Evidence review

A literature search was conducted to identify studies on the use of co-enzyme Q10 for mitochondrial disorders in children. This identified 125 references (see search strategy for full details). These references were screened using their titles and abstracts and 18 references plus a further 13 studies and case reports referenced in a Cochrane review on treatment for mitochondrial disorders (Pfeffer et al. 2012) were obtained and assessed for relevance.

Of these references 3 case reports in children (Montini et al. 2008, Papadimitriou et al. 1996 and Rotig et al. 2000) are briefly discussed in this evidence summary. No randomised controlled trials (RCTs) or higher quality observational data in children were found. Because the evidence found relating to children was very limited, the best available evidence (in this case, RCTs or the best comparative studies) on the use of co-enzyme Q10 for mitochondrial disorders in adults and young people identified from the search has also been included in this evidence summary. This comprised 2 RCTs (Chen et al. 1997 and Glover et al. 2010) and a non-randomised study (Bresolin et al. 1990).

The literature search was repeated without the children limit to identify any additional RCTs on the use of co-enzyme Q10 for mitochondrial disorders in adults. Any studies that had already been identified during the search in children were excluded. This literature search identified 262 references (see search strategy for full details). These references were also screened using their titles and abstracts and 4 were obtained and assessed for relevance. None of these 4 references were included in the evidence summary (see excluded studies).

Summaries of Bresolin et al. (1990), Chen et al. (1997) and Glover et al. (2010) are shown in table 2 (see evidence tables for full details). The 3 case reports in children (Montini et al. 2008, Papadimitriou et al. 1996 and Rotig et al. 2000) are briefly discussed in the clinical effectiveness section.

Table 2 Summary of included comparative studies in adults and young people

Study

Population

Intervention and comparison

Primary outcome

Bresolin et al. (1990) 2 part study: part 1 non-comparative part 2 comparative non-randomised vs. placebo

Mitochondrial myopathies (n=59 part 1 and n=16 part 2)

Co-enzyme Q10 2 mg per kg daily vs. placebo

Primary outcome not specified. See evidence table for full details of study outcomes.

Chen et al. (1997) RCT

Mitochondrial encephalomyopathies (n=8)

Co-enzyme Q10 160 mg daily vs. placebo

Primary outcome not specified. See evidence table for full details of study outcomes.

Glover et al. (2010) RCT

Mitochondrial disorders (n=30)

Co-enzyme Q10 600 mg twice daily vs. placebo

Primary outcome not specified. See evidence table for full details of study outcomes.

Abbreviations: RCT, randomised controlled trial.

The remaining 29 references from the 35 that were obtained and assessed for relevance were excluded. These are listed in excluded studies with reasons for their exclusion.

Clinical effectiveness

Very limited evidence was found on the use of co-enzyme Q10 for treating mitochondrial disorders in children. Three case reports in a total of 6 children (Montini et al. 2008, Papadimitriou et al. 1996 and Rotig et al. 2000) were identified, no RCTs or higher quality observational data in children were found.

Montini et al. (2008) updated a previous case report by Salviati et al. (2005) of a brother and sister with co-enzyme Q10 deficiency caused by a homozygous missense mutation in the COQ2 gene. The boy developed corticosteroid-resistant nephrotic syndrome at 12 months of age, and progressive encephalomyopathy with stroke-like episodes developed at 18 months of age. Treatment with co-enzyme Q10 (30 mg per kg daily) was started when he was 22 months of age and his neurological signs and symptoms improved. Over the following 11 months, muscle tone and strength returned to normal, myoclonus disappeared, dysphagia improved and he gradually regained lost developmental milestones. However, at the age of 7 years he still had severe neurological symptoms including cognitive impairment, seizures and hemiplegia. His sister was diagnosed with co-enzyme Q10 deficiency at 12 months of age, before any symptoms developed. Immediately after diagnosis she developed nephrotic syndrome. It was reported that treatment with co-enzyme Q10 (30 mg per kg daily) was associated with a resolution of nephrotic syndrome. Treatment with diuretics and continuous haemofiltration were also given. After 50 months' treatment her renal function and neurological status were normal.

Rotig et al. (2000) reported on a brother and sister with multiple respiratory-chain dysfunction attributed to widespread deficiency of co-enzyme Q10. The boy developed nephrotic syndrome at age 3 years and he required renal transplantation at age 9 years. He also developed visual and hearing loss and cardiomyopathy. Progressive ataxia, dystonia and muscle atrophy meant that by age 12 years he could no longer walk unaided or ride his bike. His 1-year-older sister had a milder form of the disease: she also had myopia and hearing loss and nephrotic syndrome that necessitated renal transplantation at age 8 years. However, she did not require wheelchair or motor assistance. Co-enzyme Q10 at a dose of 90 mg daily was given to both siblings, although it is unclear at what age this was started. After 2 months' treatment the boy could stand, walk unaided and ride his bicycle. His bodyweight, muscle bulk, head control and precise movements also improved. His sister's skills and general condition greatly improved; she could pronounce an increasing number of words, became able to write her name, catch a ball and interact better with her environment.

Papadimitriou et al. (1996) reported their observations on 2 children (a 12 year old boy and a 10 year old girl) with Kearns-Sayre Syndrome and hypoparathyroidism who were already being treated with alfacalcidol. Hypercalcaemia developed in both children after 2 months of concomitant treatment with co-enzyme Q10 (100mg daily). Discontinuation of both alfacalcidol and co-enzyme Q10 led to hypocalcaemia. Treatment with co-enzyme Q10 alone was reported to maintain serum calcium levels within the normal range. During treatment with co-enzyme Q10, both patients experienced a slight clinical improvement of tremor and other cerebellar symptoms, but not their muscle strength, ophthalmoplegia or cardiac abnormalities.

Three comparative studies on the use of co-enzyme Q10 for treating mitochondrial disorders in adults and young people (Bresolin et al. 1990, Chen et al. 1997 and Glover et al. 2010) were found, all of which have a number of serious limitations (see evidence strengths and limitations).

An overview of the study results for clinical effectiveness for Bresolin et al. (1990), Chen et al. (1997) and Glover et al. (2010) can be found in results tables.

Glover et al. (2010) reported a randomised double-blind placebo-controlled crossover study which compared co-enzyme Q10 at a dose of 600 mg twice daily with placebo (each taken for 60 days) in 30 adults with mitochondrial disorders. After 60 days treatment there was no statistically significant difference between co-enzyme Q10 and placebo for mitochondrial disease-specific activities of daily living, quality of life assessment scores or handgrip fatigue tests. See the results tables for more details. This study was included in a Cochrane review on treatment for mitochondrial disorders (Pfeffer et al. 2012).

Chen et al. (1997) reported a double-blind placebo-controlled crossover study which compared co-enzyme Q10 at a dose of 160 mg daily taken for 3 months with placebo taken for 1 month in 8 adults and young people (age range 17 to 68) with mitochondrial encephalomyopathy. There was no statistically significant difference between co-enzyme Q10 and placebo for sustained endurance strength (defined as the mean time taken in seconds to perform specified activities) or fatigability in activities of daily living assessment score. There was a statistically significant increase in the global Medical Research Council (MRC) muscle scale score with co-enzyme Q10 compared with placebo. However, the clinical significance of this increase (from approximately 83% to 87%) is unclear. When the proximal and distal limb muscle groups were assessed individually there was no statistical significant difference between co-enzyme Q10 and placebo for the MRC muscle scale scores for any of the muscle groups. See the results tables for more details.

In Bresolin et al. (1990), during the first phase of the study 59 adults and young people with mitochondrial myopathies took co-enzyme Q10 at a dose of 2 mg per kg daily for 6 months. Results were presented for the 44 participants who completed this phase of the study. After 6 months treatment with co-enzyme Q10 there was a statistically significant increase in the global MRC muscle scale score (from approximately 89% to 91%) and a statistically significant decrease in post-exercise serum lactate levels. However, the clinical significance of these changes is unclear. Sixteen participants classed as responders to co-enzyme Q10 (defined as at least a 25% reduction in post-exercise serum lactate after 6 months treatment) were entered into the comparative phase of the study which compared co-enzyme Q10 with placebo for 3 months. There was no significant change in the global MRC muscle scale score in either the co-enzyme Q10 or placebo groups after entering this phase of the study. In the group that received a total of 9 months treatment with co-enzyme Q10, there was a statistically significant decrease in post-exercise serum lactate from baseline to 9 months. However, in the group that received 6 months treatment with co-enzyme Q10 followed by 3 months of placebo, the decrease in post-exercise serum lactate from baseline to 9 months was not statistically significant. See the results tables for more details.

Safety and tolerability

The BNF for children (BNFc) lists side effects of co-enzyme Q10 as nausea, diarrhoea, heartburn and rarely causing headache, irritability, agitation and dizziness. The BNFc recommends that the dose of co-enzyme Q10 should be reduced in people with moderate or severe renal impairment. Co-enzyme Q10 may reduce insulin requirement in people with diabetes and may enhance or reduce the anticoagulant effect of warfarin (BNFc).

Bresolin et al. (1990), Chen et al. (1997) and Glover et al. (2010) provided no information on the safety and tolerability of co-enzyme Q10. The 3 case reports in children (Montini et al. 2008, Papadimitriou et al. 1996 and Rotig et al. 2000) also provided no information on this.

The summary of product characteristics (SPC) for idebenone (Raxone), a synthetic analogue of co-enzyme Q10 which is licensed for the treatment of visual impairment in adolescent and adult patients with Leber's Hereditary Optic Neuropathy (LHON) lists nasopharyngitis and cough as very common adverse reactions (1 in 10 or more) and diarrhoea and back pain as common adverse reactions (between 1 in 100 and 1 in 10). The SPC reports that clinical trial data in children with LHON is limited to children and young people aged 14 years and older and that no data is available on the use of this product in people with hepatic or renal impairment. It also reports that there is no available data from controlled clinical trials on treatment with idebenone for longer than 6 months. Evaluation of data from people taking idebenone for treatment of Friedreich's ataxia and LHON led to the inclusion of blood count abnormalities and abnormal liver function and hepatitis as potential safety concerns for idebenone in the European Public Assessment Report (EPAR: Raxone).

Evidence strengths and limitations

The studies included in the evidence summary have many limitations that affect their application to clinical practice. In addition, mitochondrial disorders are a heterogeneous group of rare diseases and participants included in the studies many not represent all types of patients seen in clinical practice.

Very limited evidence was found on the use of co-enzyme Q10 for mitochondrial disorders in children. Three case reports in children were found and are discussed in this evidence summary (Montini et al. 2008, Papadimitriou et al. 1996 and Rotig et al. 2000). As uncontrolled observational studies in individual patients these case reports are prone to bias and other methodological problems. Whilst they did report some changes in functional ability, they provided very limited data on objective measurable outcomes.

Because of the very limited evidence found specific to the use of co-enzyme Q10 for mitochondrial disorders in children, the best available evidence in adults and young people was also included. This comprised 2 RCTs (Chen et al. 1997 and Glover et al. 2010) and a non-randomised study (Bresolin et al. 1990). Mitochondrial diseases are chronic conditions but all 3 were short-term studies that may have been of insufficient length to fully evaluate the efficacy of co-enzyme Q10. In addition, they recruited small numbers of participants and therefore may have been insufficiently powered to reliably detect any differences between placebo and co-enzyme Q10. All 3 studies included several outcomes measures, including objective measurements of muscle power. However none of the studies defined a primary outcome measure, and the clinical relevance of some of the outcomes reported is also unclear.

In Chen et al. (1997) there was a washout period of 1 month between treatments if the participant was randomised to co-enzyme Q10 first (but not if they received placebo first). This may have allowed a blinded assessor to know whether a participant was taking treatment or placebo depending on the duration of time in the study. Bresolin et al. (1990) was a 2-part study with an open label non-comparative phase and a comparative phase. The comparative phase of the study was stated to be 'blind' but insufficient information was provided to assess whether this was single or double blinding. This phase was at high risk of bias because entry eligibility depended upon the results of the open-label phase and though comparative it was not randomised.

In Glover et al. (2010) participants had previously received nutritional supplements including co-enzyme Q10 for several years. In Chen et al. (1997) participants included in the study also had a disease duration of several years, however no information was provided in the paper about current medication (including nutritional supplements) that participants may have been taking prior to entering the study. For both studies, there was insufficient information provided in the paper to assess if the groups were similar at the start of the trial with respect to previous supplements taken.

None of the studies included in this evidence summary provided information on the safety and tolerability of co-enzyme Q10, making it difficult to establish the safety of co-enzyme Q10 for treating mitochondrial disorders from the available data. The studies used a wide variety of different doses of co-enzyme Q10.

An overview of the quality assessment for Bresolin et al. (1990), Chen et al. (1997) and Glover et al. (2010) can be found in the evidence tables.