A prospective study of sow mortality in breeding herds.

This investigation was conducted to study the incidence and the causes of sow mortality in breeding herds. Data were obtained from 24 swine breeding herds with an average inventory of 3755 sows and served gilts for the total sample. Producers were involved for 12 consecutive months and agreed to submit to the diagnostic laboratory every dead or moribund sow and served gilt. The average herd death rate was 3.3% +/- 0.5 (SEM), but varied considerably among herds, ranging from 0% to 9.2%. A total of 137 sows and mated gilts died during the year, and these females had produced an average of 4.2 litters +/- 0.2 (SEM). The number of deaths was significantly higher during the months of July, August and October. The peripartum period appeared to be when sows were most at risk, with 42% of all deaths occurring during this short period of the reproductive cycle. The three major causes of death were heart failure (31.4%), torsions and accidents of abdominal organs (15.3%) and cystitis-pyelonephritis (8.0%). Other causes included endometritis (6.6%), uterine prolapses (6.6%), pneumonia (3.6%), gastric ulcers (3.6%), downer sow syndrome (2.2%), miscellaneous (8.0%) and unknown (14.6%).     

Some observations on cardiac failure in sows

Data pertaining to factors associated with cardiac failure were collected as part of a study on sow mortality in which cardiac failure was the major cause of death. A total of 116 sows from 24 herds was used and divided into two groups: group 1 comprised sows dead of cardiac failure (n = 43), and group 2, sows dead of other causes (n = 73). Individual sow records, clinical history, backfat thickness, body weight, heart weight, selenium content in liver, and urea concentration in aqueous humor were obtained for each sow. Results from the groups were compared using Student's t-test. Maximum and minimum outside daily temperatures were obtained for the study period. A stressful or demanding event for the cardiovascular system of sows was identified in 84% of the cases of cardiac failure: nearly two-thirds of the cases occurred during the peripartum period, and other stressful events included heat, fighting, mating, and transport. More than 40% of the cardiac failures occurred during the months of July and August. Sows dead of cardiac failure were heavier and fatter than sows dead of other causes. The former group also had a significantly smaller heart weight to body weight ratio; this difference was attributed to a greater body weight since the heart weights were similar in the two groups. The hepatic selenium contents and urea concentrations in aqueous humor were not significantly different between the two groups.     

Herd and sow-related risk factors for mortality in sows in group-housed systems

Mortality of sows is a major problem for pig production worldwide. In this study, we used hierarchical multivariable logistic analyses to investigate different risk factors for mortality at the sow and herd level in herds with group-housed pregnant sows. Data included 3652 pregnant and 1266 lactating sows from 34 sow herds. A clinical examination for 16 clinical signs was carried out for each sow, and information about 16 herd related factors was obtained by interviews. Farm records were used to obtain information about whether or not sows died suddenly or were euthanized within 3 months after the clinical examination. Factors increasing the risk of sow mortality in the gestation unit were solid pen floors (OR = 1.87), presence of vulva bites (OR = 1.73) and unwillingness to stand when approached (OR = 1.62). Factors increasing the risk of sow mortality in the lactation unit were pale vulva color (OR = 12.69), body leanness (OR = 4.11), and presence of shoulder ulcers (OR = 2.89). The estimated between herd variation was small. Thus, the findings for the sow level variables may be generally applicable for sows in herds with group housed systems.     

Causes of culling and death in sows.

Over a 4-year period, the annual number of culled sows in 9 Danish herds averaged 54.8 per cent of the year-sows and the number of culled sows in per cent of total number of farrowings averaged 25.8 (Table I). -- The culling rate varied considerably from herd to herd within the same year and from year to year (Tables I and II); however, the average annual culling rate for all the herds only presented small variations (Table II). The average number of litters reared per sow before culling was 3.6. The culling rate was higher in pedigree herds than in commercial herds, and it was highest in the small pedigree herds (Table III). The hygiene level in the herds and the introduction of new female breeding stock did not influence the culling rate (Table IV). A proportionally lower percentage of the sows was culled in herds where the dry and pregnant sows were housed in stalls and/or were tethered, as compared to herds where these sows were housed in pens (Table V). -- The culling rate in the age groups of sows with less than 8 farrowings remained at approximately the same level (Table VI). The main reason for culling was infertility problems, which were recorded in 41.4 per cent of the culled sows, while 16.7 per cent of the sows were culled because their litters were poor and/or small (Table VII). The mortality rate among the culled sows was 11.9 (Table VII), and the main causes of death were chronic pyogenic infections, which occurred in 25.5 per cent of the fatal cases (Table VIII). Certain aspects concerning the recording and calculation of culling rates in the different herds are discussed and it is emphasized that the culling rate per se may not have any direct relationship to the productivity in the herd.     

Risk factors for high sow mortality in French swine herds

Episodes of high sow mortality rates affect profitability of swine farms. However, relevant control actions are difficult to implement. The objective of this study was to identify the risk factors for high levels of sow mortality rate (HM) in French swine herds. A case-control study was carried out in 102 swine herds located in Brittany (western France). Level of sow mortality of a herd was quantified by the annual mortality rate using sow-days as denominator. Fifty-five (53.9%) herds which experienced a sow mortality rate over 5% were classified as HM herds. Logistic regression was used to assess associations of managerial practices and disease prevalence with the odds of HM. High prevalence of urinary tract infections, metritis or lameness were significantly associated with a HM herd status (P < 0.10, OR ranging from 3.4 to 5.2). Multiplying herds were herds at higher risk for sow mortality than commercial farrow-to-finish herds. Providing three meals per day instead of two to dry sows decreased the odds of HM. Feeding plans where the maximum daily amount of feed provided to lactating sows was lower than 8 kg and was reached before 15 d in lactation were related to lower odds of HM (P < 0.10). Average age at weaning of 28 d or more and/or small average litter size at birth (12 piglets or less) were associated with higher odds of experiencing HM.     

Sow performance and maternal behaviour in organic and conventional herds

Abstract

The study evaluates the first three parities of 144 Swedish Landrace×Yorkshire sows from three certified organic and three conventional commercial piglet-producing herds. Sows and litters were weaned at seven and five weeks post partum (pp) in organic and conventional herds, respectively. Herdsmen judged sow behaviour and maternal ability at farrowing, two weeks pp and at weaning. Herdsmen collected registrations of litter size, litter weight and sow fat status at farrowing and at weaning. Piglet mortality and oestrous signs were registered continuously during the nursing period. Sows in organic herds were scored to be more cautious among piglets, and to have better nursing behaviour and maternal ability in early lactation than sows in conventional herds. Piglet mortality was higher in organic than in conventional herds. Oestrus during lactation was only observed among sows in organic herds, and sows showing oestrus during lactation were fatter at weaning.     

Clinical signs and their association with herd demographics and porcine reproductive and respiratory syndrome (PRRS) control strategies in PRRS PCR-positive swine herds in Ontario

The purposes of this study were to describe the clinical signs observed in PRRS positive herds during a porcine reproductive and respiratory syndrome (PRRS) outbreak in Ontario and to determine associations between these clinical signs and herd demographics and PRRS control strategies. All PRRS polymerase chain reaction-(PCR)-positive submissions to a diagnostic laboratory between September 1, 2004 and August 31, 2007 were identified (n = 1864). After meeting eligibility requirements and agreeing to voluntary study participation, producers from 455 of these submissions were surveyed for information on clinical signs observed in their herds, herd demographics, and PRRS control strategies used in their herds at the time that the PCR-positive samples were taken. Larger herd size was associated with an increased risk of reporting abortion, weakborn piglets, off-feed sows, and sow mortality in sow herds, and with an increased risk of reporting mortality in finishing herds. When disease control strategies were examined, use of a commercial PRRS vaccine in sows and gilts was associated with a decreased risk of reporting weakborn pigs and high pre-weaning mortality, while the use of serum inoculation in breeding animals was associated with an increased risk of reporting off-feed sows and sow mortality. Providing biofeedback of stillborn/mummified piglets, placenta or feces to gilts was associated with an increased risk of reporting respiratory disease and mortality in finishing pigs while all-in/all-out flow in farrowing rooms was associated with an increased risk of reporting sow mortality and weakborn piglets.     

Housing of Pregnant Sows in Loose and Confined Systems–a Field Study 1. Vulva and Body Lesions,Culling Reasons and Production Results

A field study was carried out in 18 herds with loose housing of pregnant sows (loose herds) and in 18 herds with tethered or stalled pregnant sows (confined herds). Three of the loose herds were excluded due to different kind of flooring from the rest of the herds. The remaining 15 herds had partly slatted concrete floors and electronic sow feeding. The frequencies of sows with vulva and body lesions, thin sows as well as the culling reasons and production results were used as animal welfare indicators for the herds. Vulva lesions were found only in the loose herds and the mean prevalence proportion of sows with lesions within these herds was 15.2%. All vulva lesions observed in this study seemed to be caused by biting. The relative risk of vulva lesions was 2.6 times higher in the loose herds with no roughage feeding as compared to loose herds with appetite feeding of roughage. The sows in the loose herds, that had a feeding station with a mechanical hind gate had 1.8 greater risk of vulva lesions than sows in the loose herds that used a feeding station with an electronic gate.The mean prevalence proportion of sows with body lesions was 13.1% in the loose herds and 4.0% in the confined herds. Aggression between sows seemed to be the main cause of body lesions in the loose herds, while decubitus ulcers on the shoulders were the main cause of body lesions in the confined herds. Sows in loose herds that were not fed additional roughage feeding had 1.7 times greater risk of body lesions than sows in herds that used additional roughage feeding.The main culling reasons and production results were similar in the loose and confined herds. This study showed that there were welfare problems both in confined and loose herds, however, with improved management, many of the welfare problems associated with loose housing can be reduced.     

Associations ofAscaris suum and Oesophagostomum spp. infections of sows with management factors in 83 Danish sow herds

Associations of the occurrence of Ascaris suum and Oesophagostomum spp. in sows with herd management factors were examined using logistic-binomial regression. The information used was from 11 to 78 sows sampled from each of 83 breeding herds (2961 sows in total) and examined for eggs per gram of faeces (EPG). A sow excreting at least 20 EPG was defined as ‘infected’. Management factors of the study herds were recorded using a questionnaire. A total of 263 sows from 50 herds (8.9% of sows) and 375 sows from 20 herds (12.7% of sows) were infected with A. suum and Oesophagostomum spp., respectively. For A. suum, sows from herds with more than 85 sows had significantly higher odds of being infected compared with those from 30 to 85 sows (P < 0.05). When bedding was provided for sows, the odds of A. suum infection was 5.4 compared with sows from herds in which bedding was not provided (P < 0.05). For Oesophagostomum spp., sows from herds with different specific pathogen free status had about one tenth the odds of being infected compared with those from conventional herds (P < 0.05). Sows that had been treated with anthelmintics had very low odds of being infected with Oesophagostomum spp. compared with those that were not treated (P < 0.05). The effects of these management factors in both final models did not differ when the definition of an ‘infected’ sow was changed. The present results suggest the importance of disposal of bedding material from pens in reducing the prevalence of A. suum in larger sow herds. Anthelmintic treatment is important in reducing the prevalence of Oesophagostomum spp. infection of sows.     

猪场母猪与仔猪猪圆环病毒2型感染情况调查

为了解目前母猪和仔猪猪圆环病毒2型(PCV2)感染情况,本试验对来自于15个猪场的85份母猪血清和29个猪场的55份仔猪可疑病料(脾脏、淋巴结、肾脏等),分别采用间接ELISA和PCR法进行PCV2抗体和病毒核酸检测。结果显示,在15个猪场中,1个猪场母猪抗体呈阴性,其他14个猪场母猪抗体均呈阳性,猪场PCV2阳性检出场为93.3%(14/15),85份母猪血清抗体总阳性率为75.3%(64/85)。29个猪场的仔猪,PCV2核酸检测阳性猪场为19个,猪场阳性率为65.5%(19/29),55份仔猪可疑病料病毒核酸检测总阳性率61.8%(34/55)。可见,母猪和仔猪感染PCV2相当普遍。对母猪群PCV2抗体阳性率及其仔猪群病毒核酸阳性率进行比较发现,母猪群PCV2抗体阳性率越高其仔猪群病毒核酸阳性率却越低。     

Epidemiological studies of piglet diarrhoea in intensively managed Danish sow herds. I. Pre-weaning diarrhoea

The study comprised 70,796 litters in 104 sow herds, observed from 1976 through 1982. Weaning age decreased from approx. 42 days to approx. 30 days during the observation period. Diseases and symptoms were recorded together with production parameters (feeding, barn construction, economic returns etc.). The mean incidence rate of pre-weaning diarrhoea was 6.8% of litters, with considerable inter-herd differences (incidence rates from 0 to approx. 50%). There was a slight increase in incidence during the autumn (August through October). Incidence rates increased with litter size, with a peak incidence in litters of 11-13 piglets, and decreased with increasing parity of the sow. There was a positive association between occurrence of arthritis and pre-weaning diarrhoea in the litters, and litters from sows with post parturient disease (MMA complex) had 1.8 times higher risk of getting diarrhoea than litters from healthy sows. No important differences among breeds were found. Small herds (less than 200 farrowings per year) had higher incidence rates than large herds (400-499 farrowings per year). Herds with a gilt proportion above 30% had an incidence rate of 12.3%, i.e. nearly twice as high as the overall mean (6.8%). There was a trend towards a higher incidence rate in litters kept in traditional pens (i.e. large pens with solid floor and loose sows) than in intensive pens (i.e. small pens with slatted flooring and tethered sows). The overall pre-weaning mortality rate was 16.2% of pigs born, half of which was due to stillbirths (6.3%) and overlaid piglets (2.2%). In litters with pre-weaning diarrhoea, the mortality rate was 19%, compared to 13% in litters without occurrence of diarrhoea. This difference accounts for an excess loss of 0.6 piglets from birth to weaning in diarrhoeic vs. non-diarrhoeic litters. Piglets from litters with pre-weaning diarrhoea had reduced weight gain. Thus, on the average, they were 2.2 days older at 25 kg bodyweight and weighed 0.4 kg less at 30 days than piglets from non-diarrhoeic litters. Also, litters with pre-weaning diarrhoea had a significantly increased risk of post-weaning diarrhoea. The present information forms a basis for defining acceptable disease thresholds in suckling litters in intensively managed herds.     

Mortality, death interval, survivals, and herd factors for death in gilts and sows in commercial breeding herds

The objectives of this study were to measure death intervals and survival, to determine mortality rate and mortality risks, and to investigate the association of herd factors with mortality risk in individual female pigs. This study was conducted by obtaining female data with lifetime records of 65,621 females born between 1999 and 2002, and herd data with mean measurements of 5 yr from 2000 to 2004 in 105 herds. Annualized mortality rate was calculated as the number of dead females divided by the sum of life days in all gilts and sows, multiplied by 365 d. Mortality risk was calculated as the number of dead females divided by the number of surviving females at farrowing in each parity. Death interval in gilts was defined as the number of days from birth to death, and that in sows was the number of days from the last farrowing to death. A Cox proportional hazards model was used to obtain the survival probability by parity. Logistic regression analyses were used to investigate the association of herd factors with mortality risk in individual females in each parity. Of the 65,621 females, the mortality risk was 9.9%, and the annualized mortality rate was 3.9%. Of the 6,501 dead females, death intervals in gilts and sows were 294.7 and 55.0 d, respectively. In gilts, survival probability rapidly decreased at 33 and 50 wk of age, around the first mating and the first parturition. In contrast, survival probability in sows decreased at wk 1 after farrowing, and rapidly decreased at wk 20 and 21 after farrowing in all parity groups that were around a subsequent peripartum period. The percentages of death on wk 0, 1, and 2 after the last farrowing in all the dead sows were 6.5, 23.5, and 10.1%, respectively. Approximately 10% of deaths also occurred from wk 20 to 21 after the last farrowing. Death interval in parity > or =5 was the shortest among all parity groups (49.2 d; P < 0.05). Mortality risks in parities 0 and 1 were 1.44 and 1.83%, respectively. As parity increased from 2 to > or =5, mortality risk increased from 1.63 to 5.90%. Herd factors (greater herd mortality, less herd productivity, and smaller herd size) were associated with greater mortality risk in individual females in parity 0 to > or =5, parity 4 and > or =5, and parity 1 to 4, respectively (P < 0.05). In conclusion, females in peripartum periods, gilts, and high-parity sows are at a greater risk of dying. Increased care should be implemented for prefarrowing females and early-lactating sows.     

The role of spatial factors in the success of an Aujeszky's disease eradication programme in a high pig density area (Northeast Spain, 2003–2007)

We tested the role of several spatial variables on the risk of a sow herd being Aujeszky's disease virus (ADV) seropositive in certain areas of North Eastern Spain and during different periods of the eradication programme. Distance to the nearest slaughterhouse, distance to the nearest conventional road and number of ADV serologically positive sows and ADV serologically positive fattening pigs within different distances (1000, 1500 and 2000 m) of each sow herd, were included in a hierarchical Bayesian binomial model. A variable without spatial characteristics, type of herd (farrow to weaning and farrow to finish), was also included. Presence of positive fattening pigs or positive sows up to a distance of 1500 m of a sow herd increased its risk of being seropositive, although this variable had no effect on the risk when located at distances up to 1000 or 2000 m. The number of seropositive sows increased the risk of a sow herd being ADV seropositive only in the first period of study, when the proportion of serologically positive sow herds was nearly 60%. The spatial pattern of the residuals of the hierarchical Bayesian binomial model (observed versus predicted) was very similar to the observed infection in sow herds in all of the eradication periods, showing that spatial factors might not be the main factors related to the eradication of Aujeszky's disease from sow herds. Other herd-specific risk factors might be much more strongly related to the risk of a sow herd being ADV seropositive.     

Prevalence of and risk factors for influenza in southern Ontario swine herds in 2001 and 2003

This research included 2 prevalence studies and a risk-factor investigation conducted in 2001 at 93 sites with sows only, finishers only, or both. In 2001, 1300 serum samples from sows in 65 herds and 720 serum samples from finisher pigs in 72 herds were tested for antibodies to swine influenzavirus (SIV) of H1N1 subtype with an enzyme-linked immunosorbent assay (ELISA). In 2003, 1140 serum samples from sows in 76 herds were tested for antibodies to SIV of H3N2 subtype with a hemagglutination-inhibition assay based on A/Swine/Colorado/1/77 and A/Swine/Texas/4199-2/98 isolates. The apparent pig-level H1N1 seroprevalence in 2001 was 61.1% and 24.3% in sows and finishers, respectively. The apparent pig-level seroprevalence in 2003 for H3N2 A/Sw/CO/1/77 and A/Sw/TX/4199-2/98 in sows was 0.6% and 0.7%, respectively. The factors associated with sow-herd H1N1 positivity included pig or farm density at different geographic levels, an external source of breeding pigs, number of animals on site, and decreasing proximity to other barns. Higher-parity sows had higher odds of seropositivity, but there was significant random variability in this association among herds. The odds of finisher-herd SIV positivity were higher with large herd size, high pig farm density, and farrow-to-finish type of farm. Finisher herds were SIV-positive only if source sow herds were positive. Simultaneously, 45% of finisher herds were SIV-negative although sow source herds were positive.     

Seroprevalence and risk factors associated with seropositivity in sows from 67 herds in north-west Germany infected with Mycoplasma hyopneumoniae

Risk factors for the spread of Mycoplasma hyopneumoniae in sows have not been studied although vertical transmission from sows to their offspring is considered a significant risk factor in the development of enzootic pneumonia in growers and finishers. Seropositivity for M. hyopneumoniae in sows, as assessed by commercial ELISA, is a possible indicator of infection pressure among sows. The objective of this study was to estimate seroprevalence and associated risk factors of a sow being seropositive for M. hyopneumoniae. A cross-sectional study was carried out in 2578 sows from 67 herds in north-west Germany. Data concerning general herd characteristics, acclimatisation practices, indoor and outside contacts, as well as data describing the immediate local environment were collected during a herd visit via questionnaire. Blood samples were seropositive in 65% of the 2578 sows, and all herds had ≥14% seropositive sows. Data analysis was performed in two steps. First, univariate analysis of predictor variables for the risk of a sow being seropositive for M. hyopneumoniae was performed using chi-square test. Secondly, all variables associated with the risk of a sow being seropositive (P ≤ 0.25) were included in a multivariate model using a generalised linear model. The risk of a sow being seropositive for M. hyopneumoniae was increased in herds with two- or three-site production (OR 1.50), when piglets were not vaccinated against M. hyopneumoniae (OR 1.81), in herds with a 2-week farrowing intervals (OR 1.84) and in herds without all-in/all-out management of the farrowing units (OR 1.37). The lack of an acclimatisation period for replacement boars was also associated with the risk of a sow being seropositive (OR 2.10). The results indicate that M. hyopneumoniae seropositivity is common in sows in north-west Germany and is influenced by various management factors. It is recommended that evaluation of sow herd management should be included in any strategic health plan to control M. hyopneumoniae infection.     

Salmonella seroprevalence at the population and herd level in pigs in The Netherlands

The aim of this study was to provide baseline data on the population and herd Salmonella seroprevalence in sows and finishers. For the population estimates in 1996 and 1999 and the herd prevalences for sows and gilts, blood samples from swine vesicular disease (SVD) and pseudorabies monitoring programmes were used and tested in an indirect Salmonella enzyme-linked immunosorbent assay (ELISA). The herd prevalence for finishers was determined using blood samples collected at two slaughterhouses.The population prevalence for finishers in 1996 and 1999 was 23.7 and 24.5%, respectively, and for sows 40.5 and 60.4%, respectively. The prevalence in free range (FR) finishers was significantly higher (44.6%) than in intensively housed finishers in 1999, identifying a hazard group for possible extra pork and pork product contamination. Of 406 finishing herds, 9% were completely seronegative for Salmonella (cut-off OD%>10). Of these 406 finishing herds, 69.7% had Salmonella-status I (low prevalence), 21.7% status II (moderate prevalence) and 8.6% status III (high prevalence) (cut-off OD%>40). In 46 multiplying sow herds, 20 breeding sow herds and 20 matching replacement gilt herds, the average herd prevalences were 54, 44.4 and 19.3%, respectively. Two gilt herds were completely seronegative. The prevalence in the gilt herds was never higher than in the matching breeding sow herds. Agreement on methodology and calibration of ELISA tests would make these results comparable between countries and is a prerequisite for a co-ordinated and integrated program to reduce Salmonella in pork in the European Union.     

The reproductive performance of sows raised by smallholder farmers in the Philippines.

We conducted a cross-sectional study at two sites in the Philippines to obtain baseline information about the reproductive performance of smallholder sows, and to identify important constraints and opportunities for smallholder pig raisers. Using an interview-based questionnaire, we collected data about 99 (with 128 sows) and 123 (144) herds at the northern and southern study sites, respectively. A range of animal- and herd-level measures of sow productivity were assessed, including average interfarrowing interval (9.7 and 6.6 mo in herds at the northern and southern sites, respectively), number of liveborn piglets (8.5 and 8.4) and preweaning mortality (19.0% and 12.8%). We have identified a number of animal- and herd-level factors associated with the productivity of smallholder sows in the Philippines, including aspects of breeding and piglet management and nutrition.     

Sow removal in Swedish commercial herds

The objective of this study was to describe the sow removal pattern in Swedish commercial piglet producing herds, with group-housing of non-lactating sows. The data collection, based on a PC herd monitoring program used in 21 piglet producing herds in the south-central part of Sweden, was prospective, started in 2002 and continued for 3 years. During this period, information on 14 234 removed sows was recorded. For all parameters studied a large variation between herds was found. The average annual removal rate was 49.5% (range 33.6–66.3%). Among the removed sows, 85.2% were sent to slaughter, 10.5% were euthanised and 4.3% were found dead. The most common removal reasons were ‘reproductive disorders’ (26.9%), ‘old age’ (18.7%) and ‘udder problems’ (18.1%), followed by ‘low productivity’ (9.5%), ‘lameness and/or foot lesions’ (8.6%) and ‘traumatic injuries’ (7.1%). The proportion of sows sent to slaughter was higher than the proportion euthanised for all categories of removal except ‘traumatic injuries’ (in which 81.6% of the sows were euthanised). The average parity number at removal for all sows was 4.4. Lowest average parity numbers were found for sows removed due to ‘traumatic injuries’ (2.6). Among removed sows, the proportion removed per parity decreased up to parity 6. Sows removed due to ‘old age’ had the highest piglet production, while sows removed due to reproductive disorders had the highest number of non-productive days (NPD) per parity and the longest interval from last farrowing to removal. With increasing removal parity number, the number of piglets produced increased and NPD decreased. In lower parities (1–3) most sows were removed due to ‘reproductive disorders’, but this proportion fell as parity number rose. Removals due to ‘udder problems’ were most frequent in medium parity numbers (4–6). Most sows removed after parity 7 were removed due to old age. The proportion of the sows found dead was highest just after farrowing; otherwise only a low percentage of sows were removed during the lactation period. The proportion of euthanised sows was high in the week after farrowing and during the 4-week period following weaning. The proportion of sows sent to slaughter increased considerably 4 weeks after farrowing (i.e. after weaning) and stayed high during the subsequent 3 weeks. The study revealed a considerable amount of unplanned removal of young sows. Not only the sows sent to slaughter, but also those euthanised and found dead, must be included in studies on sow removal.