Evaluación de la magnitud y la calidad de la evidencia de los estudios secundarios relacionados con la psoriasisutilidad de las revisiones sistemáticas, los meta-análisis en red y los estudios meta-epidemiológicos

Abstract

Background. Psoriasis is a highly prevalent chronic inflammatory skin disease whose moderate and severe forms are associated with a reduction of quality of life and a greater comorbidity. In relation to the classic treatments, biologics show high efficacy and better safety profile. However, they are not risk-free and their high costs threat to sustain the treatment of these patients within the public health systems. In this context, it is especially important that decisions taken by clinicians and health managers will be supported on the best evidence. Systematic reviews and a meta-analysis constitute the standard documents for summarizing scientific evidence. Methodological instruments and protocols used in the development, notification, and quality control of these documents have allowed to reduce the degree of uncertainty of drug effect estimations, thus improving the efficiency of the derived recommendations. However, it has been observed in recent years an exponential growth of systematic reviews of dubious scientific quality. By assessing the methodological quality and the risk of bias of these publications, we can select only those of the highest scientific quality. Objectives. In the present research project we proposed a series of primary and secondary objectives that were achieved through two different strategies. First, we carried out a systematic review and a network meta-analysis on the short-term effect of biologics for the treatment of moderate-severe plaque psoriasis in adults. Second, we assessed the methodological quality and the risk of bias of systematic reviews and meta-analyses published on psoriasis, by incorporating data and metadata related to articles, journals and authors, in order to develop a predictive model that can be helpful in decision-making processes. Methodology. We followed the Cochrane Handbook for Systematic Reviews on interventions and the PRISMA (Preferred reporting items for Systematic reviews and Meta- Analyses) recommendations to for systematic reviews conduction and notification respectively. The evaluation of the methodological quality and the risk of bias of the systematic reviews were performed with AMSTAR (A Measurement Tool to Assess the Methodological quality of Systematic reviews) and ROBIS (Risk of Bias In (Systematic reviews) tools, respectively. For metaepidemiological analyses different analytic approach were followed: multinomial logistic regresion models, principal component analysis, clustering analysis. Results. The results derived have been included in three original articles published in high-impact scientific journals (all ranked in the 10th percentile). In summary, after publishing a priori protocol in PROSPERO, it was carried out a systematic review and network meta-analysis, including 27 randomized-controlled clinical trials. Infliximab 5 mg/Kg every 8 weeks and secukinumab 300 mg every 4 weeks were the most effective treatments in week 10-16 accounting for PASI 75 and PASI 90 results respectively. In addition, both treatments presented the highest risk of having at least one adverse effect (42.1 %) or at least one infection (52.2 %), respectively. No differences were found in severe adverse events among the treatments. Ustekinumab 90 mg Every 12 weeks was the treatment with the best risk vs benefit profile. The quality of the evidence for efficacy estimates was high in the case of ustekinumab and moderate or low for the rest drugs. For the safety, quality of the evidence was, in general, low or very low. With respect to the metaepidemiological studies related to the evaluation and prediction of the methodological quality, 220 systematic reviews published by 741 authors of 520 different institutions from 32 countries were included. Only 17% of the reviews presented high methodological quality according to AMSTAR tool. The inclusion of a metan alisys (OR 6.22, (IC95% 2.78-14.86), funding by academid institutions (OR 2.90, IC95% 1.11-9.19), low number of authors with conflict of interests (OR 0.90, IC95% 0.82-0.99), a high journal impact factor (OR 2.14, IC95% 1.05-6.67), a high 5 years journal impact factor (OR 1.34, IC95% 1.02-1.40), and a high number of pages (OR 1.08, IC95% 1.02-1.15) were predictors of high methodological quality. In reference to the analysis carried out to compare AMSTAR and ROBIS, we evaluated 139 systematic reviews on interventions in psoriasis, of which only 22.3% presented a high methodological quality and 14 % a low risk of bias. The percentage of systematic reviews with high methodological quality and with high risk of bias was greater than 50%. The components that best explained the variability of the results of AMSTAR and ROBIS are related with the evaluation of the primary studies of systematic reviews.