Efficacy of Hypochlorite as a Disinfestant Against Fungal Pathogens in Agricultural and Horticultural Plant Production: A Systematic Review and Meta-Analysis

Hypochlorite is often used as a disinfestant of fungal pathogens in a range of agricultural and horticultural settings. However, reports of its effectiveness are variable across studies and it is unclear what factors could potentially influence the reported estimates of its efficacy. A systematic review and meta-analysis was conducted to assess the efficacy of hypochlorite against fungal pathogens and explore factors that may explain the observed heterogeneity in estimates of efficacy. Standardized mean effect size, Hedges’ g, was calculated for each of the 109 selected studies, published from 1972 to 2019, that met the criteria defined for the systematic review. A random-effects model was used to estimate the overall mean effect size (g¯+) and determine the heterogeneity in g among studies. Hypochlorite resulted in a significant (P < 0.001) reduction in either disease intensity or propagule viability with g¯+= 2.25, suggesting a large overall effect. However, 95% prediction intervals ranged from −0.18 to 4.68, indicating that hypochlorite could be ineffective against some fungi or when targeting some substrate materials. An estimate of the within-study variability, τ², was 1.48 and the proportion of heterogeneity in g among studies due to true effects was 71.5%. Inclusion of categorical moderator variables in the random effects model showed that hypochlorite treatments were significantly (P < 0.0062) more effective when used to disinfest spores in an aqueous solution (g¯+= 4.58) than when used on plastic (g¯+= 2.13), plant (g¯+= 2.13), and wood (g¯+= 0.79). Similarly, hypochlorite treatments were significantly (P < 0.0083) more effective in disinfesting fungal propagules of Thielaviopsis spp. (g¯+= 2.51) than those of Verticillium spp. (g¯+= 1.21). A meta-regression indicated that the effect of dose (β = −3.54; P = 0.0398) and contact time (β = −0.05; P = 0.0001) on g¯+ were highly significant. Further, g¯+was significantly affected by the dose × time interaction (β = −0.017; P = 0.0269). In the meta-regression models, dose and time explained 0 and 16% of the variance in true effects, respectively. In meta-regression models with a continuous variable of dose or time, a categorical variable of target or genus and their interaction term, genus and target explained an additional 7 to 19% of the variance in true effects. These results show that although the current recommended dose and contact time for commercial bleach products are expected to result in effective disinfestation, the target material and genera of the fungal pathogen of interest will likely influence their efficacy.