Prognostic indicators for survival in goats treated for toxic mastitis: A retrospective study

Abstract

Objective

The objective of this study was to retrospectively assess medical records of goats presented to the J.T. Vaughan Large Animal Teaching Hospital at Auburn University (Alabama, USA) for evaluation and treatment of toxic or gangrenous mastitis, and to identify prognostic indicators for survival to hospital discharge.

Animals and procedure

Medical records of 35 postpartum female goats treated for toxic or gangrenous mastitis from December 2012 to January 2025 were analyzed. The diagnosis was based on mastitic changes to the mammary gland and milk, accompanied by systemic compromise. Following comparisons between discharged and non-discharged goats, continuous and categorical variables were included in univariate and multivariate regression models to evaluate their associations with the outcome.

Results

Of the study animals, 74% (26/35) survived to hospital discharge and the remaining 26% (9/35) died or were euthanized. Odds of hospital discharge increased [odds ratio (OR): 2.98; 95% confidence interval (CI): 1.44 to 7.97] for each 1°C increase in rectal temperature at admission (P = 0.002). However, odds of hospital discharge increased (OR: 11.21; 95% CI: 1.31 to 231.54) for each 1 g/dL increase in serum globulin concentration at admission (P = 0.02). For each additional day of hospitalization, goats treated for toxic mastitis were 2.35× more likely to be discharged from the hospital (OR: 2.35; 95% CI: 1.43 to 2.89; P < 0.01). Intramammary antibiotic (IMA) therapy increased the likelihood of hospital discharge 5.86× (95% CI: 1.28 to 31.5) compared to no IMA therapy (P = 0.02).

Conclusion and clinical relevance

Goats treated for toxic mastitis at the J.T. Vaughan Large Animal Teaching Hospital had a favorable prognosis for survival and discharge from the hospital. Hospitalization to ensure sufficient supportive care, as well as treatment with IMA, may improve survival in select cases.

INTRODUCTION

Severe clinical mastitis, also referred to as “toxic” or “gangrenous” mastitis, is a relatively uncommon but life-threatening disease of lactating does (1). Although the disease usually affects individual animals and not the entire herd, its economic impact on caprine dairy herds can be substantial due to lost production, treatment costs, and mortality (2). According to sheep and goat inventory surveys reported by the National Agricultural Statistics Service in 2025, the goat population in the United States is > 2.5 million, including 430 000 animals on dairy operations (3).

Although the incidence of toxic mastitis is low, with < 5% of the herd affected annually (4), complications of severe illness extend far beyond decreased milk production. Goats affected by toxic mastitis are typically presented with weakness, dehydration, and occasionally, signs of severe hypovolemia and shock, including inability to stand, weak pulse, cold extremities, and collapse (5). In addition, the affected mammary gland is often enlarged, firm, and warm or cold to the touch. A subtype of toxic mastitis, termed “gangrenous mastitis,” is characterized by a reddish to bluish demarcation of the affected mammary gland and serosanguinous mammary discharge. Without prompt and adequate treatment, the case fatality rate in cases of toxic mastitis in sheep and goats can be up to 40% (6). Common causative agents of toxic mastitis in goats include Staphylococcus aureus, non-hemolytic Staphylococcus spp., and Bacillus spp. (7).

Despite the clinical importance of this disease, prognostic indicators for survival in goats treated for toxic mastitis remain poorly documented in scientific literature. Enhancing veterinarians’ access to evidence-based prognostic information may improve therapeutic decision-making and lead to better clinical outcomes. The objective of this study was to retrospectively assess medical records of goats that were presented to the J.T. Vaughan Large Animal Teaching Hospital at Auburn University (Alabama, USA) (AULATH) for evaluation and treatment of toxic or gangrenous mastitis over a period of 13 y, and to identify prognostic indicators for survival to hospital discharge.

MATERIALS AND METHODS

Data collection

The online medical records of female goats presented to the AULATH from December 31, 2012 to January 18, 2025 were retrospectively reviewed. Inclusion criteria included female goats of any breed and age admitted to the hospital for the treatment of severe clinical mastitis. A case of “severe clinical mastitis” was defined as a female goat in lactation exhibiting signs of mastitis [i.e., enlarged, firm, and painful mammary gland (in one or both halves); abnormal mammary discharge (such as serous, serosanguineous, or mucopurulent fluid); and possible discoloration of the affected gland] accompanied by dehydration and lethargy. Animals were excluded from the study if comorbidities (e.g., pneumonia, hepatitis, metritis, peritonitis) in addition to severe clinical mastitis or diagnoses of neoplasia or fibrosis of the mammary gland were recorded.

Information retrieved from each animal’s record included signalment [sex, breed, purpose (meat, dairy, or unknown), age, weight, and days post-partum]; data from initial physical examination [heart rate (HR), respiratory rate (RR), and rectal temperature (RT)]; and data from initial blood work [complete blood (cell) count (CBC), serum biochemistry panel, or blood gas analysis]. Additional information obtained from individual medical records included time elapsed from when the owner noticed clinical signs to admission to AULATH, days of hospitalization, fecal parasite egg count, milk culture results, treatments, and outcome (discharged from the hospital or not discharged).

Statistical analysis

Statistical analyses were done using RStudio integrated development environment version 4.3.3 (RStudio, Boston, Massachusetts, USA). Data distribution plots and the Shapiro-Wilk test were employed to determine normality of the data, and comparisons were made between the discharged and non-discharged groups. Normally distributed continuous variables were analyzed using the Student’s t-test and results are presented as mean ± standard deviation. Non-normally distributed continuous variables were compared using the Wilcoxon test and results are presented as median with interquartile range. Categorical variables were analyzed using the Fisher exact test. Differences with a P-value < 0.05 were considered statistically significant.

Additional analyses aimed to examine how different factors were associated with 1 binary outcome: whether an animal was discharged from the hospital or not. Due to the relatively small dataset (n = 35; 26 discharged, 9 not discharged), with some categorical variables having sparse levels and evidence of potential separation (e.g., days of hospitalization nearly separating outcomes), we used the Firth penalized logistic regression for univariate analyses rather than standard logistic regression. This approach helps reduce bias and instability in maximum likelihood estimates under small sample sizes and separation (8). Odds ratios (OR) with 95% confidence intervals (CI) were calculated, using a reference category for categorical predictors.

Missing data were also a concern, with several variables containing a high proportion of missing values. Variables with > 50% missingness were excluded. For remaining variables, missing values were imputed using a random-forest-based imputation method. This method iteratively predicts missing values for each variable using all other variables as predictors, continuing until convergence. Accuracy of imputations was assessed using out-of-bag error estimates: normalized mean squared error for continuous variables and proportion falsely classified for categorical variables. Variables with both high missingness and poor imputation performance were discarded, whereas those with lower error rates were retained for analysis. Predictors that demonstrated significant associations (P < 0.05) in univariate models were considered for multivariate analysis.

RESULTS

Signalment and history

Thirty-five does treated for toxic mastitis met the inclusion criteria; 74% (26/35) of the animals survived to hospital discharge and the remaining 26% (9/35) died or were euthanized during hospitalization. Meat breeds represented 34% (12/35) of the study population, followed by dairy (26%; 9/35), unknown (26%; 9/35), and mixed (14%; 5/35) breeds. Of those discharged, 48% (10/21) were meat, 33% (7/21) were dairy, and 19% (4/21) were mixed breeds. Of those not discharged, 40% (2/5) were meat, 40% (2/5) were dairy, and 20% (1/5) were mixed breeds. Ages of animals in the study ranged from 1 to 11 y, with a median of 3 y. Body weights ranged from 16 to 93 kg (mean: 56 kg). Breed, age, and body weight were not different between survivors and non-survivors (P > 0.05). Goats were presented to the hospital between 8 h and 10 d after clinical signs appeared. The median time from noticing of clinical signs to admission to the hospital was not different (P = 0.6) between survivors and non-survivors (24 vs 36 h, respectively). Days post-partum at admission ranged from 1 to 730 d and median days post-partum at admission was not different (P = 0.3) between survivors and non-survivors (7 vs 4.5 d, respectively).

Physical examination and laboratory results

The HR, RR, and RT at admission ranged from 72 to 162 bpm, 20 to 112 rpm, and 36.6 to 40.9°C, respectively. Nineteen animals were presented with tachycardia (HR > 120 bpm), including 46.1% (12/26) of survivors and 77.7% (7/9) of non-survivors; 20 animals were presented with tachypnea (RR > 36 rpm), including 57.7% (15/26) of survivors and 55.5% (5/9) of non-survivors. The mean HR and RR at admission for survivors (HR: 120.4 bpm and RR: 46.7 rpm) and non-survivors (HR: 129 bpm and RR: 58.5 rpm) were not different (P = 0.1 and P = 0.4, respectively). In contrast, the mean RT at admission was lower (P = 0.01) in non-survivors (38.1°C) compared to survivors (39.5°C). Twelve animals were febrile (RT > 39.7°C) at admission, including 42.3% (11/26) of survivors and 11% (1/9) of non-survivors. Data from CBC and serum chemistry panel at admission were available for 57% (20/35) of animals (17/26 survivors and 3/9 non-survivors), and the results did not differ significantly between groups (Table 1). Mean fecal egg count was numerically greater (P = 0.3) in non-survivors compared to survivors (8400 vs 3010 eggs per gram, respectively).

TABLE 1.

Complete blood (cell) count and serum chemistry results compared between discharged and non-discharged goats treated for toxic mastitis.

Variable (unit; reference range)	Non-discharged groupa	Discharged groupa	P-value
AST (U/L; 167 to 513)	935 (583 to 974.5)	148 (109.75 to 315.75)	0.06
Glucose (mg/dL; 50 to 75)	43 (42 to 50)	66 (59.5 to 95.25)	0.07
Globulin (g/dL; 2.7 to 4.1)	3.39 (3.35 to 3.75)	4.24 (3.99 to 4.53)	0.08
TP (g/dL; 6.4 to 7.0)	6.26 ± 0.64b	6.65 ± 0.86b	0.34
Albumin (g/dL; 2.7 to 3.9)	2.3 (2.15 to 2.34)	2.43 (2.12 to 2.64)	0.42
SDH (U/L; 14 to 23.6)	15.9 (13.85 to 27.6)	13.9 (12.1 to 19.85)	0.72
GGT (U/L; 20 to 56)	43 (38 to 52.5)	39.5 (31.5 to 59)	0.82
BUN (mg/dL; 10 to 20)	36.3 (35.55 to 39.9)	35.05 (21.53 to 43.28)	0.67
Creatinine (mg/dL; 1.0 to 1.8)	1.4 (1.1 to 1.7)	1.35 (0.98 to 1.58)	0.92
Na (mg/dL; 142 to 155)	146 (145 to 147)	143 (140.25 to 145.75)	0.13
K (mmol/L; 3.5 to 6.7)	2.9 (2.7 to 3.7)	3.9 (3.5 to 4.6)	0.14
Cl (mg/dL; 99 to 110)	104 (100 to 104.5)	105 (101 to 107.75)	0.29
Ca (mg/dL; 8.9 to 11.7)	7.3 (6.8 to 8.6)	8.2 (8.1 to 8.8)	0.34
Bicarbonate (mmol/L; 20 to 25)	18.4 (14.3 to 26.7)	19.5 (18.15 to 20.88)	0.80
Anion gap (mmol/L)	26.5 (23 to 29.9)	23.95 (20.78 to 26.73)	0.43
PCV (%; 22 to 38)	28 (26 to 30)	31 (29 to 33)	0.55
Hematocrit (%; 22 to 38)	28.3 (26.8 to 31.15)	30.2 (26.3 to 33.1)	0.77
Hemoglobin (g/dL; 8 to 12)	9.8 (9.35 to 10.65)	10.2 (9.4 to 11.4)	0.63
WBC (×103/μL; 4 to 13)	6.16 (4.53 to 11.10)	6.63 (3.73 to 12.35)	0.89
Neutrophils (×103/μL; 1.2 to 7.2)	1.48 (1.42 to 6.83)	1.94 (1.09 to 9.15)	1.00
Bands (Yes/No)	1 (33.3%); 2 (66.7%)c	4 (23.5%); 13 (76.5%)c	1.00
BHB (mmol/L; 0 to 0.7)	0.1 (0.1 to 0.1)	0.2 (0.03 to 0.45)	1.00

AST — Aspartate aminotransferase; BHB — Beta-hydroxybutyrate; BUN — Blood urea nitrogen; Ca — Calcium; Cl — Chloride; GGT — Gamma-glutamyl transferase; K — Potassium; Na — Sodium; PCV — Packed cell volume; SDH — Sorbitol dehydrogenase; TP — Total protein; WBC — White blood cell count.

^aData presented as median (interquartile range), unless otherwise indicated.

^bData presented as mean ± standard deviation.

^cData presented as number of population (percentage).

Reference ranges obtained from published reference works (21,22).

A milk culture was done on 71% (25/35) of goats, including 21 survivors and 4 non-survivors. A single organism was isolated in 76% (19/25) of cultures, with the remaining 24% (6/25) of cultures growing multiple organisms. Staphylococcus aureus was the most frequently isolated pathogen, identified in 48% (12/25) of cases. Other isolates included Escherichia coli in 20% (5/25), Enterococcus spp. in 16% (4/25), and Bacillus cereus in 12% (3/25). Additional Staphylococcus species, e.g., S. intermedius and S. chromogenes, were recovered in 12% (3/25) of cases. Less commonly isolated organisms included Pseudomonas aeruginosa (8%; 2/25), Klebsiella pneumoniae (4%; 1/25), and Streptococcus agalactiae (4%; 1/25).

Patient management during hospitalization

The median hospitalization time was greater (P = 0.0001) for survivors compared to non-survivors (12 vs 2 d, respectively). Most cases (94%; 33/35) were treated with systemic antibiotics; however, the number of animals treated with systemic antibiotics was not different (P = 0.1) between survivors (24/26) and non-survivors (9/9). Systemic antibiotic therapy consisted of either individual use or combinations of the following antimicrobials: oxytetracycline, penicillin, florfenicol, ceftiofur, and tulathromycin. In contrast to findings for systemic antibiotic therapy, a greater proportion of survivors received intramammary antibiotic (IMA) therapy compared to non-survivors [77% (20/26) vs 33% (3/9), respectively; P = 0.04]. Intramammary antibiotic therapy consisted of the individual use of pirlimycin or ceftiofur in the affected mammary gland; however, type of IMA was not associated with outcome. Dexamethasone, nonsteroidal anti-inflammatory drugs, and intravenous fluids were administered in 34, 91, and 86% of cases, respectively. The proportion of animals receiving each treatment did not differ between survivors and non-survivors (P > 0.3).

Univariable analysis

Results from the univariate analysis are shown in Table 2 for continuous and categorical variables. The odds of hospital discharge increased (OR: 2.98; 95% CI: 1.44 to 7.97) for each 1°C increase in RT at admission (P = 0.002). The odds of hospital discharge increased (OR: 11.21; 95% CI: 1.31 to 231.54) for each 1 g/dL increase in serum globulin concentration at admission (P = 0.02). For each additional day of hospitalization, goats treated for toxic mastitis in our study were 2.35× more likely to be discharged from the hospital (OR: 2.35; 95% CI: 1.43 to 2.89; P < 0.001). Intramammary antibiotic therapy increased the likelihood of hospital discharge 5.86× (95% CI: 1.28 to 31.5) compared to no IMA therapy (P = 0.022). The odds of hospital discharge increased 4.2× (OR: 4.20; 95% CI: 0.77 to 43.56) in goats treated with dexamethasone; however, this association only approached statistical significance (P = 0.10). The odds of hospital discharge decreased by 89% in goats with no appetite at admission (OR: 0.11; 95% CI: 0.001 to 1.16); this association only approached statistical significance (P = 0.07). Other physical examination, diagnostic, and treatment variables were not significantly associated with the outcome (Table 2).

TABLE 2.

Univariable logistic regression analysis of patient signalment, physical examination, diagnostic, and treatment variables for prediction of survival to discharge from the hospital in goats with toxic mastitis.

Variable	Coefficient	Std error	OR	95% CI (OR)	P-value
Days of hospitalization	0.86	0.31	2.35	(1.43, 2.88)	< 0.01
Rectal temperature (°C)	1.09	0.41	2.98	(1.44, 7.97)	< 0.01
Globulin (g/dL)	2.42	1.18	11.21	(1.31, 231.54)	0.02
Intramammary antibiotics (Y/N)	1.77	0.80	5.86	(1.28, 31.50)	0.02
Appetite: Absent vs normal	−2.23	1.54	0.11	(< 0.01, 1.16)	0.07
Dexamethasone (Y/N)	1.44	0.97	4.20	(0.77, 43.56)	0.10
K (mmol/L)	0.91	0.67	2.49	(0.74, 12.53)	0.16
Ca (mg/dL)	0.72	0.58	2.05	(0.70, 8.62)	0.2
Appetite: Reduced vs normal	−1.71	1.59	0.18	(< 0.01, 2.32)	0.21
IV fluids (Y/N)	−1.58	1.53	0.21	(< 0.01, 2.13)	0.22
TP (g/dL)	0.68	0.59	1.98	(0.66, 7.58)	0.24
RR (respirations/min)	−0.02	0.02	0.98	(0.95, 1.01)	0.25
HR (beats/min)	−0.02	0.02	0.98	(0.93)	0.37
Breed: Meat vs dairy	0.43	0.83	1.53	(0.30, 8.33)	0.61
Days after parturition	< 0.01	< 0.01	1.00	(0.1, 1.58)	0.62
Flunixin (Y/N)	0.55	1.11	1.73	(0.14, 15.01)	0.63
Systemic antibiotics (Y/N)	−0.66	1.60	0.52	(< 0.01, 7.15)	0.66
Age (y)	0.05	0.17	1.05	(0.78, 1.56)	0.75
Hemoglobin (g/dL)	0.06	0.18	1.06	(0.75, 1.65)	0.76
Albumin (g/dL)	0.22	0.74	1.25	(0.21, 5.69)	0.78
Breed: Mixed vs dairy	−0.26	0.97	0.77	(0.12, 5.79)	0.79
Creatinine (mg/dL)	0.03	0.50	1.03	(0.40, 3.59)	0.96

Ca — Calcium; CI — Confidence interval; HR — Heart rate; IV — Intravenous; K — Potassium; OR — Odds ratio; RR — Respiratory rate; TP — Total protein.

Multivariate analysis

Days of hospitalization, RT at admission, serum globulin concentration, treatment with IMA (yes/no), and appetite at admission (yes/no) were retained in the final multivariate model. When holding all other variables constant, the odds of hospital discharge increased 23× for each additional 1 g/dL increase in serum globulin concentration at admission (OR: 22.99; 95% CI: 1.59 to 942.97; P = 0.01), and the goats treated with IMA were 10× more likely to be discharged from the hospital compared to goats not treated with IMA (OR: 10.04; 95% CI: 1.67 to 99.85; P = 0.01).

DISCUSSION

Mastitis is a relatively common disease in dairy-producing animals and is increasingly recognized in the growing companion goat industry. Although many cases are subclinical, clinical mastitis presentation may vary from mild forms characterized by abnormal milk to severe cases involving systemic compromise and marked inflammation of the mammary gland. Severe clinical mastitis, often referred to as “acute toxic mastitis,” is distinguished by profound dehydration, septicemia, endotoxemia, altered mentation, anorexia, and clinical signs consistent with shock. Affected animals may deteriorate rapidly and face a high risk of mortality without timely and appropriate intervention (5). Consequently, evidence-based guidelines for the effective management of these cases are critically needed.

To the authors’ knowledge, this is the first retrospective study evaluating prognostic indicators for survival in goats treated for toxic mastitis at a referral veterinary hospital. Of 35 goats treated for toxic mastitis at this institution, the overall rate of survival to hospital discharge was 74%. This survival rate seemed greater than the 60% survival rate reported in 2800 cases of gangrenous mastitis in goats from a previous study (6), perhaps due to advancements in diagnosis and treatment over the past 60 y. Severe inflammation of affected mammary glands, particularly in cases of gangrenous mastitis, can lead to irreversible damage to glandular tissue and permanent loss of milk production. In dairy production systems, this often results in humane euthanasia or culling of affected goats without treatment (5). In this study, most goats were companion animals, which likely influenced the willingness of owners to pursue additional treatment and try to save their pets’ lives, regardless of prognosis for production.

Univariate and multivariable regression analyses yielded inconsistent results across most variables. Such discrepancies are common in retrospective evaluations of medical records, which often contain variable data quality and missing information. To address this, multiple models were explored to identify the best fit, accounting for potential interactions among variables. The severity and duration of toxic mastitis often lead to pronounced systemic inflammation, resulting in clinical signs such as fever, lethargy, inappetence, tachycardia, and tachypnea. These signs typically result from cytokine release, dehydration, and pain. In this study, however, only RT at admission and inappetence were significantly associated with hospital discharge or approached statistical significance, respectively.

A higher RT at admission was associated with increased odds of survival. Transient pyrexia is common in the early stages of toxic mastitis (9), and its association with survival in our study may have reflected admission during the initial phase of the acute inflammatory response, early in the stages of the disease, when systemic compromise was less severe. In contrast, animals admitted in more advanced stages, characterized by signs of shock including severe dehydration, hypovolemia, and lower RT, may have had reduced ability to respond to treatment.

Results from a previous study demonstrated increased pro-inflammatory cytokines, e.g., interleukin-6, interleukin-1b, and tumor necrosis factor-α, in does affected by toxic mastitis (10). A common sequela of systemic cytokine release is reduced appetite, as observed in several goats in this study. As documented in other veterinary species (11,12), inappetence can be a prognostic indicator of poorer outcomes in systemically compromised animals.

Clinical compromise due to inflammation, toxin production, and dehydration, as well as electrolyte and metabolic imbalances occurring in cases of toxic mastitis, warrants aggressive therapy. Frequently, intravenous fluids, anti-inflammatories, analgesics, and systemic antibiotics and/or IMA are used to improve the likelihood of patient survival. In this study, only IMA therapy was significantly associated with hospital discharge; however, administration of systemic steroids appeared to also improve the likelihood of hospital discharge and approached statistical significance. The severity of clinical compromise at the time treatment was initiated may have had a greater influence on patient outcomes than the specific therapeutic interventions used for managing toxic mastitis.

In addition to clinical signs of systemic disease, bacterial infection of the mammary gland initiates either a primary or memory immune response, leading to production of specific immunoglobulins (antibodies) and cell-mediated immunity. Results from previous studies demonstrated lower serum total protein and globulin concentrations in goats affected with toxic mastitis compared with healthy goats (13,14). Higher serum globulin concentrations observed at admission for survivors in this study may have reflected a stronger and faster inflammatory and antibody response in affected animals. This enhanced immune activity could have contributed to more effective neutralization of bacteria and their toxins, thereby increasing the likelihood of hospital discharge. Similarly, results from previous studies demonstrated that elevated levels of serum antibody against Escherichia coli toxins, including lipopolysaccharides, reduced clinical severity and culling in dairy cows with toxic mastitis (15,16).

Treatment of cases of toxic mastitis with IMA is controversial because the overwhelming inflammatory response and neutrophil infiltration into the affected mammary gland usually eliminate the infection without the need for antimicrobials (17). However, in the present study, treatment with IMA increased survival to discharge in goats with toxic mastitis. This was consistent with previous studies in which the use of IMA in addition to systemic antibiotics resulted in successful treatment of gangrenous mastitis in affected does (9,18). Intramammary antibiotic therapy likely reduced proliferation and translocation of bacteria and their toxins from the mammary gland into the circulation, decreasing the risk of septicemia, systemic compromise, and mortality. The use of IMA for treatment of mastitis in goats is extra label in the United States, requiring longer milk and meat withdrawal times (19). Antimicrobial therapy is recommended for animals affected by toxic mastitis and should, in most cases, begin before the causative agent is identified. Findings from this study suggest that combining intramammary with systemic antibiotic therapies may enhance the likelihood of survival in goats with toxic mastitis.

The odds of survival to hospital discharge in goats treated for toxic mastitis in our study increased for each additional day of hospitalization, and hospitalization time was significantly greater in survivors compared to non-survivors. Similar results were previously reported in a study evaluating prognostic indicators for survival in downer beef cows (20). Factors such as initial response to treatment could have influenced this finding. Animals with toxic mastitis admitted during the early stages of disease, with only mild systemic compromise, were more likely to respond favorably to initial treatment. This positive response likely increased the willingness of veterinarians and owners to continue with therapy and hospitalization. In contrast, animals admitted during more advanced stages of disease, with greater systemic compromise, were more likely to deteriorate rapidly, resulting in earlier death or euthanasia.

The CBC and serum biochemistry panels are valuable diagnostic tools that guide clinicians in treating systemically compromised patients. In this study, there was a trend of greater serum aspartate aminotransferase and lower glucose concentrations in non-survivors at hospital admission; however, other differences were not statistically significant. The relatively small sample size, along with numerous missing data points, may have limited the ability to detect significant associations. Nevertheless, clinicopathologic abnormalities identified through CBC and serum biochemistry remain essential for guiding treatment decisions and establishing prognosis in severely ill animals, such as those with toxic mastitis.

Staphylococcus aureus was the most common isolate identified in goats with toxic mastitis in this study. This finding was consistent with previous reports (9,14); however, the high proportion of missing milk culture data may have influenced this result. Beyond its role as the primary cause of toxic mastitis, S. aureus is also well recognized for causing subclinical and chronic mastitis in goats, owing to its ability to persist within macrophages and resist antimicrobial therapy (5). Its importance as a causal agent of chronic, treatment-refractory mastitis, coupled with its zoonotic potential, underscores its clinical and production relevance and may influence treatment decisions in dairy production settings (4).

Escherichia coli, Enterococcus spp., and Bacillus cereus are other environmental agents of importance in the pathogenesis of toxic mastitis in cattle and goats (10,15). In a previous report, 6/6 goats affected with toxic mastitis caused by B. cereus survived to hospital discharge after treatment (18). The higher survival rate reported in that study may have reflected its smaller sample size and differences in severity of systemic compromise at admission.

Intestinal parasitism, which is highly prevalent in small ruminant populations in the southeastern United States, may impair the host’s ability to mount effective immune responses against other infectious diseases, such as toxic mastitis in the early postpartum period. In this study, mean parasite fecal egg counts at admission were higher in non-survivors, though the difference was not statistically significant. Clinically, however, a heavier parasite burden could have compromised the capacity of these animals to respond to treatment, thereby reducing the likelihood of hospital discharge.

Limitations of this study included the retrospective design and the inherent incompleteness of medical records, which resulted in missing data points for several variables. Such gaps can reduce statistical power, restrict the ability to detect significant associations, and introduce potential bias if missing information is not random. Consequently, these factors may have influenced robustness of the analyses and should be considered when interpreting the results of this study. Further investigation into prognostic indicators for survival in goats with toxic mastitis is warranted to strengthen the validity of our findings. Larger sample sizes, achieved through multi-institutional and cross-country retrospective studies, would reduce geographical and case-selection biases.