Background: Decision making in health care should be based on evidence - in individual patient consultations as well as in healthcare systems. Systematic reviews, health technology assessments and evidence-based clinical practice guidelines have been shown as powerful means to support decisions. Nevertheless, evidence-based judgements are not value-free descriptive statements but interwoven with inherent normativity - from the concept of effectiveness to the derivation of recommendations. Since normativity is context dependent, its consideration is of key importance for practice implementation and (trans-contextual) adaptation of the results from evidence-based medicine. Objectives: To discuss the systematic guideline review (SGR) as a method to deal with normativity in guidelines. Chronic heart failure (CHF) in primary care serves as a model. Methods: The SGR involves the following steps: (1) Systematic search for (electronic databases, handsearches) and (2) selection of eligible source guidelines according to predefined criteria (guidelines on CHF management of adults in primary care, published in English or German after 2000), (3) assessment of methodological quality (AGREE-instrument), (4) systematic definition of clinical questions, (5) data extraction into evidence tables, and (6) analysis for consistency in the recommendations, (7) synthesis of the data, (8) validation procedure to verify the guideline judgments in the main primary evidence sources and to justify the recommendations, (9) reformulation of the draft guideline and identification of evidence gaps. Results: A total of 16 guidelines was included (two reviewers, k = 0.95), 5/16 of high quality. Within the identified framework of 35 recommendations: 25/35 were consistent, 9/35 inconsistent, 1/35 unratable. Out of 25 consistencies n = 14 were based on consensus (diagnostics, non-pharmacologic therapy, diastolic CHF), n = 7 on evidence, n = 4 differed in grading. Major inconsistencies were found in 3/9 recommendations (brain natriuretic peptides [BNP], AT1-blockers). In 17 statements we re-evaluated the evidence; the majority was congruent. Using BNP-testing as a model, analysis of inconsistencies and incongruencies can reveal different normative constraints. Conclusions: The SGR is a systematic approach to the development of evidence-based guidelines. It identifies grey zones of clinical practice and discloses contentious normative issues to prepare the following discourse (e.g. formal consensus process).