds are depicted in Figure 4. There were no significant differences in seroprevalence among the age groups in either the case herds (P = 0.0667) or control herds (P = 0.749). Only after analyzing seroprevalence data in the individual herds were significant differences per age group found in some herds. In 6 case herds, the young animals (age Groups 1, 2) had a significantly (P < 0.05) lower seroprevalence to N. caninum than the adult cows (age Groups 3,4, and 5). An even more extreme situation was observed in 1 control herd where none of the 14 young animals proved seropositive, whereas the proportion of seropositive adult cows was 16 out of 20 (80%). In another control herd 9 of the 11 animals in age Group 2 were seropositive versus 2 of 27 animals in all the other combined age groups. After comparing 25 case herds from which blood samples were collected within 6 mo after the abortion storm (mean 1.9 mo) with 25 herds that were blood sampled later than 6 months after the storm (mean 12.9 mo) no significant differences in seroprevalence were found between the early and the late sampled herds for each of the 5 age groups (P = 0.49).
Figure 4. Seroprevalence to Neosnora caninum in different age groups in 50 case herds and 100 control herds (Group 1 = calves 0 to 1 yr; Group 2 = heifers 1 to 2 yr; Group 3 = Parity 1 cows; Group 4 = Parity 2 cows; Group 5 = Parity 3 and older cows).
N. caninum Antibody
Levels in Aborting and Nonaborting
Cows in Case Herds
Aborting cows had significantly higher antibody levels (S/P ratios) than nonaborting cows in all age groups (Figure 5). Antibody values in calves ranged to higher levels than in nonaborting animals in the older age groups.
73eriogenology
241
:g 4I 2
2
z
oa
0
8 *
Nomborting
-2. N=
rAborting
305.
m-31 1
m-72
2
174
3
a7 4
267
182 5
Figure 5. N. caninum antibody levels (sample to positive ratio) in aborting and nonaborting cows in different age groups of 50 dairy herds which had experienced abortion storms (Group 1 = calves; Group 2 = heifers 1 to 2 yr; Group 3 = Parity 1 cows; Group 4 = Parity 2 cows; Group 5 = Parity 3 and older cows).
DISCUSSION Abortion storms in 50 Dutch dairy herds are reported in which there was a strong association with N. caninum-infection. In all herds taken together, a diagnosis of infection with N. caninum was made in 175 of the 226 (77.4%) submitted fetuses. In 46 herds N. caninum was the only identified cause of abortion, while in 4 herds N. caninum-infection was the main diagnosis, with a bacterial or fungal infection being demonstrated in an additional fetus. Moreover, 215 of the 240 (89.6%) cows aborting during the outbreaks were seropositive for & caninum, and the antibody levels in aborting cows were significantly higher than in nonaborting cows. There was an apparent seasonal influence on the occurrence of N. caninum-associated abortion storms in the 50 dairy herds of this study. Most abortion storms occurred during the summer and early autumn. Possible seasonal risk factors are discussed in a companion paper (6). The average gestational age of the aborted fetuses was 180 d (n = 371). This finding is similar to that of earlier reports on N. canimun-associated abortion storms (16,21,22). The average gestational age of 95 N. caninum-infected fetuses from California dairies was 5.4 mo (164 d) with a range of 3.5 to 8 mo (1). In a cohort study of congenitally infected cows Thurmond and Bietala (23) concluded from fetal survival curves that the period at risk of s. caninum abortion was predominantly between 90 and 180 d of gestation. They found that fetal age at abortion increased and that the length of the period of high risk decreased with
242
Theriogenology
each subsequent pregnancy. The relatively old age of the fetuses in our study compared with those in the Californian study may be related to the underrepresentation of heifers in the abortion storms. A possible explanation for this underrepresentation of heifers is the fact that these nonlactating animals are less frequently under the observation of the farmer. The feeding regimen may also be important. Heifers usually are pastured without additional feeding during the summer months. It has been reported that the prevalence of -.N -caninum abortions is much lower in pastured beef cattle than in drylot dairy cattle (4). In general,the incidence of abortion in the cows in the present study increased with age. This could be related to the level of production and the associated metabolic activity, which increase with age. After analyzing the N_.caninum seroprevalence data, the most noteworthy results were 1) seroprevalence was significantly higher in case herds than in control herds, 2) except for 6 case herds and 2 control herds, there were no significant differences in seroprevalence among the different age groups. In other words, generally speaking, seropositive animals were equally distributed in all age groups, but at a much higher level in the case herds than in the control herds. The high seroprevalence of N. caninum in case herds compared with control herds indicates a strong relationship between the prevalence of N. caninum-infection in the herds and the occurrence of an abortion storm. However, it is impossible to draw conclusions as to the cause and effect of this relationship because the seroprevalence was not known before the abortion storms. Theoretically there are 2 major possibilities: 1) A high prevalence was the effect of a recent introduction of the infection into the herd, which then caused the abortion storm. 2) A high prevalence already existed before the onset of the abortion storm and then formed the basis for recrudescence of the infection in a number of cows by factors causing immune suppression. In most herds similar proportions of all age groups were infected. It is hard to imagine that a recent introduction caused simultaneous infection of all age groups. Particularly the fact that young animals and adult cows are usually kept separately and are fed different rations argues against such simultaneous infection. A more plausible explanation for the similarity of seroprevalence in all age groups is vertical transmission of the infection, which has been established as the major route of infection in dairy herds (2, 18, 19,24,27). Assuming that an equal distribution of seropositives over all ages was already present at the time of the abortion storms, the infection must have been perpetuated by vertical transmission for some time before the onset of the abortion storms. This would favor the hypothesis that the abortion storms were due to a reactivation of neosporosis in a number of chronically infected cows. Further evidence for this hypothesis might lay in higher abortion incidence in the case herds before the abortion storms as compared with the control herds. This might indicate that the infection was already endemic in @art of) the case herds before the onset of the abortion storms. Indeed, some case herds were known to have had N. caninum abortions for several years. On the other hand, the high levels of prevalence of N. caninum found strongly suggest that postnatal transmission had occurred at some time point. Through vertical transmission alone N. caninum-infection could not reach and maintain such high prevalence levels, as has been demonstrated by modeling (11). The significant overrepresentation of seropositives in
Theriogenology
243
adult cows in 7 herds (6 case and 1 control herd) and in l- to 2-yr-old heifers in 1 control herd also favors a preferential exposure of these animals to the infection, which can only be explained by assuming postnatal infection. The 6 case herds with a disproportionately high seroprevalence in adult cows all had had an explosive abortion storm within a period of 4 wk (range 9 to 28 d, mean 15.3 d), suggesting a point source infection. The possibility that part of the herd (only the aborting cows) may have had a primary infection leading to an abortion storm, as suggested in a previous report (16) cannot be excluded. Pre-abortion storm serology is needed to clarify this. The N. caninum infection appears to be widespread in Dutch dairy cattle population as seropositive animals were found in 78 of 100 control herds. In addition, due to the sampling policy of 20% of animals per herd, some herds may have been misclassified into the zero prevalence group, although the chance for such misclassification with an average seroprevalence of 14% is low. Similar findings have been reported from Canada (17) and Denmark (13). In a Canadian study, 16 out of 22 (73%) herds with no known history of N. caninum abortions had at least 1 seropositive cow (17) whereas Danish researchers (13) found seroprevalence values ranging from 6 to 59% in 8 out of 16 herds with no history of abortions. Although case herds in our study had, on average, a much higher seroprevalence than control herds (51.5 vs 13.9%), there was a clear overlap in seroprevalence among cases and controls. The control herds with the 30% highest seroprevalence were within the range of the case herds. In our estimation this would make these herds at risk of an abortion storm. In 1 control herd with an estimated seroprevalence of 20%, a minor outbreak of abortions was recorded 4 mo after the cross-sectional herd blood sampling. Four cows aborted within 1 wk and all 4 had high levels of antibodies to N. caninum in post abortion blood samples. Two fetuses were submitted and both had evidence of N. -~ caninum infection. Unfortunately, the preabortion serostatus to N. caninum of the 4 aborting cows in this herd had not been determined. It appears from this study that the knowledge of the seroprevalence in herds before the onset of an abortion storm may contribute significantly to the understanding of the epidemiology of neosporosis. Therefore, serobanking should be encouraged. REFERENCES 1. Anderson ML, Blanchard PC, Barr BC, Dubey JP, Hoffman RL and Conrad PA. Neosnora-like protozoan infections as a major cause of abortion in California dairy cattle. J Am Vet Med Assoc 1991; 198: 241-244. 2. Anderson ML, Reynolds JP, Rowe JD, Sverlow KW, Packham AE, Barr BC, Conrad PA. Evidence of vertical transmission of Neosnora sp infection in dairy cattle. J Am Vet Med Assoc 1997;210:1169-1172. 3. Barr BC, Anderson ML, Blanchard PC, Daft BM, Kinde H, Conrad PA. Bovine fetal encephalitis and myocarditis associated with protozoa1 infections. Vet Path01 1990;27:354-361. 4. Barr BC, Bjerkas I, Buxton D, Conrad PA, Dubey JP, Ellis JT, Jenkins MC, Johnston SA, Lindsay DS, Sibley D, Trees AJ, Wouda W. Neosporosis - Report of the International Neospora Workshop. Compend Contin Educ Pratt Vet 1997; 19:5 1205126,5144.
244
Theriogenology
5. Barr BC, Conrad PA, Breitmeyer R, Sverlow IS, Anderson ML, Reynolds J, %hauvet AE, Dubey JP, Ardans AA. Congenital Neosnora infection in calves born from cows that had previously aborted Neospora-infected fetuses: four cases (1990-1992). J Am VetMed Assoc 1993;202:113-117. 6. Bartels CJM, Wouda W, Schukken YH. Risk factors for Neosnora caninum-associated abortion storms in dairy herds in The Netherlands (1995 to 1997). Theriogenology 1999;52:247-257. 7. Bjiirkman C, Johansson 0, Stenhmd S, Holmdahl JM, Uggla A. Neosnora species infection in a herd of dairy cattle. J Am Vet Med Assoc 1996;208:1441-1444. 8. Conrad PA, Sverlow K, Anderson M, Rowe J, BonDurant R, Tuter G, Breitmeyer R, Palmer C, Thurmond M, Ardans A, Dubey JP, Duhamel G, Barr B. Detection of serum antibody responses in cattle with natural or experimental Neosnora infections. J Vet Diagn Invest 1993;5:572-578. 9. Dubey JP, Lindsay DS. A review of Neosnora caninum and neosporosis. Vet Parasitol 1996;67:1-59. 10. Fetrow J, McClary D, Harman R, Butcher K, Weaver L, Studer E, Ehrlich J, Etherington W, Guterbock W, Klinborg D, Reneau J, Williamson N. Calculating selected reproductive indices: recommendations of the American Association of Bovine Practitioners. J Dairy Sci 1990;73:778-90. 11. French NP, Davison HC, Clancy D, Begon M, Trees AJ. Modelling of Neosnora species infection in dairy cattle: the importance of horizontal and vertical transmission and differential culling. Thrustield MV, Goodall EA (eds), Proc Sot Vet Epidemiol Prev Med, 1998;113-122. 12. Jenkins MC, Wouda W, Dubey JP. Serological response over time to recombinant Neosnora caninum antigens in cattle after a neosporosis-induced abortion. Clin Diagn Lab Immunol 1997;4:270-274. 13. Lind P, Kjeldsen AM, Bj?&man C, Uggla A, Willadsen C, Wedderkopp A, Jensen AM. Sero-epidemiological studies of Neospora infection in Danish cattle herds. VIIth International Coccidiosis Conference and European Union COST 820 Workshop, 1997:78 abstr. 14. McAllister MM, Huftinan EM, Hietala SK, Conrad PA, Anderson ML, Salman MD. Evidence suggesting a point source exposure in an outbreak of bovine abortion due to neosporosis. J Vet Diagn Invest 1996; 8:355-357. 15. McAllister MM, Dubey JP, Lindsay DS, Jolley WR, Wills RA, McGuire AM. Dogs are definitive hosts ofNeosoora caninum Int J Parasitol 1998;28:1473-1478. 16. Moen AR, Wouda W, Mul MGt EAM, Werven T van. Increased risk of abortion following Neosnora caninum abortion outbreaks: a retrospective and prospective cohort study in four dairy herds. Theriogenology 1998;49:1301-1309. 17. Pare J, Fecteau G, Fortin M, Marsolais G. Seroepidemiologic study of Neoswra caninum in dairy herds. J Am Vet Med Assoc 1998;213:1595-1598. 18. Pare J, Thurmond MC, Hietala SK. Congenital Neosnora canimtm infection in dairy cattle and associated calfhood mortality. Can J Vet Res 1996;60: 133-l 39. 19. Pare J, Thurmond MC, Hietala SK. Neosnora caninum antibodies in cows during pregnancy as a predictor of congenital infection and abortion. J Parasitol 1997;83:8287. 20. Schares G, Peters M, Wurm R, Tackmann K, Henning K, Conraths FJ. Neoswra caninum causes abortions in a cattle herd in North Rhine Westphalia. Dtsch Tieranzl Wochenschr 1997;104:208-212.
Theriogenology 21.
245
Thilsted JP, Dubey JP. Neosporosis-like abortions in a herd of dairy cattle. J Vet Diagn Invest 1989;1:205-209. 22. Thornton RN, Gajadhar A, Evans J. Neosnora abortion epidemic in a dairy herd. N Z Vet J 1994;42:190-191. 23. Thurmond M, Hietala S. Effect of congenitally acquired Neospora caninum infection on risk of abortion and subsequent abortions in dairy cattle. Am J Vet Res 1997;58: 13811385. 24. Thurmond M, Hietala S, Blanchard PC. Herd-based diagnosis of Neosnora caninuminduced endemic and epidemic abortion in cows and evidence for congenital and postnatal transmission. J Vet Diagn Invest 1997;9:44-49. 25. Wouda W, Brinkhof J, Maanen C van, Gee ALW de, Moen AR. Serodiagnosis of neosporosis in individual cows and dairy herds, a comparison of three enzyme linked immunosorbent assays. Clin Diagn Lab Immunol; 1998;5:711-716. 26. Wouda W, Ingh TSGAM van den, Knapen F van, Sluyter FJH, Koeman JP, Dubey JP. Neosnora abortus bij het rund in Nederland. Tijdschr Diergeneeskd 1992; 117599-602. 27. Wouda W, Moen AR, Schukken YH. Abortion risk in progeny of cows that experienced a Neosnora caninum abortion epidemic. Theriogenology 1998;49: 13 ll1316. 28. Wouda W, Moen AR, Visser IJR, Knapen F van. Bovine fetal neosporosis, a comparison of epizootic abortion cases and different age classes with regard to lesion severity and immunohistochemical identification of organisms in brain, heart and liver. J Vet Diagn Invest 1997;9:180-185. 29. Yaeger MJ, Shawd-Wessels SC, Leslie-Steen P. Neosnora abortion storm in a midwestem dairy. J Vet Diagn Invest 1994;6:506-508.