Revisão sistemática e meta-análise de insulinas análogas e avaliação da efetividade e custos da insulinoterapia em diabéticos tipo 1 no estado do Paraná

Abstract

Resumo: Diabetes mellitus tipo 1 (DM tipo 1) é uma disfunção metabólica resultante da destruição das células beta pancreáticas tendo como consequência a deficiência de insulina, representando de 5% - 10% dos casos de diabetes. O objetivo deste trabalho foi avaliar a efetividade do uso de insulinas análogas de longa duração no controle glicêmico de pacientes com diabetes mellitus tipo 1 no Estado do Paraná. Foi realizada uma revisão sistemática e meta-análise de insulinas análogas com pacientes diabéticos tipo 1, por meio da metodologia proposta pala Cochrane e um estudo de coorte retrospectiva com pacientes pertencentes a 2ª Regional de Saúde (Curitiba e região metropolitana) que fazem parte do Programa de Otimização do Diabetes Tipo 1 do Estado do Paraná. Os custos diretos do tratamento com insulinas análogas foram avaliados com base nos dados de compra dos medicamentos e insumos, pela CEMEPAR, por meio de processos licitatórios e de valores pagos pelo SUS para os exames laboratoriais de acompanhamento dos pacientes com DM tipo 1, disponibilizados na base de dados SIGTAP. Após busca sistemática foram encontrados 1238 artigos de insulinas análogas de ação lenta e 2823 artigos de insulinas de ação rápida. A avaliação revelou que os artigos incluídos são de qualidade pobre a moderada. Para o desfecho de eficácia a insulina análoga glargina não apresentou resultados estatisticamente significativos que a favoreçam quando comparadas a insulina NPH, enquanto a insulina análoga detemir na posologia duas vezes ao dia apresentou resultados significativos em seu favor (peso da média da diferença -0.14, 95% IC [-0.21, -0.01]). Não existem diferenças entre as duas insulinas análogas (glargina e detemir) com relação eficácia. Os resultados de segurança e tolerabilidade mostraram-se inconsistentes. O tratamento com insulina aspart é mais efetivo quando comparado com insulina regular MD -0,13, IC95% (- 0,17;-0,09). Não foram encontradas diferenças significativas entre insulina lispro e insulina glulisina quando comparadas a insulina regular para o desfecho de eficácia. Nas comparações, direta e indireta, de insulinas análogas de ação rápida observase que existe diferença estatísticamente significativa no desfecho eficácia, entre a insulina regular e a aspart [MD -0,15 IC95%(-0,24;- 0,06)]. A análise conjunta de comparações diretas e indiretas de insulinas análogas de ação rápida mostra que existe diferença estatísticamente significativa entre a insulina regular e aspart [MD - 0,15 IC95%(-0,24;- 0,06)] no desfecho eficácia. Para avaliação de efetividade das insulinas análogas em pacientes DM tipo 1 pertencentes ao programa de insulinas análogas da 2ª Regional de Saúde do Estado do Paraná, foram incluídos 148 pacientes maiores de 18 anos. O parâmetro avaliado foi a redução de HbA1c, que entre os pacientes foi de 0.36±2,75%. Os resultados médios dos custos diretos totais do tratamento com DM tipo 1 durante 24 meses é de R$ 7334,54, para se obter redução em 0,36% dos valores de HbA1c. O custo teórico para a redução em 1% da hemoglobina glicada é de R$ 20.373, 72. Não sendo observadas diferenças significativas para os desfechos de eficácia, segurança e tolerabilidade entre as nsulinas análogas de ação longa vs. NPH ou insulinas análogas de ação rápida vs. insulina regular. Os dados de efetividade desta coorte confirmam a evidência encontrada na meta-análise, o que nos leva a refletir a necessidade de outros estudos com os demais desfechos que permitam concluir se os investimentos do sistema de saúde estão sendo bem empregados.

Abstract: Type 1 Diabetes mellitus (type 1 DM) is a metabolic disfunction which affects 5%- 10% of the diagnosed diabetes population. This endocrine disease has origin in pancreatic beta cells destruction, consequently resulting in insulin production deficiency. The objective of this work is to evaluate the effectiveness of long-acting nalogue insulins to control the glycemic values from patients with type 1 DM in the state of Paraná. In accordance with the methodology proposed by Cochrane, it was performed a systematic review and a meta-analysis to analyze the results from the use of analogue insulins, used by patiens with type 1 DM. A retrospective cohort study was done with data collected from the State Health Secretary's 2nd Regional Health Area (Curitiba and metropolitan region), which participates in the Paraná state’s Otimization Program for type 1 DM. To analyze direct costs of the treatment with analogue insulins, it was collected data from licitatory processes and other documented informations about payments done by SUS for laboratorial follow up exams (SIGTAP’s data) from patients with type 1 DM. After the systematic search, it was found 1238 articles about slow-acting insulins and 2823 rapid-acting insulins.The analysis revealed the quality of the included articles were poor or moderate. The results indicate that there are no statistically significance which could favor Glargine or NPH insulin, whereas Determir analogue insulin presented the most significant results by using it b.i.d. (mean difference -0.14, 95% IC [-0.21, -0.01]). There are no efficacy differences between the two analogue insulins. The results of safety and tolerability have shown inconsistencies. The treatment with Aspart insulin is the most effective when compared with regular insulin MD -0,13, IC95% (-0,17;-0,09). There were no significant differences between Lyspro insulin e Glulysine insulin. It was not possible to perform direct meta-analysis for these two insulins because there was only one clinical trial found. The direct and indirect comparison analysis for efficacy from rapid-acting insulins had significant statistical differences between regular insulin and Aspart [MD -0,15 IC95%(-0,24;- 0,06)]. To evaluate the effectiveness of analogue insulins from the 2nd Health Regional from Paraná State, 148 type 1 DM patients with more than 18 year old were included in the cohort study. The HbA1c reduction was 0.36±2,75% and the direct costs to reduce this parameter in 1%, along 24 months, was R$ 7334,54. Therefore the theoretical costs to reduce in 0.36% the values of glycosilated Hb is R$ 20.373,72. There are neither significant differences in efficacy, safety and tolerability outcomes between NPH insulin vs. long acting analogue insulins nor rapid-acting insulins vs. regular insulins. The effectiveness data of this cohort study confirmed the evidence found in the meta-analysis, which lead us to re-think the need for additional outcomes studies whether the investments by the health system are properly being used.

Abstract: Type 1 Diabetes mellitus (type 1 DM) is a metabolic disfunction which affects 5%- 10% of the diagnosed diabetes population. This endocrine disease has origin in pancreatic beta cells destruction, consequently resulting in insulin production deficiency. The objective of this work is to evaluate the effectiveness of long-acting analogue insulins to control the glycemic values from patients with type 1 DM in the state of Paraná. In accordance with the methodology proposed by Cochrane, it was performed a systematic review and a meta-analysis to analyze the results from the use of analogue insulins, used by patiens with type 1 DM. A retrospective cohort study was done with data collected from the State Health Secretary's 2nd Regional Health Area (Curitiba and metropolitan region), which participates in the Paraná state’s Otimization Program for type 1 DM. To analyze direct costs of the treatment with analogue insulins, it was collected data from licitatory processes and other documented informations about payments done by SUS for laboratorial follow up exams (SIGTAP’s data) from patients with type 1 DM. After the systematic search, it was found 1238 articles about slow-acting insulins and 2823 rapid-acting insulins. The analysis revealed the quality of the included articles were poor or moderate. The results indicate that there are no statistically significance which could favor Glargine or NPH insulin, whereas Determir analogue insulin presented the most significant results by using it b.i.d. (mean difference -0.14, 95% IC [-0.21, -0.01]). There are no efficacy differences between the two analogue insulins. The results of safety and tolerability have shown inconsistencies. The treatment with Aspart insulin is the most effective when compared with regular insulin MD -0,13, IC95% (-0,17;-0,09). There were no significant differences between Lyspro insulin e Glulysine insulin. It was not possible to perform direct meta-analysis for these two insulins because there was only one clinical trial found. The direct and indirect comparison analysis for efficacy from rapid-acting insulins had significant statistical differences between regular insulin and Aspart [MD -0,15 IC95%(-0,24;- 0,06)]. To evaluate the effectiveness of analogue insulins from the 2nd Health Regional from Paraná State, 148 type 1 DM patients with more than 18 year old were included in the cohort study. The HbA1c reduction was 0.36±2,75% and the direct costs to reduce this parameter in 1%, along 24 months, was R$ 7334,54. Therefore the theoretical costs to reduce in 0.36% the values of glycosilated Hb is R$ 20.373,72. There are neither significant differences in efficacy, safety and tolerability outcomes between NPH insulin vs. long acting analogue insulins nor rapid-acting insulins vs. regular insulins. The effectiveness data of this cohort study confirmed the evidence found in the meta-analysis, which lead us to re-think the need for additional outcomes studies whether the investments by the health system are properly being used.