Effectiveness of Coenzyme Q10 Supplementation for Type 2 Diabetes Mellitus: A Systematic Review and Meta-Analysis

Objective To evaluate the effectiveness and safety of coenzyme Q10 for patients with type 2 diabetes mellitus (T2DM). Methods Data from randomized controlled trials were obtained to assess the effects of coenzyme Q10 versus placebo or western medicine on patients with T2DM. The study's registration number is CRD42018088474. The primary outcomes included glycosylated hemoglobin, fasting blood glucose, and fasting insulin. Result Thirteen trials involving 765 patients were included. Compared with the control group, coenzyme Q10 may decrease the HbA1c (WMD −0.29; 95% CI −0.54, −0.03; P = 0.03) and the fasting blood glucose (WMD −11.21; 95% CI −18.99, −3.43; P = 0.005). For fasting insulin, there is also not strong evidence that confirms which one is better because there was no statistical difference (WMD −0.48; 95% CI −2.54, 1.57; P = 0.65). Conclusion Based on current evidence, coenzyme Q10 may assist glycemic control, decrease TG, and improve HDL-C in patients with T2DM.


Introduction
Diabetes is one of the major global health issues. According to the World Health Organization, in 2014, about 422 million adults were suffering from diabetes. In the US, more than 9% of the US population were affected by diabetes [1][2][3]. In the global population of diabetes, 90%-95% are with type 2 diabetes mellitus (T2DM). Since 1980, the global prevalence of diabetes in adults has nearly doubled. Diabetes caused 1.5 million deaths in 2012 as it relates to increased risk of cardiovascular and other diseases [4,5]. And according to estimates of the Diabetes Federation Diabetes Atlas, the number of patients with diabetes in 2035 will reach 592 million [6]. Patients with T2DM are at high risk of developing hyperglycemia-related cardiovascular complications such as heart disease, hypertension, stroke, retinopathy, and nephropathy. In the US, disability, loss of work, and premature death caused by diabetes have resulted in a huge direct and indirect medical costs [2]. Without effective prevention and management programs, a further significant increase in diabetes will have grave consequences on the health and lifespan of the world population [6]. At present, the treatment of diabetes is still lacking etiological treatment; the treatment of diabetes is mainly achieved through diabetes management including healthy eating, weight control, appropriate physical activity, antiglycemic medications, and multifactorial risk reduction [7].
T2DM occurs when the organism fails to respond to the increased blood glucose caused by the impaired β cell secretion and/or insulin resistance [1,8]. Therefore, T2DM is a complex chronic disease. Oxidative stress and insulin resistance are recognized pathogenic mechanisms in the development and progression of T2DM and its complications [1,9]. Insulin, secreted by β cells, plays crucial roles in many metabolic processes such as regulating the glucose uptake [1,10]. Therefore, when insulin secretion is impaired, glucose uptake will also be affected. Persistent hyperglycemia will lead to the overproduction of reactive oxygen species, which can cause oxidative damage to deoxyribonucleic acid (DNA), proteins, and lipids [1,10,11]. Because the mitochondria are in proximity to the sources of reactive oxygen species, oxidative stress often leads to mitochondrial damage, resulting in mitochondrial dysfunction. An increasing number of studies have shown that in obese and T2DM patients, reactive oxygen species can aggravate the insulin resistance status and interfere with the insulin signaling pathway through impairing the mitochondrial ability to oxidize fat [12,13]. Meanwhile, the abnormality in mitochondrial functions secondary to oxidative stress is one of the mechanisms leading to T2DM and T2DM-related complications [11].
Coenzyme Q10 is a lipid-soluble nutrient widely present in living cells. Coenzyme Q10, as an effective antioxidant, can scavenge free radicals and protect cells from oxidation. Recent studies have found that T2DM patients have significantly lower levels of coenzyme Q10 than healthy people [1,[14][15][16], which indicates that coenzyme Q10 deficiency may reduce the organism's ability to counter hyperglycemiainduced oxidative stress in T2DM [17,18]. This suggests that coenzyme Q10 plays an important role in the pathogenesis of T2DM [19]. Therefore, exogenous coenzyme Q10 supplements may improve the oxidative stressinduced abnormalities in mitochondrial functions, thereby bettering glycemic control in patients with T2DM [20].
Therefore, the purpose of this paper is to review the available randomized controlled trials (RCTs) to evaluate the effectiveness of coenzyme Q10 for T2DM. Although systematic reviews and meta-analyses of the effects of coenzyme Q10 on the metabolic profile of diabetes mellitus [21] and diabetes-related biomarkers [22] have been performed, the analysis in 2015 only showed that coenzyme Q10 may reduce triglyceride levels, while the analysis in 2016 only included RCTs before 2014. Over time, more RCTs about coenzyme Q10 were published between 2014 and 2018. Therefore, the results of systematic reviews and meta-analyses need to be updated. This systematic review and meta-analysis is a registered review with protocol (CRD42018088474) in PROSPERO, which is aimed at evaluating the effects of coenzyme Q10 on T2DM.  Table 1.

Materials and Methods
Studies meeting the inclusion criteria were included in this review (see Box 1).

Data Extraction.
Three reviewers (Shi-ying Zhang, Kai-lin Yang, and Liu-ting Zeng) independently selected the studies and extracted the data from the studies. Disagreements were resolved by discussion of five reviewers (Shi-ying Zhang, Kai-lin Yang, Liu-ting Zeng, Xiao-he Wu, and Huiyong Huang). We first reviewed the titles and abstracts of each of the studies and excluded the articles that do not meet the eligibility criteria. Then, we assessed the full texts of studies that meet the criteria. A customized form was used to record authors, year of publication, intervention, control group, outcomes, AEs, and duration.
If there is missing information in the article, reviewers would attempt to contact the authors to obtain the data or impute the missing data according to the Cochrane Handbook 5.1.0 [23]. If P < 0 05, the missing standard deviations would be imputed by a P value; if P > 0 05 or P = NS, it would be imputed by using the average of candidate standard deviations [23].

Study Quality Assessment.
We assessed the risk of bias by using the risk of a bias assessment tool based on the Cochrane Handbook [24]. The criteria consist of 7 items: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective outcome reporting, and other sources of bias. The study quality assessment was performed by three reviewers (Shi-ying Zhang, Kai-lin Yang, and Liu-ting Zeng) independently. Disagreements were resolved by consensus among all five reviewers (Shi-ying Zhang, Kai-lin Yang, Liu-ting Zeng, Xiao-he Wu, and Hui-yong Huang).

Statistical
Analysis. The data were analyzed by RevMan 5.3 software. The dichotomous variable measure was summarized by risk ratio (RR) with a 95% confidence interval (CI). The continuous outcomes underwent meta-analysis using mean differences (MD) and 95% CI. Heterogeneity among studies was assessed using Cochrane's Q and I 2 statistic [25]. The fixed effect model would be used when P > 0 1 and I 2 < 50%. We would explore the reasons for heterogeneity, perform the subgroup analysis, or use the random effects model when P < 0 1 and I 2 > 50%.

3.1.
Results of the Search. Two hundred ninety-seven articles were found in the initial search; two hundred eighty of them were excluded based on the title and abstract and seventeen of them were retrieved for more detailed evaluation. Finally, thirteen studies were included in this systematic review and meta-analysis, and four were excluded ( Figure 1).

Description of Included Trials and Risk of Bias in Included
Studies. Thirteen RCTs with 765 participants met the inclusion criteria. All of them were parallel-group RCTs. In  [27], there were two trial groups and two control groups. According to the Cochrane Handbook 5.1.0, we split the shared trial and control groups into two groups with a smaller sample size [23] and include the four reasonably independent comparisons [ Table 2.
With regard to the selection bias, eight RCTs [19, 26-32] did not describe the randomization procedures, and twelve RCTs [19,[26][27][28][30][31][32][33][34][35][36][37] did not describe an acceptable method of allocation concealment; thus, we thought that their risk of bias is unclear. One RCT [33] used block randomization, three RCTs [34,36,37] utilized permuted random block allocation, and one RCT used a website [35] to conduct randomization; we therefore rated them as having a low risk of bias. One RCT [29] describes an acceptable method of allocation concealment, and it was rated as having a low risk of bias. As for the performance bias and detection bias, two trials were unclear [32,33], but they used objective measures (e.g., HbA1c, fasting insulin, and fasting glucose) and the outcome is not likely to be influenced by the lack of blinding, while the remaining three studies used blinding; thus, we gave a low risk of bias for all. None of trials missed data and incompletely reported the outcomes; therefore, we gave a low risk of bias. Other sources of bias were at low risk in all of the included studies. A graphical summary of the risks of bias assessment is presented in Figure 2.

Discussions
This study is a systematic review and meta-analysis that regards the effects of coenzyme Q10 on T2DM patients. It synthesizes the results from 13 RCTs (including the newest RCTs [29][30][31][32][33][34]36] after 2014) involving 765 participants to draw an overall conclusion. Although significant differences between groups were found for some outcomes, the available evidence shows that coenzyme Q10 may improve the glycemic control (decreasing the HbA1c and fasting blood glucose) and blood lipids (decreasing TG and increasing HDL-C), suggesting that coenzyme Q10 may assist glycemic control and protect the cardiovascular system. However, these findings should be interpreted with caution although it seems like that they are prospective. The unclear risk of bias for selection bias (random sequence generation and allocation concealment), the small number of patients, and the high heterogeneity limited the promotion of the results. In addition, the lack of statistical significance of fasting insulin, HOMA-IR, TC, LDL-C, and adiponectin does not equate to no medical significance. Instead, it may mean that coenzyme Q10 may be the safer or cheaper treatment options.
Coenzyme Q10 deficiency, particularly ubiquinol (the reduced form of coenzyme Q10) deficiency, is often observed among patients with T2DM. The ubiquinol/ubiquinone ratio is often utilized as an indicator to react to the body's oxidative stress [38]. Decreased ubiquinol levels are often accompanied by increased ubiquinone levels, suggesting that there is an ineffective conversion between ubiquinone and ubiquinol. It also indicates that the body's ability to scavenge free radicals is reduced. Meanwhile, the impaired conversion of ubiquinone to ubiquinol is often found in many diseases [39]. Román-Pintos et al. [15] found that coenzyme Q10 levels in T2DM patients were significantly lower than those in healthy people, while their MDA levels were significantly higher than those in healthy one. Furthermore, the ubiquinone/ubiquinol ratio in T2DM patients was continuously higher than that in normal heathy people throughout the day after breakfast, which indicates that postprandial hyperglycemia is associated with increased oxidative stress [16].    coenzyme Q10 level is beneficial for this cycle [40]. Therefore, based on the evidence above, the abnormalities in mitochondrial functions secondary to oxidative stress in T2DM patients may be potentially relieved by restoring the coenzyme Q10 level through exogenous coenzyme Q10 supplementation, ultimately improving the glycemic control.
None of RCTs reported information about AEs, which means that there is currently a lack of reports on AEs. However, this does not mean that the intervention of coenzyme Q10 is safe [41]. Thus, although, based on current evidences, we consider that coenzyme Q10 is a relatively safe treatment, we cannot assure it. Future clinical trials are required to report AEs with more explanations [42].
Compared with previous reviews [21,22], the strengths of this systematic review and meta-analysis are that it is strictly conducted in accordance with the protocol registered on PROSPERO and it assesses more outcomes. It also included seven new RCTs [  this systematic review and meta-analysis also has limitations, including but not limited to the number of patients participated which is only 765 and the heterogeneity of some outcomes. Such heterogeneity confounds the interpretation of statistical findings. The heterogeneity may come from the potential discrepancies in the pharmacological effects of various coenzyme Q10 preparations which may result from different standardizations of the coenzyme Q10 manufacturing process, dosage, duration of treatment, units of laboratory tests, and races of the selected patients or other places. Therefore, the random effects model was adopted, although it cannot completely eliminate heterogeneity. Meanwhile, the study duration is generally short-to-medium term (mostly 12 weeks), and the long-term efficacy of coenzyme Q10 is temporarily uncertain. Furthermore, in the included studies, the dose of CoQ10 in the experimental group was not necessarily    the same, which could lead to differences in the effect of lowering blood glucose; the difference in dosage makes it difficult to determine the minimum effective dose of coenzyme Q10. Finally, due to the fact that none of the trials reported AEs, the safety of coenzyme Q10 should be interpreted with caution. In the future, more similar highquality randomized controlled trials are needed to amend the results of this systematic review and meta-analysis.

Conclusion
Coenzyme Q10 may assist glycemic control, decrease TG, and improve HDL-C in patients with T2DM. However, the limitations in RCTs, including small sample sizes and short duration, make the result be interpreted cautiously. The benefits from long-term treatment of coenzyme Q10 beyond 6 months remain to be defined by future studies. Meanwhile, more randomized, double-blind, large-sample-size trials of coenzyme Q10 for T2DM are needed in the future to validate or revise the result of this work.

Conflicts of Interest
We declare no competing interests.