Endocrine diseases in Brazilian small-animal hospitals and referral caseloads: A systematic review and meta-analysis of proportions.

Endocrinopathies are common chronic disorders in small-animal practice, and accurate estimation depends on standardized diagnostic criteria and denominator-defined populations. Brazilian epidemiological evidence is mostly drawn from hospital and referral settings, resulting in fragmented coverage, limited generalizability, and marked heterogeneity between studies. Conversely, VetCompass in the United Kingdom integrates standardized primary-care electronic records and supports denominator-based population inference. Searches were updated in January 2026 and covered PubMed, Web of Science, Scopus, VETINDEX, LILACS (via BVS-Vet), SciELO, and gray literature (2005-2025). A total of 2,501 records were identified, of which 13 met the eligibility criteria. Two reviewers independently screened the studies, extracted the data, and appraised the methodological quality using the JBI checklist. Random-effects meta-analyses of proportions were undertaken where appropriate, with heterogeneity quantified using I². When quantitative synthesis was not appropriate, results were synthesized narratively. In Brazilian dogs, reported frequencies varied widely across studies and were strongly setting-dependent, consistent with referral and selection effects. Estimates across outcomes ranged from low frequencies in broad hospital caseloads to substantially higher proportions in specialized endocrinology referral cohorts (e.g., diabetes mellitus, hypothyroidism, and hypercortisolism). Hyperparathyroidism was consistently rare (0.01%-0.07%). Hypoadrenocorticism was uncommon (0.09% in a multicenter cohort). In cats, the frequency of hyperthyroidism also varied substantially by setting. Evidence on diabetes mellitus was limited to a single endocrinology referral cohort, reflecting a selected population. Therefore, Brazilian results are reported as study-specific frequencies. Where pooled, these estimates should be interpreted strictly as setting-specific pooled proportions from Brazilian hospitals and referral caseloads, not as population prevalence. VetCompass estimates are used as a primary-care benchmark to contextualize Brazilian hospital-based frequencies and illustrate setting effects (primary versus tertiary care) and not to infer true between-country differences in population prevalence. As a primary-care benchmark, UK studies report canine hypercortisolism at 0.28% (95% confidence intervals: 0.25-0.31), hypothyroidism at 0.23%, and diabetes mellitus at ~0.34%. The prevalence of feline hyperthyroidism is 2.4% overall and 8.7% among cats aged ≥10 years. Overall, the Brazilian evidence base largely reflects tertiary-care caseloads, and most studies were at high risk of bias. This reinforces the need for multicenter primary-care surveillance in Brazil supported by harmonized diagnostic criteria and interoperable electronic records.