An assessment of the impacts of litter treatments on the litter quality and broiler performance: A systematic review and meta-analysis

Abstract

Objective: The choice of the most suitable litter treatment should be based on scientific evidence. This systematic review assessed the effectiveness of litter treatments on ammonia concentration, pH, moisture and pathogenic microbiota of the litter and their effects on body weight, feed intake, feed conversion and mortality of broilers.

Methods: The systematic literature search was conducted using PubMed (Medline), Google Scholar, ScienceDirect and Scielo databases to retrieve articles published from January 1998 to august 2019. Means, standard deviations and sample sizes were extracted from each study. The response variables were analyzed using the mean difference (MD) or standardized mean difference (SMD), (litter treatment minus control group). All variables were analyzed using random effects meta-analyses.

Results: Subgroup meta-analysis revealed that acidifiers reduce pH (P<0.001), moisture (P = 0.002) ammonia (P = 0.011) and pathogenic microbiota (P <0.001) of the litter and improves the weight gain (P = 0.019) and decreases the mortality rate of broilers (P<0.001) when compared with controls. Gypsum had a positive effect on ammonia reduction (P = 0.012) and improved feed conversion (P = 0.023). Alkalizing agents raise the pH (P = 0.035), worsen feed conversion (P<0.001), increase the mortality rate (P <0.001), decrease the moisture content (P<0.001) and reduce the pathogenic microbiota of the litter (P<0.001) once compared to controls. Superphosphate and adsorbents reduce, respectively, pH (P<0.001) and moisture (P = 0.007) of the litter compared to control groups.

Conclusion: None of the litter treatments influenced the feed intake of broilers. Meta-analyses of the selected studies showed positive and significant effects of the litter treatments on broiler performance and litter quality when compared with controls. Alkalizing was associated with worse feed conversion and high mortality of broilers.

Fig 1. Diagram adapted from the PRISMA-P guidelines [24], showing the total number of reports identified and the number of reports filtered at each stage of the study selection process from the systematic review on poultry litter treatments.

Fig 2. Forest plot of the effect size or standardized mean difference and 95% confidence interval of the effect of litter treatments on feed intake of broilers.

The solid vertical grey line represents a mean difference of zero, or no effect. Points to the left of the solid vertical line represent a reduction in feed intake, while points to the right of the line indicate an increase in feed intake in broilers reared on treated litter.

Fig 3. Forest plot of the effect size or standardized mean difference and 95% confidence interval of the effect of litter treatment on weight gain of broilers.

The solid vertical grey line represents a mean difference of zero, or no effect. Points to the left of the solid vertical line represent a reduction in weight gain, while points to the right of the line indicate an increase in the weight gain of broilers reared on treated litter.

Fig 4. Forest plot of the effect size or mean difference and 95% confidence interval of the effect of litter treatment on feed conversion of broilers.

The solid vertical grey line represents a mean difference of zero, or no effect. Points to the left of the solid vertical line represent an improvement in feed conversion, while points to the right of the line indicates a worsening in feed conversion of broilers reared on treated litter.

Fig 5. Forest plot of the effect size or mean difference and 95% confidence interval of litter treatment on mortality rate in broilers.

The solid vertical grey line represents a mean difference of zero, or no effect. Points to the left of the line represent a reduction in mortality, while points to the right of the line indicate an increase in mortality in broilers reared on treated litter.

Fig 6. Forest plot of the effect size or standardized mean difference and 95% confidence interval of litter treatment on ammonia concentration and volatization.

The solid vertical grey line represents a mean difference of zero, or no effect. Points to the left of the line represent a reduction in ammonia, while points to the right of the line indicate an increase in ammonia on treated litter.

Fig 7. Forest plot of the effect size or mean difference and 95% confidence interval of litter treatment on pH.

The solid vertical grey line represents a mean difference of zero, or no effect. Points to the left of the line represent a reduction in pH, while points to the right of the line indicate an increase in pH on treated litter.

Fig 8. Forest plot of the effect size or mean difference and 95% confidence interval of litter treatment on moisture.

The solid vertical grey line represents a mean difference of zero, or no effect. Points to the left of the line represent a reduction in moisture, while points to the right of the line indicate an increase in moisture on treated litter.

Fig 9. Forest plot of the effect size or standardized mean difference and 95% confidence interval of litter treatment on pathogenic microbiota.

The solid vertical grey line represents a mean difference of zero, or no effect. Points to the left of the line represent a reduction in pathogenic microbiota, while points to the right of the line indicate an increase in pathogenic microbiota on treated litter.

Fig 10. Publication bias analysis, funnel plot.

Egger's Linear regression test of funnel plot asymmetry: A) Feed intake: bias±s.e. = 0.43 ± 0.68; slope = -0.27 (t = 0.63, p-value = 0.53). B) Weight gain: bias±s.e. = 0.29 ± 0.82; slope = 0.16 (t = 0.36, p-value = 0.72). C) Feed conversion: bias±s.e. = -0.06 ± 0.54; slope = -0.01 (t = -0.11, p-value = 0.91). D) Mortality rate: bias±s.e. = 0.36 ± 1.0; slope = -0.52 (t = 0.36, p-value = 0.72). E) Amonnia: bias±s.e. = -2.26 ± 0.6; slope = 1.27 (t = -3.76, p-value < 0.001). F) pH: bias±s.e. = 0.09 ± 0.81; slope = -0.32 (t = 0.12, p-value = 0.91). G) Moisture: bias±s.e. = -1.04 ± 1.22; slope = -1.27 (t = -0.86, p-value = 0.39). H) Pathogenic microbiota: bias±s.e. = -2.50 ± 0.65; slope = 0.92 (t = -3.85, p-value = 0.001).