Defining phenotypes of long-term lithium and valproate response, including combination therapy: a modified application of the Alda scale in patients with bipolar disorders

Jinyoung Lee, Ji Hyun Baek, Dongbin Lee, Sung Woo Ahn, Soyung Yang, Yujin Choi, Yong Chun Bahk, Kyung Sue HongInternational Journal of Bipolar Disorders 2020 · DOI

TL;DR

This study investigates the effectiveness of lithium and valproate, commonly used mood stabilizers for bipolar disorder, in patients receiving monotherapy versus combination therapy. It reveals that patients on combination therapy tend to have poorer long-term treatment responses, suggesting that excluding them from studies might introduce bias. The research also identifies distinct clinical factors associated with the effectiveness of lithium and valproate individually.

What Problem Does This Paper Address?

Bipolar disorder (BD) is a chronic condition requiring long-term mood stabilizer treatment to prevent relapses and manage symptoms. Lithium and valproate are cornerstone medications for this purpose. However, a significant challenge in studying their long-term efficacy is how to handle patients who receive combination therapy (e.g., lithium and valproate). Traditionally, these patients are excluded from studies because combination therapy is seen as a confounding factor. This exclusion might be problematic, as patients requiring combination therapy could represent a group with more severe or treatment-resistant illness, meaning their exclusion could skew the results and lead to an overestimation of monotherapy effectiveness. This paper aims to address this by applying and modifying the Alda scale, a standard tool for assessing long-term treatment response, to include patients on combination therapy and to explore whether this inclusion reveals important differences and clinical correlates.

Methods

The researchers recruited 102 patients diagnosed with bipolar I or II disorder who had been on lithium and/or valproate maintenance therapy for at least two years. They divided patients into three groups: lithium monotherapy, valproate monotherapy, and lithium-valproate combination therapy.

The primary method for assessing long-term treatment response was the Alda scale. This scale retrospectively measures global treatment outcome (A score) and confounding factors (B score), with the total score being A-B. A higher Alda score indicates a better response. The study also used the Clinical Global Impressions-Bipolar Illness, overall (CGI-BP-III-O) score.

To investigate clinical correlates, simple regression analyses were performed, using the total Alda score as the dependent variable. These analyses were conducted separately for lithium and valproate users, excluding the combination group.

Datasets

Not discussed in the paper. The study utilized retrospective clinical records and patient interviews from a single tertiary-care university-affiliated hospital.

Key Insights and Ideas

  • Combination Therapy is Associated with Poorer Response: A critical finding is that patients on lithium-valproate combination therapy showed significantly worse long-term treatment outcomes compared to those on monotherapy. This challenges the practice of excluding these patients from research.
  • Exclusion of Combination Therapy Users May Introduce Bias: The study strongly suggests that excluding patients on combination therapy could lead to biased results, as these individuals may represent a group of poorer responders. This highlights the need to develop methods that can accommodate combination therapy in research.
  • Distinct Clinical Correlates for Lithium and Valproate: The research identified different clinical factors associated with treatment response for lithium and valproate. This implies that the choice of mood stabilizer might be influenced by specific patient characteristics, and that a one-size-fits-all approach to understanding response predictors is insufficient.
  • Alda Scale Adaptability: The study demonstrates that the Alda scale, with potential modifications, can be applied to evaluate treatment response in patients receiving combination therapy, opening avenues for more inclusive research.

Key Results

  • Treatment Response Differences: The combination therapy group exhibited the worst treatment response across all measured outcomes (Alda A score, total Alda score, and CGI-BP-III-O). They also had a significantly higher rate of "poor responders" compared to the valproate monotherapy group.
  • Baseline Characteristics: Importantly, there were no significant differences in baseline demographic or illness characteristics among the three groups, suggesting that the poorer response in the combination group was not due to more severe illness at the outset, but rather to their treatment needs.
  • Clinical Correlates for Lithium: In lithium users, older age at onset and a (hypo)manic episode at onset were positively associated with better treatment response (higher Alda score).
  • Clinical Correlates for Valproate: In valproate users, comorbid anxiety disorders, obsessive-compulsive disorder, and mixed episodes were negatively associated with treatment response, indicating that these comorbidities are linked to poorer outcomes with valproate. Increased appetite during depressive episodes was also associated with poorer valproate response.

Limitations and Future Work

The authors acknowledge several limitations:

  • Small Sample Size: The overall sample size was relatively small, and the BD-II subgroup analysis was not possible due to limited numbers. Similarly, evaluating clinical correlates for the combination group was not feasible.
  • Retrospective Design: The study relied on retrospective data collection, which inherently carries limitations compared to prospective studies. However, the authors note that this is often unavoidable for long-term treatment response evaluations.
  • Physician Preference: Individual treatment strategies can be influenced by physician preference, which might indirectly affect treatment response.
  • Single-Center Study: Recruitment from a single tertiary academic hospital may limit the generalizability of the findings.

Future work should focus on developing improved measurement methods and study designs that can effectively handle the complexities of combination treatment in both clinical and genetic studies of bipolar disorder. Further research is warranted to explore the distinct clinical factors influencing response to different mood stabilizers.

Summary updated

Abstract

The combination group had poorer long-term treatment response but did not show distinct clinical characteristics compared to the monotherapy groups. When exploring the long-term effects of mood stabilizers, excluding patients undergoing combination treatment could result in bias because they may represent a poor response group. The long-term treatment responses of lithium and valproate had different clinical correlates.
BibTeX
@article{lee2020defining,
  author = {Jinyoung Lee and Ji Hyun Baek and Dongbin Lee and Sung Woo Ahn and Soyung Yang and Yujin Choi and Yong Chun Bahk and Kyung Sue Hong},
  title = {{Defining phenotypes of long-term lithium and valproate response, including combination therapy: a modified application of the Alda scale in patients with bipolar disorders}},
  journal = {International Journal of Bipolar Disorders},
  year = {2020},
  doi = {10.1186/s40345-020-00199-w},
}
Figure