Role of the CINP 2017 Treatment Guidelines for Bipolar Disorder in clinical practice

In developing the CINP 2017 Treatment Guidelines for Bipolar Disorder, the guidelines authors — Professors Konstantinos Fountoulakis from Thessaloniki, Greece; Eduard Vieta from Barcelona, Spain; Allan Young from London, UK; Lakshmi Yatham from Vancouver, Canada; Heinz Grunze from Salzburg, Austria; Pierre Blier from Ottawa, Canada; Hans Jurgen Moeller from Munich, Germany; and Siegfried Kasper from Vienna, Austria — decided to follow a strict evidence-based approach based on randomized controlled trials (RCTs) and related meta-analyses, explained Professor Moeller.

Distilling the complex research data into a clinically meaningful tool

A detailed analysis of the evidence base distilled into clinical recommendations

Evidence-based medicine is the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients,¹ Professor Moeller said.

Guidelines are derived from a detailed analysis of the evidence base distilled into clinical recommendations to assist clinicians in their treatment decisions.

The CINP 2017 Treatment Guidelines for Bipolar Disorder is published in four separate articles^2–5 addressing:

• the background and methods for developing the guidelines
• reviewing and grading of the evidence and development of a precise algorithm
• the clinical guidelines
• the unmet needs — to reflect the many issues for which there are no clear answers

  The workgroup analyzed the many grading systems used by different guidelines groups — for example the eight levels of evidence and four grades of recommendation used by the Scottish Intercollegiate Guidelines Network⁶ — and decided to create a new simple grading system with:

• four Levels of efficacy together with a fifth category for negative data:
  
  • Level 1 — good research-based evidence supported by at least two good-quality placebo-controlled RCTs
  • Level 2 — fair research-based evidence from one RCT
  • Level 3 — comparative studies without placebo arm, post-hoc analyses
  • Level 4 — inconclusive data or poor-quality RCTs
  • Level 5 — negative data
• three Levels for safety and tolerability:
  
  • Level 1 — very good tolerability with few non-enduring nonlife-threatening side effects
  • Level 2 — moderate tolerability with many (possibly enduring) side effects that are not life-threatening; or good tolerability with rare life-threatening side effects
  • Level 3 — poor tolerability with many enduring side effects that cause distress, compromise health or are life-threatening; or moderate tolerability with rare life-threatening side effects

Very good safety and tolerability is required for a Level 1 recommendation

Treatment is a decision based on benefits vs risk (efficacy vs tolerability)

The workgroup decided that treatments meeting very good safety and tolerability were of high importance and would be required for a Level 1 recommendation. The algorithm they used for recommendations was as follows:

• Level 1 — Level 1 or 2 for efficacy and Level 1 for safety and tolerability
• Level 2 — Level 1 or 2 for efficacy and Level 2 for safety and tolerability
• Level 3 — Level 3 for efficacy and Level 1 or 2 for safety and tolerability
• Level 4 — Level 4 for efficacy and Level 3 for safety and tolerability
• Level 5 — Level 5 for efficacy (not recommended)

The requirement for very good safety and tolerability for a Level 1 recommendation has, however, resulted in some anomalies in Level 1 recommendations compared with Level 1 recommendations in other guidelines using different algorithms not requiring such a high level of safety and tolerability.

Guidelines inform but should not replace professional knowledge and clinical judgement

Guidelines are not derived from real-world patients

Co-author of the guidelines, Professor Vieta, cautioned that evidence-based guidelines:

• may end up as “evidence-biased” depending upon which randomized controlled trials, systematic reviews and other studies meet the methodological criteria to be included in the evidence base and who funded them
• are based on the treatment for average patients defined within strictly controlled criteria including adherence to therapy — but real-world patients are individuals with comorbidities, use or abuse substances, may be pregnant, and do not adhere to therapy

Professor Vieta recommends that everyone reads the guidelines to become familiar with the current state of knowledge and to guide therapy decisions, but emphasized that they should not:

• replace professional knowledge and clinical judgement
• override the responsibility of the healthcare professional to make appropriate decisions
• take precedence over the consensual decisions made with patients and families

Treatment choice needs to be individualized

Adherence remains a challenge in the real world

Treatment choice for patients with bipolar disorder needs to be individualized according to predictors of response, predominant polarity, treatment adherence and patient preferences for tolerability, said Professor Vieta. 

The sequence of episodes and predominant polarity — for example, mania followed by depression, depression followed by mania — is not normally considered by guidelines but can make a difference: depression may be prevented by treating preceding mania, he explained.

Furthermore, guidelines do not address adherence, which remains a challenge in the real world, said Professor Vieta.

This has been well demonstrated by a Finnish registry study of 18,000 patients, which showed that the use of long-acting injectables decreased the risk of relapse for patients with bipolar disorder by 30% compared with the use of the oral formulations of antipsychotics;⁷ but this effect has not been seen in clinical trials because adherence is largely not an issue in clinical trials.

Another issue for treatment in the real world is that guidelines generally recommend monotherapy in the first instance, but patients are often on a combination of treatments for a variety of reasons, and therapies tend to be added rather than removed when patients relapse. 

The guidelines are therefore helpful for patients who have been newly diagnosed but perhaps not as helpful for complex patients on multiple therapies, concluded Professor Vieta.