Vetnuus | November 2025 21 Article number of cases of UO. Two independent reviewers manually screened records to identify male cats presenting clinical signs consistent with FLUTD, including ischuria, hematuria, stranguria, pollakiuria, periuria, or dysuria. From this initial cohort, only cases diagnosed with UO were selected, regardless of the underlying aetiology. Exclusion criteria included absence of LUT signs, female sex, diagnosis of non-obstructive FLUTD, comorbidities of infectious, parasitic, fungal, or traumatic nature, and duplicate records. A total of 386 male cats with UO met all inclusion criteria and were retained for final analysis. Among the selected animals, data on sex, age, breed, body weight, reproductive status (neutered or intact), and clinical signs presented in the emergency room were noted. Statistical analysis Data distribution was analysed using the Shapiro-Wilk test. Parametric analyses were performed on the variables body weight and number of UO cases. Comparison of body weight between castrated and intact animals was performed using the unpaired t-test. Data are shown as the mean ± standard deviation (SD). Significant differences were considered when p < 0.05. The nonparametric Mann-Whitney test was applied to compare the BCS values between castrated and intact animals; the results are shown as median (Q1–Q3). Significant differences were considered when p < 0.05. Comparison between the cause of UO, BCS, and body weight was also performed. Diet and breeds were considered for descriptive analysis. The obstruction rate, calculated as the ratio between the number of UO cases per year and the number of cats per year, was analysed using ARIMA (autoregressive integrated moving average) time series models. ARIMA models are effective for modelling stationary data; they capture temporal dependencies through autoregressive (AR), differencing (I), and moving average (MA) components. Model selection and parameter estimation (p, d, q) were guided by autocorrelation function (ACF) and partial autocorrelation function (PACF) plots. Likewise, the Akaike Information Criterion (AIC) was used to optimise the model, with parameters set to h = 24 h and a 95% confidence level. This modelling strategy allows for the identification of autocorrelated structures and seasonal fluctuations. The finalised ARIMA model was then used to project UO cases in cats for the years 2021, 2022, 2023, and 2024. All analyses were conducted using the “stats,” “tseries,” and “forecast” packages in R (version 4.1.3). The Holt-Winters exponential smoothing model was applied to capture trend and seasonality in the time series data. This model assigns greater weight to more recent observations while exponentially decreasing the influence of older values9. It decomposes the data into three components: a baseline value (representing the overall mean of the period), a trend, and seasonal components, which improves forecast accuracy, especially in cases of non-stationary time series. Based on these patterns, the model provides an exponential forecast for the trend in the next cycle. Likewise, projections for the period 2021–2024 were employed. Finally, a simple linear regression model was developed with year as the dependent variable and case rate as the independent variable to determine the regression equation. A scatter plot with the fitted regression line was created for visualisation. The regression model is represented as follows18. Yi represents the response variable, B is the slope coefficient indicating the average change in Y per one-unit increase in X, A is the intercept (the value of Y when X = 0 e Ei), and Ei denotes the error term for each predicted value. To further illustrate these analyses, boxplots and time series plots were generated. A 5% significance level was applied, and all analyses were conducted using R software (version 4.1.0; R Core Team, 2021). Results Study population data In the survey of cases, 5,230 cats were seen between January 2010 to December 2020 at the Veterinary Teaching Hospital of the UNESP in Botucatu, São Paulo. Of the 5,230 records of cats found in the hospital database, 4,158 (79.5%) were excluded due to not presenting signs of the LUT. Furthermore, among the remaining cats, 295 (5.64%) were females and were subsequently excluded from the study. Among the remaining male cats, 198 (3.7%) were diagnosed with nonobstructive FLUTD. Out of the male cats with UO, 92 (1.75%) presented with other comorbidities, which included infectious, parasitic, fungal, or traumatic conditions (car accident, fight with other animals). Additionally, 101 (1.9%) cats with duplicate records were excluded from the analysis (Fig. 1). Following the application of the exclusion criteria, 386 male cats (7.4%; 95% CI: 6.7–8.1%) were retained for analysis and diagnosed with UO (Fig. 2). The mean body weight of these cats with UO was 4.24 ± 1.11 kg, and the mean age was 98.40 ± 53.94 months. Of the cats with information on reproductive status, 60.94% were castrated (181/297) and 39.06% were intact (116/297). In our study, castrated male cats exhibited a greater body weight (4.66 kg ± 1.13 kg) compared to intact male cats (3.77 kg ± 1.11 kg) (p < 0.001). Additionally, when considering reproductive status and BCS, castrated male cats had higher BCS compared to intact cats, 6 (5–7) vs. 5 (4–5.75) (p < 0.0001). In the population of cats with UO of our study, the distribution frequency according to the breed is presented in Fig. 3a. Notably, the most common type of diet provided by owners, primarily dry food, is depicted in Fig. 3b. In investigating the causes of UO among the 386 cats, an underlying cause was identified in 15.3% (n = 59) of cases. These underlying causes were found either in isolation or in combination within the same animal. FIC appears to be the most common cause of UO in the study population. Incomplete medical records prevented the determination of the underlying causes of UO in 327 cats (Fig. 3c). LUT and systemic clinical signs, including depression, anorexia or hyporexia, polydipsia, hypothermia, and erection, were evaluated (Supplementary Fig. S1). In our study, each cat (n = 386) presented at least two clinical signs upon emergency care. The frequency of clinical signs categorised by the cause of UO is presented in Table 1. Prevalence and incidence The calculated mean annual prevalence of UO considering the total cat population (n = 5,230) was 7.4% (95% CI: 6.7–8.1%). Among the total cat population, 1,072 showed LUT signs. Yi = A+BXi+Ei >>>22
RkJQdWJsaXNoZXIy OTc5MDU=