Background & Aims

Mitochondrial diseases (MD) are rare diseases characterized by dysfunctions of the mitochondria. Due to the crucial role of mitochondria in supplying energy for cellular functioning, those diseases show a broad range of symptoms, such as enduring exercise-induced muscle fatigue, exercise intolerance, and general fatigue (DiMauro et al., 2013). Pain as a symptom of MDs has been underestimated, although recent findings indicate that up to 91% of patients with MD are affected (Löffler et al., 2020). Questionnaire-based studies provide first evidence that the nature of pain may be neuropathic rather than musculoskeletal in most patients. However, the underlying mechanisms, potential subgroups of patients, and the nature of pain in different types of MD have yet to be investigated. The present study aims to differentiate types of pain profiles in patients with MD to identify subgroups of patients based on the nature of their pain, potentially uncovering the underlying mechanisms.

Methods

We re-analyzed data from a previous study on pain in 96 patients with 15 different types of MD (Löffler et al., 2020). Diagnoses were confirmed by the German Network for Mitochondrial Disorders (mitoNET). Pain quality was assessed using the German version of the pain experience scale (PES; Geissner, 1996) with 24 verbal descriptors of pain, such as “burning”, “cutting” or “numbing”. Missing values were addressed using multiple imputation. To identify subgroups of patients with MD with distinct pain profiles we then conducted a cluster analysis using the k-means algorithm. Silhouette coefficients were calculated to find the most suitable number of clusters. The k-means algorithm was performed with 20 randomized starting points to find the best solution regarding the minimized within-cluster sum of squares (WSS).

Results

Based on pain descriptors of the PES, we found two distinct clusters of patients with MD. Different types of MDs were equally distributed across the two clusters. Among the few exceptions were all nine patients with Leber’s hereditary optic neuropathy (LHON) which were assigned to the first cluster. Clusters therefore seem to represent transdiagnostic categories of patients with MD.

Conclusions

The nature of pain seems to vary independent of the specific type of MD. This is in line with the high phenotypic variability within a specific type of MD despite common genetic mutations and depletions. Our findings need to be viewed in the light of several limitations, such as partly small and varying subgroup sizes. Further, different statistical approaches, such as machine learning, may be better able to discern subgroups of patient-reported pain descriptions and the mechanisms underlying pain in different types of MD. Future studies should take a transdiagnostic approach to identify mechanisms of pain in MD.

References

DiMauro, S., Schon, E. A., Carelli, V., & Hirano, M. (2013). The clinical maze of mitochondrial
neurology. Nat Rev Neurol, 9(8), 429-444. https://doi.org/10.1038/nrneurol.2013.126
Geissner, E., & Schulte, A. (1996). SES: Die Schmerzempfindungs-Skala. Hogrefe.
Löffler, M., Gamroth, C., Becker, S., & Flor, H. (2020). Chronic pain as a neglected core
symptom in mitochondrial diseases. Neurology, 94(8), 357-359. https://doi.org/10.1212/WNL.0000000000009011

Presenting Author

Benjamin Dickmann

Poster Authors

Benjamin Dickmann

MSc

Heinrich-Heine-Universität Düsseldorf

Lead Author

E-mail

Herta Flor

Interdisciplinary Center for Clinical Trials, Johannes Gutenberg University Medical Center Mainz

Lead Author

E-mail

Martin Löffler

Clinical Psychology, Department of Experimental Psychology, Heinrich Heine University Düsseldorf

Lead Author

E-mail

Susanne Becker

Heinrich Heine University Düsseldorf, Institute of Experimental Psychology

Lead Author

E-mail

Topics

  • Mechanisms: Psychosocial and Biopsychosocial