Integrating expert-based content into guideline development when literature is limited

Article type
Authors
Migliavaca CB1, Miguel SRPDS1, Stein C1, Cruz LN1, Colpani V1, Falavigna M1
1Hospital Moinhos de Vento
Abstract
Background: Chagas disease (CD) is a neglected tropical disease caused by the protozoan Trypanosoma cruzi. Two drugs are available for its treatment: benznidazole and nifurtimox. Although these medications prevent CD vertical transmission, guidelines do not recommend their use during pregnancy due to risk of teratogenicity. In a systematic literature search we found only two case reports of pregnant women using benznidazole, both showing no congenital abnormalities. This evidence is very limited to provide clinical recommendations.

Objectives: To systematically collect physicians’ experience with benznidazole and nifurtimox in pregnant women with CD using structured forms.

Methods: We conducted an online survey with physicians with experience in CD treatment, between December 2016 and March 2017. Forms included questions related to physicians’ previous experience with the use of benznidazole and nifurtimox in pregnant patients with CD, including infant outcomes if known.

Results: We contacted 35 physicians of which 17 completed the survey. Only two reported use of
benznidazole in a total of four pregnant patients, and there was no report of vertical transmission, congenital abnormalities or pregnancy complications. Data are still being collected and will be used for the development of the Brazilian Guideline for Diagnosis and Treatment of Chagas Disease. Full results will be presented at the Conference.

Conclusions: Although clinicians do not have much experience in prescribing antiparasitic drugs to pregnant women with CD, in this survey we identified twice the number of cases published in the literature. It is important to recognise the high risk of recall bias in this approach, resulting in low-quality evidence. However, the systematic and structured collection and analysis of experts’ experience can be considered similar to a case series report and may provide complementary evidence for guideline development when there is limited evidence in the literature.