Most goats in the United States are raised for meat, and the meat goat industry is one of the fastest growing sectors of livestock production in the United States.1 Historically, US goats were kept principally for milk or fiber production or occasionally for brush control or as pets, but current management and production goals for meat goats are more diverse and include exhibition and development of breeding stock in addition to meat production. The accompanying evolution in breed predominance and management approaches has likely caused changes in disease-related epidemiological factors. To the authors’ knowledge, such changes have not been explored and reported. Moreover, the changes in management and subsequent improvement in genetic characteristics have resulted in individual animals having high financial value. This often justifies more aggressive diagnostic and therapeutic options in providing veterinary care, and guidance regarding these activities is sometimes lacking in the scientific literature.
Pregnancy toxemia is often associated with high mortality rates in small ruminants, typically occurring in overconditioned or underconditioned animals carrying multiple fetuses.2,3 The pathophysiologic mechanisms of PT are poorly understood, although disruption of glucose homeostasis may play a role.2 Inadequate energy intake results in fat mobilization and ketonemia, which negatively impacts appetite, depressing feed intake and thus perpetuating the disease cycle. Older, multiparous sheep and goats are generally considered at increased risk for clinical PT.2,4
Treatment goals have traditionally included correction of perceived negative energy balance in the dam, generally achieved by increasing energy intake or decreasing energy demand associated with the pregnancy (through induction of labor or by C-section). Additional therapeutic goals include addressing electrolyte and acid-base disturbances and remedying any concurrent disease processes. Commonly recommended practices for increasing energy intake in small ruminants include provision of a highly digestible high-energy diet, IV dextrose administration, and oral administration of propylene glycol.2,3 Unfortunately, these treatment approaches often fail, and the prognosis for goats with PT is usually poor, especially if the diagnosis is made late in the disease process.
Several fairly recent publications5–8 have described PT in dairy goats; however, information pertaining to PT in goats of other breeds (eg, those raised for meat) is lacking. To the authors’ knowledge, risk factors and prognostic indicators for survival have not been previously reported for this population of animals. The objective of the retrospective study reported here was to describe the signalment, history, clinical and clinicopathologic findings, treatments, and survival rates in a mixed population of does (meat and dairy breeds) with PT and to determine prognostic indicators for survival of these patients and their offspring.
Materials and Methods
Case selection criteria
Electronic and hardcopy medical records of all goats of any breed admitted to the Oklahoma State University Boren Veterinary Medical Teaching Hospital between August 1, 2002, and August 1, 2012, were reviewed to identify does with a diagnosis of PT. Keywords for the electronic search included “pregnancy toxemia” and “caprine.” Does that were reported or determined to be in late gestation and were identified by attending clinician as having PT on the basis of examination findings as well as ketonemia, ketonuria, or both were included in the study. No animals were included in the study more than once.
Medical records review
Information regarding signalment (including age and body weight at the time of hospital admission, breed, parity, and expected parturition date of does), history, and the results of initial physical examination, CBC and serum biochemical analysis, blood BHB concentration and pH (when available), urinalysis, and ancillary diagnostic tests (eg, ultrasonography) were recorded on a commercially available spreadsheet.a General treatment approaches (eg, IV administration of dextrose, oral treatment with glucose precursors, insulin administration, prostaglandin treatment to induce parturition, or delivery of offspring [kids] by C-section), time from hospitalization to delivery of offspring, number of days that patients were hospitalized, and outcomes (whether does survived to delivery of offspring, whether vaginal delivery [spontaneous or induced parturition] occurred or C-section was performed, and whether does and offspring survived to hospital discharge).
Procedures
Treatments for PT were determined at the clinician's discretion. Ketones in blood (specifically BHB) and urine were measured by means of a handheld glucometer-ketone meterb and urine dipstick,c respectively. Complete blood count and serum biochemical analysis were performed by the facility's in-house diagnostic laboratory. Blood glucose concentration and pH were assessed by a handheld glucometer-ketone meterb and point-of-care blood analyzer,d respectively. Body condition score was assessed by the attending clinician on a scale of 1 to 5 (including half scores).9 Body temperature of does was measured per rectum. Results for clinicopathologic tests and examinations performed at the time of hospital admission were used for study purposes.
Statistical analysis
Potential risk factors for death prior to hospital discharge were converted to categorical data for analysis. Results of CBC, serum biochemical, and urine variable analyses at the time of hospital admission were generally categorized as low, high, or within the reference range, with some exceptions.
Cut-off points for these additional study categorizations were assigned post hoc on the basis of natural breakpoints that were scrutinized for clinical relevance, the authors’ clinical experience, and the need to convert data to categories for statistical analyses. Serum bicarbonate concentration was categorized as severely low (< 15 mEq/L), low (15 to < 20 mEq/L), high (> 25 mEq/L), or within the reference range (20 to 25 mEq/L). Acid-base status was categorized on the basis of blood pH as moderate to severe acidosis (6.96 to 7.24), mild acidosis (7.25 to 7.40), alkalosis (> 7.48), or within the reference range (7.41 to 7.48). Ketonemia as assessed by BHB concentration was categorized as absent to trace (< 2 mmol/L), low to moderate (2 to 6 mmol/L), and high (> 6 mmol/L). Ketones in urine were categorized as low (5 to 15 mg/dL), moderate (16 to 80 mg/dL), and high (> 80 mg/dL).
Other variables related to doe health at the time of admission and analyzed for association with outcome included recumbency (not recumbent [not present or not reported] or recumbent [identified on initial examination or reported as relevant clinical history]), BCS (appropriate [2.5 to 3.5 on a scale of 1 to 5] or not appropriate [< 2.5 or > 3.5]), dehydration (not detected, mild to moderate [5% to 7%], or severe [≥ 8%]), duration of clinical signs (≤ 3 days, 4 to 6 days, or > 6 days), and time to expected parturition (≤ 7 days, 7 to 14 days, 15 to 21 days, and > 21 days). Other analyzed conditions or variables not necessarily noted at admission but determined during management of the case included diagnosis of dystocia (present or absent) and number of fetuses present (1, 2, 3, or 4).
Other factors related to doe health and treatment during hospitalization that were analyzed included propylene glycol administration, dextrose administration, insulin administration, or prostaglandin treatment for induction of parturition. Because cases were managed according to the clinicians’ judgement, no protocol was consistently used for frequency or duration of treatment, so these were categorized in a binary fashion (present or absent). Performance of C-section was categorized as not done, done on day 1 (the first day of hospitalization), done on day 2, done on days 3 through 6, or done on or after day 7. Other patient-specific factors included the number of days hospitalized (< 3, 3 to 5, 6 to 8, or ≥ 9), time in days from admission until delivery of offspring (1 to 2, 3 to 4, 5 to 7, 8 to 14, > 14, or death without giving birth), and rate of survival to hospital discharge for offspring (0%, 25% to 50%, 66% to 75%, or 100%). Peak and nadir fetal heart rates (high, low, or within the reference range [130 to 152 beats/min]) and whether vaginal delivery was assisted (yes vs no) were also analyzed.
Primary outcomes of interest were survival (to hospital discharge) versus death of the dam and percentage of offspring that survived. Relationships between variables of interest and these outcomes were examined by contingency table analysis. A Fisher exact test was performed for variables with 2 categories, and the likelihood ratio was determined for variables with ≥ 3 categories.
To assess the relationship between breed and a diagnosis of PT, the most frequently represented breed in the study was determined. The percentage of goats of this breed that had a diagnosis of PT in the final year of the study was compared with the percentage of goats of the same breed evaluated at the study hospital for any reason during the same year (October 1, 2011, to October 1, 2012) by the Fisher exact test. Comparison with the general patient population for breed across all years of the study was not possible owing to changes in medical record coding for breed. The distribution of number of fetuses per pregnancy for the study population was also compared with that for a collaborating university's herd of Boers during the same period by χ2 analysis to investigate a potential association with PT.
All analyses were performed with commercially available spreadsheet software.a Values of P < 0.05 were considered significant.
Results
Fifty-six does were initially identified in the records search. All 56 met the selection criteria and were included in the study. Breeds included Boer (n = 52), Pygmy (1), Saanen (1), La Mancha (1), and Anglo-Nubian (1). The Boer breed was overrepresented among PT patients (52/56 [93%]), compared with the general patient population in the last year of the study (83/150 [55.3%]; P < 0.001). Mean age of the population of does with PT at admission was 2.7 years (median, 2.9 years; range, 9 months to 6 years; n = 51). Mean body weight (n = 54) was 79.5 kg (174.9 lb; median, 80.5 kg [177 lb]; range, 29.1 to 107.3 kg [64 to 236 lb]), and mean BCS (36) was 2.7 (median, 2.5; range, 1.5 to 4.5 on the 5-point scale), with 15 of 36 (42%) does having scores between 2.5 and 3.5 (considered appropriate BCS). Ten of 17 does for which parity was recorded were primiparous, and all 10 were Boers. The median number of pregnancies, including that at the time of diagnosis, was 1 (range, 1 to 3; n = 17).
The overall rate of survival to hospital discharge for does with PT was 39 of 56 (70%). Forty-nine of 56 (88%) does survived to parturition or delivery of offspring by C-section, and 39 of these 49 (80%) survived to hospital discharge.
The number of clinical signs noted and reported by the owner at the time of hospital admission (the primary reason for evaluation, often described as chief complaint) ranged from 0 (n = 1) to 5 (2). Hyporexia or anorexia was reported for 36 of 55 (65%) does with data recorded. Twenty-two (40%) does were described as recumbent and 15 (27%) were described as weak, depressed, or lethargic (alone or in combination). Eleven (20%) had swollen limbs, lameness, or difficulty ambulating. Additional reported signs included respiratory abnormalities (rapid breathing, respiratory distress, or pneumonia; n = 3), weight loss (3), diarrhea (3), dehydration or not drinking (3), prolonged gestation (3), straining (3), bruxism (2), vocalizing (2), vaginal prolapse (2), rectal prolapse (2), dystocia (1), vaginal discharge (1), bloating (1), and abduction of the forelimbs (1). The mean reported duration of clinical signs prior to admission (n = 48) was 4.3 days (median, 2.75 days; range, 0.5 to 21 days).
On clinical examination, mean body temperature (n = 55 does) was 39.38°C (102.9°F; median, 39.44°C [103°F]; range, 37.5° to 41.8°C [99.5° to 107.2°F]; reference range, 38.6° to 39.7°C [101.5° to 103.5°F]). Sixteen of 55 (29%) does were hyperthermic, 7 (13%) were hypothermic, and 32 (58%) had results within the reference range. Tachycardia was identified in 43 of 55 (78%) does. Mean heart rate was 129 beats/min (median, 132 beats/min; range, 60 to 184 beats/min; reference range, 70 to 110 beats/min). Tachypnea was reported for 50 of 55 (91%) does (mean respiratory rate, 80 breaths/min; median, 70 breath/min; range, 24 to 200 breaths/min; reference range, 15 to 40 breaths/min). Auscultation of the number of rumen contractions was recorded for 42 does, with rumen hypomotility identified in 33 (79%). The mean value for all does was 1.4 contractions/2 min (median, 1 contraction/2 min; range, 0 to 5 contractions/2 min; expected value, 3 contractions/2 min). Of 43 does with hydration status reported, 24 (56%) were considered dehydrated on the basis of physical examination results (mean estimated degree of dehydration, 7.5%; median, 7.5%; range, 5% to 10%). Scleral injection was detected in 24 of 56 (43%) does, edema of the distal aspect of the limbs was noted for 17 (30%), and bruxism was observed in 9 (16%). Fifteen of 56 (27%) does were recumbent on initial examination.
Concurrent diseases included gastrointestinal parasitism (13/56 [23%] does), pneumonia (8 [14%]), vaginal prolapse (3 [5%]), suspected type 2 abomasal ulceration (2 [4%]), rectal prolapse (2 [4%]), mastitis (1 [2%]), and free gas bloat (1 [2%]). Twenty-nine does had dystocia. Death of does with PT prior to hospital discharge was not significantly associated with duration of signs prior to hospital admission, recumbency status, dehydration status, BCS, or diagnosis of dystocia.
Does for which an expected parturition date was known had a significantly (P = 0.03) higher survival-to-discharge rate (28/35 [80%]) than those without an expected parturition date (11/21 [52%]). Mean time to expected parturition was 11.2 days (median, 9.5 days; range, 1 to 30 days). None of the does with PT had a singleton pregnancy; the number of fetuses was recorded as 2 (11/55 [20%]), 3 (37 [67%]), or 4 (7 [13%]). For 1 doe, the number was not recorded but the report indicated > 1 kid was present; this doe was excluded from analyses related to the number of offspring. Neither of these variables was associated with death prior to discharge. However, the frequency of triplets and quadruplets per pregnancy in the study group was significantly (P < 0.001) different from that for the collaborating university's Boer herd during 2011 and 2012, which comprised 1 (70/429 [16.3%]), 2 (242 [56.4%]), 3 (105 [24.5%]), and 4 (12 [2.8%] fetuses/pregnancy.
Selected clinicopathologic data were summarized (results obtained for all does that had a CBC or serum biochemistry profile performed; Table 1). The most common hematologic abnormalities (detected in > 50% of animals tested) included leukocytosis (81%), neutrophilia (100%), hyperfibrinogenemia (58%), hyperglycemia (80%), hypoproteinemia (56%), hypoalbuminemia (78%), hyperbilirubinemia (59%), hypocalcemia (74%), hypomagnesemia (75%), acidosis (53%), and low serum bicarbonate concentration (73%). Of the 19 does with low serum bicarbonate concentrations, 8 and 11 had results classified as low and severely low, respectively. Seven does with low blood pH were subcategorized as having mild (n = 4) or moderate to severe (3) acidosis. Blood ketone (BHB) concentrations were reported for 18 does and were classified as absent to trace in 4, low to moderate in 6, and high in 8 does. The median blood BHB concentration (n = 18) was 5.6 mmol/L (range, 0.6 to > 8 mmol/L). Urine ketone concentrations were recorded for 46 does and were classified as low in 8 (17%), moderate in 8 (17%), and high in 30 (65%) does. Findings for blood glucose and BHB concentrations as well as serum total protein, albumin, creatinine, and calcium concentrations were not significantly associated with doe survival. However, 9 of 12 does with a high BUN concentration and 4 of 15 does with BUN concentration within or below the reference range died prior to discharge, and death was significantly (P = 0.021) more likely (OR, 8.25; 95% CI, 1.45 to 46.86) for does with high BUN concentrations. Death was significantly (P = 0.041) more common for does with low serum potassium concentration than for does with serum potassium values within the reference range (4/4 vs 9/23). Nineteen does had low serum bicarbonate concentrations; 11 of these had severely low values. Death was significantly (P = 0.013) more likely for does with severely low serum bicarbonate concentrations than those in other categories (9/11 vs 4/15; OR, 12.4; 95% CI, 1.83 to 83.77). The 3 does with moderate to severe acidosis died; the outcome of death was significantly (P = 0.029) more common for does of this category than for does with mild or no acidosis (3/3 vs 2/10). Does with moderate urine ketone concentrations were significantly (P = 0.008) more likely to die than does in other urine ketone categories (5/8 vs 7/38; OR, 7.14; 95% CI, 1.37 to 37.23).
Summary of selected hematologic variables and proportions of tested does with values above or below the reference ranges in a retrospective study to evaluate prognostic indicators for 56 goats with PT (52 Boers and 1 each of 4 other breeds) and their offspring.
| Variable | No. of does | Result > RR (No. [%]) | Result < RR (No. [%]) | Mean | Median (range) | RR |
|---|---|---|---|---|---|---|
| Blood values | ||||||
| WBCs (X103 cells/μL) | 26 | 21 (81) | 0 (0) | 19.5 | 19.6 (11.7–34.7) | 7–15 |
| Neutrophils (cells/μL) | 26 | 26 (100) | 0 (0) | 16,277 | 16,017 (8,925–32,965) | 1,800–4,500 |
| Band cells (cells/μL) | 26 | 7 (27) | NA | 159 | 0 (0–2,097) | ≤ 200 |
| Lymphocytes (cells/μL) | 26 | 0 (0) | 14 (54) | 2,287 | 2,196 (466–4,872) | 2,500–7,500 |
| Monocytes (cells/μL) | 26 | 8 (31) | 1 (4) | 719 | 655 (0–2,695) | 50–800 |
| Fibrinogen (mg/dL) | 26 | 15 (58) | 0 (0) | 333 | 303 (100–1,200) | 100–300 |
| Hct (%) | 26 | 3 (12) | 0 (0) | 24 | 25 (16–31) | 19–38 |
| Glucose (mg/dL) | 40 | 32 (80) | 2 (5) | 111.5 | 107 (42–315) | 50–75 |
| BUN (mg/dL) | 27 | 12 (44) | 1 (4) | 22 | 19 (8–59) | 10–20 |
| pH | 13 | 3 (23) | 7 (53) | 7.34 | 7.39 (6.96–7.57) | 7.41–7.48 |
| Serum values | ||||||
| Total protein (g/dL) | 27 | 0 (0) | 15 (56) | 6.3 | 6.2 (5.3–7.3) | 6.4–8.8 |
| Albumin (g/dL) | 27 | 0 (0) | 21 (78) | 2.4 | 2.3 (1.3–3.3) | 3.0–4.0 |
| Globulins (g/dL) | 27 | 0 (0) | 0 (0) | 3.9 | 4 (3.2–4.7) | 2.5–5.8 |
| GGT (U/L) | 25 | 2 (8) | 0 (0) | 37 | 33 (23–98) | 20–40 |
| Total bilirubin (mg/dL) | 27 | 16 (59) | NA | 0.3 | 0.3 (0.1–0.7) | ≤ 0.1 |
| Creatinine (mg/dL) | 27 | 3 (11) | 4 (15) | 1.2 | 0.9 (0.5–4.6) | 0.6–1.6 |
| Phosphorus (mg/dL) | 25 | 9 (36) | 0 (0) | 7.8 | 7 (4.1–13.1) | 3.4–8.6 |
| Calcium (mg/dL) | 27 | 0 (0) | 20 (74) | 8.2 | 8.4 (5.7–9.7) | 8.9–11.7 |
| Magnesium (mg/dL) | 24 | 1 (4) | 18 (75) | 2 | 2 (1.3–3.8) | 2.2–3.4 |
| Potassium (mg/dL) | 27 | 0 (0) | 4 (15) | 3.9 | 3.9 (3.1–4.6) | 3.4–5.5 |
| Bicarbonate (mEq/L) | 26 | 4 (15) | 19 (73) | 17 | 16 (5.0–33.0) | 20–25 |
Data were obtained at the time of hospital admission. Blood glucose concentration and pH were assessed with point-of-care equipment; remaining hematologic analyses shown were performed by the facility's in-house diagnostic laboratory.
GGT = γ-Glutamyltransferase. NA = Not applicable. RR = Reference range.
Commonly administered medical treatments included propylene glycol (PO; 11/56 [20%]), 2.5% to 10% dextrose solution (by continuous rate infusion; 53 [95%]), insulin (both long- and short-acting insulin products were used; 28 [50%]), and prostaglandins (for induction of parturition; 21 [38%]). Pregnancy toxemia was not considered resolved in any cases until delivery of offspring, and all does with PT required hospitalization for the remainder of pregnancy. Mean duration of hospitalization was 6.2 days (median, 5 days; range, 0.5 to 22 days). Survival-to-discharge rates of does hospitalized for < 3 days (4/14), 3 to 5 days (10/16), 6 to 8 days (12/13), and ≥ 9 days (13/13) differed significantly (P = 0.001), with does hospitalized for the shortest interval more likely to die, compared with those hospitalized longer (OR, 12.5; 95% CI, 3.04 to 51.47). Twenty-nine of 49 (59%) does that survived to the delivery of offspring had dystocia, and 7 of 49 (14%) underwent C-section. Four of 7 had prostaglandin treatment the day before surgery, and 3 did not. Three of the 7 (43%) does that underwent C-section and 36 of 42 (86%) does that did not require C-section survived to hospital discharge. Among the 49 does that survived to delivery of offspring, death prior to discharge was significantly (P = 0.025) more common for those that had a C-section than for those that did not (4/7 vs 6/42). Postparturient uterine abnormalities (retained fetal membranes, metritis, or both) developed in 18 of 49 (37%) does that survived to delivery of offspring.
Among the 17 does that did not survive to hospital discharge, 7 had dystocia, 4 underwent C-section, and 6 had prostaglandin treatment. Necropsy was performed in 15 does that died. Hepatic lipidosis was a common finding (12/15). Abomasal ulceration was present in 4 of the 15 does; 2 does with suspected gastrointestinal ulceration on initial examination were not included among these (1 had severe hemonchosis with intraluminal gastric hemorrhage, and 1 was not necropsied). Retained fetal membranes, metritis, or both were found in 3 of the 15 does. Death was attributed to perforating abomasal ulcers in 2 does, uterine rupture in 1, and uterine artery rupture in 1.
Propylene glycol administration, dextrose administration, and prostaglandin treatment were not associated with survival to discharge for the doe; however, does treated with insulin were significantly (P = 0.008) more likely to die than does that did not receive insulin (13/27 vs 4/29; OR, 5.80; 95% CI, 1.59 to 21.25). Insulin administration was significantly (P = 0.001) more common among does that subsequently underwent C-section (7/7) than among does that did not (20/49).
Survival to discharge for the doe was significantly (P < 0.007) associated with that of offspring. Twenty-two of 23 does with 100% offspring survival rates lived to hospital discharge, compared with 8 of 9, 5 of 5, and 4 of 18 does that had 66% to 75%, 25% to 50%, and 0% of offspring survive, respectively.
Survival of offspring to hospital discharge was assessed only if the dam survived to parturition or C-section. In this group (offspring of 48 does), survival was not associated with most clinical variables recorded for the dam, including recumbency, duration of clinical signs, BCS, presence or degree of dehydration, time to expected parturition (when known), or duration of hospitalization. Similarly, the dam's blood or urine ketone status, WBC count results (band neutrophil, neutrophil, monocyte, or lymphocyte concentrations), blood pH, Hct, BUN concentration, and serum concentrations of total protein, albumin, globulins, γ-glutamyltransferase, bilirubin, creatinine, magnesium, and potassium, and many treatment variables including administration of propylene glycol, dextrose, insulin, or prostaglandins were not associated with survival of offspring. Finally, the categories for peak fetal heart rate (which ranged from 106 to 244 beats/min), nadir fetal heart rate (which ranged from 80 to 169 beats/min), and whether vaginal delivery was assisted were not associated with this outcome.
Fetal survival was associated with the dam's glucose status at admission (P = 0.043). Of the 48 does that survived to delivery (of a known number of offspring), those that were hyperglycemic at admission had better offspring survival rates than those with normo- or hypoglycemia. Two of 7 does that underwent C-section had offspring that survived, compared with 35 of 41 does that did not have the procedure performed (P = 0.009). Survival rates were higher overall for offspring delivered vaginally (84/117) than for those delivered by C-section (3/20). Survival rates were significantly (P = 0.039) higher for twins (18/22 [82%]) than quadruplets (11/20 [55%]), with triplets being intermediate (58/96 [60%]). Does with serum phosphorus concentration within the reference range had greater offspring survival rates than those with hyperphosphatemia (P = 0.001), as did does with serum calcium concentration within the reference range, compared with hypercalcemic does (P = 0.044). Does with severely low or low serum bicarbonate concentrations had poorer offspring survival rates than did does with bicarbonate values within or greater than the reference range (P = 0.036).
Discussion
Our study population was primarily of the Boer breed (52/56 [93%]). Boers were overrepresented among PT patients, compared with the general population examined at our facility during the final year of the study. This finding should be interpreted cautiously, because we were unable to investigate breed prevalence in other years. Boer does may be at greater risk of PT than other goat breeds because of genetic selection or management practices, as many of the Boer does were raised and managed primarily for exhibition purposes prior to being bred. Several of the historical, clinical, and clinicopathologic findings as well as prognostic indicators (risk factors for death prior to discharge) in this study population deviated from many findings previously reported in ewes and dairy does with PT,2–6,8,10 potentially indicating a different disease pathogenesis in meat goats or in the Boer breed, although further investigation is needed to confirm these findings.
In the present study, the overall survival rate of does with PT was 39 of 56 (70%), and aggressive medical treatment was often employed. Investigation may be warranted to identify predisposing factors (which may be targets for modification to reduce incidence) and delineate effective management of PT in this population of animals.
Anorexia or hyporexia; recumbency; weakness, depression, or lethargy; and swollen limbs, lameness, or difficulty ambulating were common in our study population and have often been described as historical or clinical findings in small ruminants with PT.2,3 However, other primary reasons for evaluation for a proportion (each ≤ 3/56 [5%]) of our study population were less indicative of PT and included vocalizing, prolonged gestation, and potential respiratory signs (respiratory distress, rapid breathing, or pneumonia). Other reported signs, including straining (3 [5%]), diarrhea (3 [5%]), rectal or vaginal prolapse (2 [4%] each), and bloat (1 [2%]), were more indicative of gastrointestinal disease or dystocia. These results indicated that some owners may not perceive that PT is the primary disease process and may attribute the clinical signs to other conditions. It may therefore be prudent to rule out PT in all does evaluated during the last month of gestation, regardless of the clinical signs observed or reported.
Does for which an expected parturition date was known were more likely to survive to hospital discharge than does for which this date was unknown (P = 0.03). This likely reflected on herd management, suggesting that producers with better breeding oversight and recordkeeping were also more astute in observation of animals and detected clinical signs of illness earlier in the disease process. The number of days to expected parturition, when known, was not associated with doe survival, and reported duration of clinical signs prior to admission was not associated with survival of does or their offspring.
Most does in the present study for which parity was recorded (10/17) were primiparous. However, given the small number for which these data were available, this result may not have been representative of the larger population. Despite this caveat, it may be worth noting that this was in contrast to previous reports2,4 in which multiparity was identified as a risk factor for development of PT in small ruminants. This finding may have been attributable to management factors, including nutritional factors; additional investigation is warranted to assess predisposing causes of PT in primiparous does.
Body condition score was not associated with survival to discharge in this population of does; a relatively large proportion of animals that had the information available (15/36) were determined to have an appropriate BCS (2.5/5). Development of PT in these animals may have been attributable to a decreased plane of nutrition in late gestation or a primary metabolic derangement. Our findings suggested that PT in meat goats is infrequently related to the typically described scenarios for PT (ie, starvation PT and voluntary decrease in nutritional intake of obese animals resulting from reduced rumen volume following intra-abdominal fat accumulation). We speculate that PT in this population could have resulted from high basal energy demands, particularly among the meat goat breeds, which might have made it more difficult for them to meet demands for maintenance as well as fetal growth. Alternatively, deficiencies or defects in metabolism could predispose such goats to development of PT, even when seemingly managed appropriately, or PT could have been secondary to other disease processes, which were common among these patients. Such secondary development of PT could result from another disease process negatively impacting appetite or nutrient use (eg, gastrointestinal parasitism), inability to access feed (eg, lameness), or catabolic activities associated with an active immune response.
Common clinical findings in our study population of goats with PT included tachycardia, tachypnea, rumen hypomotility, and some degree of dehydration. Additional, less commonly reported clinical signs included scleral injection, edema at the distal aspect of limbs, and bruxism. Clinical signs related to potential neurologic dysfunction (eg, signs of depression) were rarely reported in this group of primarily Boer does, although they are commonly recognized as occurring in sheep and goats with PT.3 One reason for this finding could have been related to glucose status at admission; systemic hypoglycemia resulting in cerebral hypoglycemia has been postulated as the cause of neurologic dysfunction in ewes with PT,11 but most (38/40 [95%]) does in the present study that had blood glucose data recorded were normoglycemic or hyperglycemic.
Some clinical signs detected in does with PT, such as hyporexia or anorexia, bruxism, and diarrhea, could have been related to abomasal ulceration. Although not definitively diagnosed in any animals that survived, abomasal ulceration was present in 4 of 15 does at necropsy, and death for 2 of these does was attributed to perforating ulcers. It may be advisable to perform fecal occult blood testing in does with PT; prophylactic administration of effective gastroprotectants and acid-reducing medications may be warranted.
Most (37/55 [67%]) does in the study population were pregnant with triplets, and only 11 (20%) were pregnant with twins. The frequency of triplets and quadruplets in the study population (44/55 [80%] pregnancies) was significantly (P < 0.001) greater than that for a university herd of Boers chosen as a reference herd (117/429 [27.3%]). One report12 describing reproductive performance in Boer does under semi-intensive management conditions indicated that the mean number of kids expected per parturition was 1.8. However, the potential relationship between number of offspring present and survival of does was not explored in the study.
Mature neutrophilia was present in all (26/26) does that had a CBC performed at the time of admission, and leukocytosis was present in 21 (81%.) A transient physiologic stress response from endogenous steroid release could have accounted for this in some cases, as all animals were transported to the hospital; concurrent lymphopenia, an expected finding if this was the cause, was present in 14 (54%) does. Additionally, pregnancy in people often causes physiologic leukocytosis, primarily related to increased circulation of neutrophils.13 In some does, neutrophilia was likely related to an inflammatory condition, as a regenerative left shift was present in 7 (27%).
Previous reports4,5 have indicated that hypoglycemia is frequently found in ewes and dairy does with PT. However, hypoglycemia was detected at admission in only 2 of 40 (5%) does of the present study. Dysregulation of glucose metabolism, rather than negative energy balance alone, could account for this finding. Duehlmeier et al14 found reduced pancreatic insulin response and impaired insulin-dependent inhibition of ketone body formation during late gestation in ewes with PT and postulated that insulin resistance might play a role in the pathogenesis of ovine PT. Results of another study15 of ewes suggested that during late pregnancy, plasma insulin concentrations decrease with an increased number of fetuses. Similar mechanisms might explain the common finding of hyperglycemia in our study population of primarily Boer does that were pregnant with multiple fetuses. A stress response associated with transport to the hospital could also have masked preexisting hypoglycemia or normoglycemia. It has been suggested that hypoglycemia in dairy does with PT might indicate that the fetuses are viable, whereas hyperglycemia might indicate fetal death.6 In the present study, survival of offspring to hospital discharge was assessed only when the dam survived to their birth (48 pregnancies); among these, offspring of dams with hyperglycemia at admission had significantly (P = 0.043) greater survival rates than kids of does that were normo- or hypoglycemic.
Low serum potassium, high BUN, or severely low serum bicarbonate (< 15 mEq/L) concentrations as well as moderate to severe acidosis (blood pH < 7.25) on admission were each associated with death of does prior to hospital discharge. Hypokalemia is common in dairy goats with PT5,6 and could be partly attributable to many affected does being hyporexic or anorexic and thereby having reduced dietary potassium intake. However, in our study, hypokalemia was uncommon, affecting only 4 of 27 (15%) does. It was possible that presumed metabolic acidosis in many of these does resulted in transcellular potassium-hydrogen exchange with extracellular potassium shift, falsely increasing circulating potassium concentrations and potentially masking an overall potassium deficit (although blood pH [the measure of acidosis for this study] was only determined for a subset of does, other data such as blood BHB and serum bicarbonate concentrations suggested that additional animals may have been affected by metabolic acidosis). This could also partially explain the higher mortality rate detected in hypokalemic does in the study (4/4 vs 9/23 normokalemic does). Although serum creatinine concentrations were not associated with outcome, does with high serum BUN concentrations were 8.25 times as likely to die as does with BUN concentrations within or below the reference range. Does with very low serum bicarbonate concentrations were 12.4 times as likely to die as does of other bicarbonate categories. Previous research has suggested that blood pH, circulating bicarbonate concentration, base excess, and anion gap may be useful indicators for early detection of PT in does,7 and that does with naturally occurring PT have lower venous blood bicarbonate concentrations and lower blood pH than healthy pregnant control does.16 In another study,5 all dairy does that had PT with blood pH < 7.122 died. Increasing severity of metabolic acidosis has only recently been identified as associated with lower probability of survival in dairy does with PT8; to the authors’ knowledge, no other reports describing odds of survival or death in sheep or goats with PT on the basis of clinicopathologic variables have been published.
The degree of ketonemia, specifically BHB concentration, was not associated with death of does in the present study. Mixed results regarding this have been reported in small ruminants with PT. A previous report10 described higher circulating BHB concentrations in sheep with spontaneous PT that died, compared with results for those that survived. Although BHB concentration data were recorded for only 18 of 56 does in the present study, our findings were consistent with those recently reported for dairy goats by Lima et al,8 suggesting circulating BHB concentration is not a useful prognostic indicator for dairy does with PT. Moderate urine ketone concentration at admission (but not low or high concentration) was associated with increased likelihood of death for does in our study. This finding was not intuitive. However, ketone concentrations in urine may be more concentrated than those in blood. Therefore, this result might have been related to hydration status at admission or renal concentrating ability, rather than a true association with degree of ketonuria. This finding could also have been confounded by other variables.
Intensive medical management of does in our study was common, and this was likely beneficial to patient outcome. However, no single treatment was identified as necessary or sufficient for survival to hospital discharge. The only treatments significantly associated with outcome for does or their offspring were insulin administration and C-section. Treatment with insulin at any time during hospitalization was significantly (P = 0.008) associated with death of does. This was potentially an artifact, considering that does perceived to be more severely affected on admission were likely to have received more aggressive medical treatment earlier in the course of their stay. Additionally, does that later underwent C-section were significantly (P = 0.001) more likely to have received insulin than were does that did not have a C-section performed. This may have been the greatest potential confounder for the association between insulin administration and death and should be considered a possible explanation for the finding. However, it should also be considered that insulin could have been detrimental to case management, meaning that does treated with insulin were more likely to have a decline in condition warranting emergency C-section. Henze et al10 found that treatment with a combination of insulin, glucose and fructose infusions, and orally administered glucose precursors increased survival rates in ewes with spontaneous PT, compared with results for ewes that received glucose and fructose infusions or oral glucose precursor treatment alone. Additional investigation is required to determine the effects of insulin administration (with or without supplementary dextrose treatment) on survival rates of does with PT.
For the 49 does that survived to the delivery of offspring, undergoing C-section was associated with a lower rate of survival to hospital discharge, compared with does that did not have the surgery (3/7 vs 36/42, respectively; P = 0.025). This finding was not surprising and was in agreement with other reports of high mortality rates in small ruminants with PT undergoing C-section. Brounts et al17 described the overall rate of survival to discharge in small ruminants undergoing C-sections as 89 of 110 (81%); however, all ewes that had PT and underwent C-section died. Similarly, Lima et al5,8 reported mortality rates of 7 of 8 to 10 of 10 for dairy does with PT that underwent C-section. This could be attributable in part to small ruminants perceived as severely affected with PT often having a C-section performed as an emergency measure.
Dystocia (n = 29) and postparturient uterine abnormalities (18) were fairly common in does of our study population. Many (21/56) does had treatment to induce parturition, which is a risk factor for retained fetal membranes and can lead to metritis. Hypocalcemia was also common and could have resulted in some degree of uterine inertia. The authors of a study18 that included 9 pregnant goats with experimentally induced PT concluded that PT may adversely affect humoral immune responses, with significant (P < 0.01) decreases in serum concentrations of IgG, IgM, and IgA detected. Additionally, PT has been described as a predisposing factor for development of mastitis in sheep during the immediate post-partum period.19 Whereas post-partum mastitis was not common in the present study, it was possible that PT predisposed some animals to metritis because of adverse effects on the immune system.
Longer hospitalizations were significantly (P = 0.001) associated with a lower overall mortality rate in does; this could have indicated detection of the disease earlier in the process, permitting more effective management. However, it could have been attributable to the nature of the disease and the fact that does were not considered cured, and therefore not released from the hospital, until after parturition. Many severely affected does died soon after admission, lowering the mean days of hospitalization in that population. Longer hospitalization did not guarantee successful outcome, as death occurred after several days or even after a week of hospitalization. In addition, the duration of hospitalization may have been greater for does that survived because these animals were typically kept in the hospital until all surviving offspring were doing well.
Survival of does to hospital discharge was not significantly associated with reported duration of clinical signs, days to expected parturition, BCS, recumbency, or degree of dehydration on admission, nor was it associated with the number of fetuses present or diagnosis of dystocia. Lack of an association with outcome was surprising in some cases. For example, recumbency was often reported by owners (22/56 [39%]) or described as a clinical finding (15 [27%]) but was not associated with survival to discharge. This was in conflict with previous descriptions of PT in other small ruminants, for which recumbency has been associated with a poor prognosis.8 It would generally be assumed that more severely affected animals or those later in the course of the disease are likely to be refractory to treatment. However, it is also possible that lameness or familiarity with human interaction made some of these does reluctant to rise, which could help explain why recumbency was not associated with outcome. Further work is needed to confirm the findings reported here, but the study results may justify a shift away from relying on traditional findings as prognostic indicators for survival vs death in this population of animals.
Among does that survived to the delivery of offspring, survival to hospital discharge was significantly (P < 0.007) associated with survival of their offspring. This suggested that focusing treatment on attempting to save either the doe or the offspring may not be realistic. Rather, the choice appears to be attempting intensive and potentially costly treatment with the intent of saving both or attempting minimal treatment with the understanding that the prognosis is more likely to be poor or guarded for the doe and its offspring. Overall doe survival rate in this population of animals was 39 of 56 (70%). This was a higher survival rate than those reported for dairy goats with PT (1/10 [10%] to 8/32 [25%]) in other studies.5,6,8 This may reflect more intensive management practices and earlier identification of signs of PT than in other study populations, or it may suggest better resiliency in Boers than other breeds.
Survival of offspring in our study was not related to the blood or urine ketone status of dams at admission. In a report20 of 4 pregnant ewes with PT, low alveolar partial pressure of oxygen was present in 3 of 4 fetuses, with mildly or moderately increased circulating l-lactate concentrations in 4 of 4, suggesting poor placental perfusion. However, circulating BHB concentrations were not high in those fetuses, indicating that BHB likely does not cross the placenta in substantial amounts or cause fetal ketoacidosis in sheep.20 If this is also true in goats, it may help explain why maternal blood or urine ketone concentrations did not directly correspond to the survival rates of offspring. We also found that survival rates (among the offspring of does that survived to their delivery) were higher for twins (18/22) than for triplets (58/96) and quadruplets (11/20) and were higher for offspring delivered vaginally (84/117) than for those delivered by C-section (3/20).
Hepatic lipidosis was present in 12 of 15 does in the present study that underwent necropsy. In another report,5 7 of 7 dairy does with PT that died and were necropsied had evidence of hepatic lipidosis. Severe hepatic lipidosis has also been associated with negative outcomes for goats with PT, compared with results for those in which the disease was mild.21 Ultrasonographic examination with liver biopsy or liver biopsy during C-section could potentially offer prognostic information for does with PT, but these were not performed in any of the does in the present study.
Potential shortcomings of the study were largely a result of the retrospective nature of the report, in that not all data were available for every patient. Future studies are indicated that could confirm the findings that suggested Boer does were overrepresented in our facility population, compared with the general hospital population in 1 year. If these results are confirmed in a larger population, further study of disease pathogenesis and confirmation of prognostic indicators for disease outcome in this breed and in other meat goat breeds may be warranted.
Acknowledgments
Presented in abstract form at the American College of Veterinary Internal Medicine Forum in Seattle, June 2013.
No third-party funding or support was received in connection with this study or the writing of the publication. The authors declare that there were no conflicts of interest.
ABBREVIATIONS
| BCS | Body condition score |
| BHB | β-Hydroxybutyrate |
| CI | Confidence interval |
| C-section | Cesarean section |
| PT | Pregnancy toxemia |
Footnotes
Microsoft Excel, 2012, Microsoft Corp, Redmond, Wash.
Precision Xtra Ketone and glucose meter, Abbott Laboratories, Abbott Park, Ill.
Multistix 10 SG Reagent Strips, Siemens Diagnostics, Erlangen, Germany.
VetScan iStat, Abaxis North America, Union City, Calif.
References
1. USDA APHIS. US Meat Goat Operations. Veterinary Services Fact Sheet #641.0112. Fort Collins, Colo: USDA APHIS, 2012;1–4.
2. Rook JS. Pregnancy toxemia of ewes, does, and beef cows. Vet Clin North Am Food Anim Pract 2000;16:293–317.
3. Brozos C, Mavrogianni VS, Fthenakis GC. Treatment and control of periparturient metabolic diseases: pregnancy toxemia, hypocalcemia, hypomagnesemia. Vet Clin North Am Food Anim Pract 2011;27:105–113.
4. Pregnancy toxemia in sheep. In: Radostitis OM, Gay CC, Hinchcliff KW, et al, eds. Veterinary medicine: a textbook of the diseases of cattle, horses, sheep, pigs, and goats. 10th ed. Philadelphia: Saunders-Elsevier, 2007;1668–1671.
5. Lima MS, Pascoal RA, Stilwell GT, et al. Clinical findings, blood chemistry values, and epidemiologic data from dairy goats with pregnancy toxemia. Bovine Pract 2012;46:102–110.
6. Lima MS, Pascoal RA, Stilwell GT. Glycaemia as a sign of the viability of the foetuses in the last days of gestation in dairy goats with pregnancy toxaemia. Ir Vet J 2012;65:1–6.
7. González FHD, Hernandez F. Acid-base and electrolyte status during early induced pregnancy toxemia in goats. Vet J 2012;193:598–599.
8. Lima MS, Silveira JM, Carolino N, et al. Usefulness of clinical observations and blood chemistry values for predicting clinical outcomes in dairy goats with pregnancy toxaemia. Ir Vet J 2016;69:16.
9. Rankins DL, Pugh DG. Feeding and nutrition. In: Pugh DG, Baird AN, eds. Sheep and goat medicine. 2nd ed. St Louis: Elsevier, 2012;32.
10. Henze P, Bickhardt K, Fuhrmann H, et al. Spontaneous pregnancy toxaemia (ketosis) in sheep and the role of insulin. Zentralbl Veterinarmed A 1998;45:255–266.
11. Scott PR, Sargison ND, Penny CD, et al. Cerebrospinal fluid and plasma glucose concentrations of ovine pregnancy toxemia cases, inappetant ewes and normal ewes during late gestation. Br Vet J 1995;151:39–44.
12. Duričić D, Grizelj J, Dobranić T, et al. Reproductive performance of Boer goats in a moderate climate zone. Vet Arh 2012;82:351–358.
13. Paidas MJ, Hossain N. Hematologic changes in pregnancy. In: Paidas MJ, Hossain N, Shamsi TS, et al, eds. Hemostasis and thrombosis in obstetrics & gynecology. Oxford, England: Wiley-Blackwell, 2010;1–11.
14. Duehlmeier R, Fluegge I, Schwert B, et al. Insulin sensitivity during late gestation in ewes affected by pregnancy toxemia and in ewes with high and low susceptibility to this disorder. J Vet Intern Med 2013;27:359–366.
15. Moallem U, Rosov A, Gootwine E, et al. Plasma concentrations of key metabolites and insulin in late pregnant ewes carrying 1 to 5 fetuses. J Anim Sci 2012;90:318–324.
16. Tharwat M, Al-Sobayil F. Cord and jugular blood acid–base and electrolyte status and haematobiochemical profiles in goats with naturally occurring pregnancy toxaemia. Small Rumin Res 2014;117:73–77.
17. Brounts SH, Hawkins JF, Baird AN, et al. Outcome and subsequent fertility of sheep and goats undergoing cesarean section because of dystocia: 110 cases (1981–2001). J Am Vet Med Assoc 2004;224:275–279.
18. Hefnawy AE, Youssef S, Shousha S. Some immunohormonal changes in experimentally pregnant toxemic goats. Vet Med Int 2010;2010:768438.
19. Barbagianni MS, Mavrogianni VS, Katsafadou AI, et al. Pregnancy toxaemia as predisposing factor for development of mastitis in sheep during the immediately post-partum period. Small Rumin Res 2015;130:246–251.
20. Gomez DE, Kuthiala S, Liu HL, et al. Effect of maternal ketoacidosis on the ovine fetus. Can Vet J 2015;56:863–866.
21. Kotb AG, El-Razik WMA, El-Nagar EAM, et al. Ultrasonography and ultrasonographic guided biopsy of the liver in pregnant toxemia does. Adv Anim Vet Sci 2015;3:466–472.
