Results

Study Population
Fifty-six cats were screened to participate in the study. Of the 56 cats, 4 did not meet the initial pre-diagnostic inclusion criteria. On day 0, 52 cats were randomized to 1 of the 2 study diets, of which 17 were found to either violate the inclusion or meet the exclusion criteria based on the results of the diagnostic screening. Seven cats dropped out of the study prior to the day 28 visit. Reasons for dropping out included unwillingness to eat the diet in four cats (two cats fed Diet A and two cats fed Diet B), withdrawn by owner due to continued diarrhea in two cats (one fed Diet A for 1 day and one fed Diet B for 4 days), and one owner declined to participate following screening for one cat. The final study population of 28 cats included 13 cats fed Diet A and 15 cats fed Diet B. One of the cats in the Diet A group was inadvertently given the wrong diet after 2 weeks; therefore, data for this cat were included for only the first 2 weeks of the feeding period.

Demographic and Baseline Characteristics
Mean age was 8.0 years (SD = 3.9, range 2–15 years) and 6.5 years (SD = 3.6, 1.0–14 years) in cats fed Diets A and B, respectively. Spayed females represented 62% (n = 6) and 40% (n = 6), while neutered males represented 38% (n = 5) and 60% (n = 9) of cats fed Diets A and B, respectively. Mean body weight was 4.9 kg (SD = 1.3, range 3.0–7.3) and 4.6 kg (SD = 1.2, range 2.9–7.0); and mean BCS was 3.2 (SD = 0.6, range 2–4) and 3.1 (SD = 0.6, range 2–4) on a 5-point scale in cats fed Diets A and B, respectively. Results of the PCR fecal panel showed that 16 of the 28 cats (Diet A n = 9; Diet B n = 7) tested positive for C. perfringens enterotoxin gene; 1 cat (Diet A n = 1; Diet B n = 0) tested positive for C. coli; and 17 cats (Diet A n = 9; Diet B n = 8) tested positive for FCoV. Eighteen of the 28 cats (Diet A n = 10; Diet B n = 8) tested positive for FCoV antibody. There were no statistically differences in the baseline characteristics of the cats between diet groups, and diagnostic screenings showed no clinically meaningful differences.

Comparison of Dietary Interventions
Cats fed both Diets A and B had a numerical reduction in the mean weekly number of vomiting episodes, with little to no change in weekly diarrhea episodes, over the 4-week period (Table 2). There were no differences between diet groups in the mean weekly number of vomiting or diarrhea episodes (p > 0.05). Significant differences were seen in vomiting episodes over time within cats fed Diet A between week 0 and weeks 2–4. These had fold changes in the number of vomiting occurrences of 3.24, 3.74, and 2.72 (indicating reduction of 69.1, 73.3, and 63.2%), respectively, with p values of 0.008, 0.003, and 0.029, respectively (Table 3). Changes in vomiting episodes within cats fed Diet B did not reach significance over the 4-week study period (p > 0.05) (Table 3). There were no statistical differences in the change of diarrhea episodes over the 4-week study (p > 0.05).
Table 2  Estimated means and 95% confidence intervals (CIs) for weekly vomiting and diarrhea episodes in 28 cats with chronic gastroenteropathy.
Table 3  Estimated fold changes, 95% confidence intervals (CIs), and p values for significance of change in vomiting over the 4-week study period in 28 cats with chronic gastroenteropathy.  Significant p-values are listed in bold text.  The vomiting probability also numerically declined in both groups of cats over the 4-week study. At weeks 0–4, cats fed Diet A had a probability of vomiting of 0.98 [95% confidence interval (CI) 0.42–1.00], 0.80 (95% CI 0.19–0.99), 0.3 (95% CI 0.03–0.87), 0.20 (95% CI 0.01–0.82), 0.20 (95% CI 0.01–0.82), respectively. At weeks 0, 1, 2, 3, and 4 cats fed Diet B had a probability of vomiting of 0.96 (95% CI 0.45–1.00), 0.86 (95% CI 0.27–0.99), 0.92 (95% CI 0.35–1.0), 0.78 (95% CI 0.20–0.98), and 0.57 (0.10–0.94), respectively. Statistical differences were seen between time points within cats fed Diet A between week 0 and weeks 2–4, with odds ratios of 0.008, 0.005, and 0.005 respectively (p values of 0.038, 0.023, and 0.023) (Table 4). The change in probability of vomiting within cats fed Diet B did not reach significance over the 4-week study period (p > 0.05) (Table 4).
Table 4  Estimated fold changes, 95% confidence intervals (CIs), and p values for significance of all contrasts for vomiting probability.  Significant p-values are listed in bold text.  Fecal Consistency Scale scores were available for only 4 (31%) cats fed Diet A and 10 (67%) cats fed Diet B. The model of FCS against day and diet over the 4 weeks showed no significant differences between the diets in the change in fecal score (p > 0.05). There was no correlation between the frequency of vomiting or diarrhea over the 4-week period and the presence of fecal enteropathogens detected via PCR or antibody testing (p > 0.05). Similarly, neither Spec fPL results nor age of the cat correlated with the frequency of vomiting over the 4-week period (p > 0.05).