Mitochondrial diseases (MDs) vary considerably in terms of their manifestations and effects on different organ systems, as well as in terms of age of onset and rate of progression (1). Clinical pediatric reports have described the spectrum of signs and symptoms associated with MDs. Classification relies on accurate clinical, biochemical, and genetic information and may be based on either genotype or phenotype, but there is significant overlap. Although the prognosis for MD is dependent on the underlying diagnosis, there is large phenotypic variation (2, 3). Thus, the diagnostic evaluation is necessarily multitiered and broad based, with a focus on integrating information from many avenues (4). In particular, children with encephalopathy experience many limitations in daily life (5). However, information on the long-term clinical course and developmental outcome of this heterogeneous group of diseases is lacking (5–7).

One of the key needs of families caring for children with rare metabolic or genetic impairments is to understand the long-term effect of the disorder on the child’s development (8). Specifically, a better understanding of the needs, problems, and long-term developmental trends of these children is essential for effective care planning and improving the quality of life (1).

Accordingly, the aims of this study were to evaluate the clinical course and long-term developmental trends in pediatric patients with MDs. The present study is the first to attempt to use a developmental scale for retrospectively assessing and describing the long-term developmental trends of children with MDs.

In the present study, the global developmental trends were evaluated over a follow-up period of 7.70 years at nine time points and showed a decline over all periods. However, even with consistently declining patterns, the characteristics of developmental deterioration were disparate. The overall developmental decline pattern could be classified into five phases: (1) pre-diagnostic initial decline phase; (2) pre-diagnostic accelerated decline phase; (3) post-diagnostic alleviated phase; (4) post-diagnostic reaccelerated decline phase; and (5) post-diagnostic stagnant phase.

Considering the pre-diagnostic phase as lead time, the DQ declined continuously, with the initial decline starting from the first onset of symptoms, followed by an accelerated deteriorating phase until diagnosis. During the initial period from diagnosis to the first post-diagnostic evaluation (~1 year), an impeded phase of developmental decline was noted, which appeared alleviated compared with the decline noted in previous phases. This was followed by an accelerated period of decline for ~2 years. Finally, ~3 years after the diagnosis, the decline reached a stagnant phase. In this phase, the developmental deterioration appeared to be reaching the lowest and terminal level of 10% of the DQ. These declining patterns were similar for the overall development and all types of skills, including the gross motor, fine motor, social–personal, language, and cognitive-adaptive. However, a more detailed analysis revealed that the delayed development in gross and fine motor skills exhibited more significant deterioration even in the pre-diagnostic initial decline phase.

In terms of the lead time, the developmental decline accelerated from the first onset of symptoms to the time of confirmation of diagnosis. In contrast, right after the diagnosis, a post-diagnostic alleviated phase was manifested. These facets of the developmental course highlight the importance of shortening the lead time as well as of early detection of the first significant symptom. A shorter lead time is important in achieving a longer developmentally alleviated phase or a prolonged period before reaching the stagnant phase and the lowest and terminal level of development.

As developmental delay is one of the first common symptoms in children with MD (16), vigilance and careful observation of any motor delay in children are integral to early diagnosis. With more structural attention and information, motor delays might be easier to detect by caregivers because of their comparatively evident developmental milestones in earlier childhood compared with language, cognitive, or social development (7). In addition, any motor delay is usually enough for a child to be referred to professionals, and additional signs of any developmental delays in the language, social, and cognitive domains should be further evaluated for etiological diagnosis. Hence, early detection of motor delay and other developmental signs of delay is important in children with MD to shorten the lead time.

Furthermore, the effects of diagnosis and intervention should not be overlooked in children with MD. Such a standardized screening tool to detect MDs early at patients’ routine visits to the pediatricians would be ideal. However, making the correct diagnosis is exceedingly problematic because of the heterogeneous features of this disorder (17, 18). MD may present at any age, with a spectrum of symptoms and signs, to several medical specialties. And there are no standard guidelines for the investigation of MD (2, 3). Diagnostic difficulty results not only from the wide spectrum of symptoms and signs that an individual patient may have but also from the absence of a reliable screening or diagnostic biomarker that is both sensitive and specific in all cases of MD (4). Discussion with a clinician who has expertise in MD is advised in order to guide investigation, in particular with respect to laboratory facilities and handling of specimens (2). Therefore, it might not be easy to adopt a standardized screening tool to detect them early in MD.

Confirming the diagnosis of MD is important to both the parents and physician, because confirmation of diagnosis could relieve the parents’ anxiety and helplessness caused by the uncertainty of an undiagnosed condition and encourage them to seek various therapies for their children. For physicians, confirming the diagnosis of MD is fundamental in planning therapeutic intervention (17, 18).

The appearance of a post-diagnostic alleviated phase in this study is notable and possibly even inspiring to physicians who have only been using symptomatic or supportive treatment for MD. The possibility of a positive effect of the treatment on patients’ development, including mitochondrial cocktails, such as coenzyme Q, carnitine, and various vitamins, which were provided to the children with MD in this study, should not be dismissed. Notably, the children with MD in this study had been taking a mitochondrial cocktail immediately after receiving a confirmed diagnosis. Accordingly, the necessity of an early diagnosis, consecutive symptomatic treatment, general supportive care, and early administration of possible medicine to shorten the lead time to diagnosis and alleviate the patients’ developmental regression is suggested.

We found that diffuse brain atrophy, the clinical rating provided by the physician, and the age at first symptom onset significantly affected the developmental level and decline. However, no significant effect was noted for syndromic diagnosis and epilepsy severity, suggesting that these aspects may not directly reflect the developmental condition of the patients. Because of the heterogeneous symptoms of MDs (14), syndromic diagnosis based on a combination of particular symptoms might not be strongly associated with the developmental function of children with MDs. Additionally, epilepsy-related factors are not the main factors that influence developmental function in such patients, but are nonetheless important manifestations of MDs (14). Instead, diffuse brain atrophy might largely account for the developmental level as a measure of brain function with structural change (16, 19). In particular, all pediatric patients included in the present study presented with central nervous system involvement. In addition, more than half had diffuse brain atrophy, which was associated with lower developmental level at all-time points and over all periods considered, from the early stage of the pre-diagnostic period through the lowest, stagnant, and terminal phase. The concept of clinical ratings (20) provided by the treating physician could be regarded as a good predictor of long-term developmental prognosis because it reflects the characteristics of the patient’s general state and multiorgan involvement.

This study has several limitations. This is a preliminary study with a small number of patients, and the overall cohort was not followed up consistently since we reviewed the data retrospectively. Moreover, in spite of the advantage of other developmental scales such as Bayley scale of Infants Development, KICDT was used as the most frequently and repeatedly used measure in outpatient or inpatient clinical settings in the current study. It still has a strength as a possible long-term and repeatedly useful screening tool, however, other developmental measures might be able to give more reliable information about the patients’ developmental function. In addition, patients’ anthropometric values were not included in the current study due to the small number of patients as a preliminary study. Despite such limitations, the present study addresses the lack of data regarding the long-term developmental trajectory and clinical course of pediatric patients with MDs, as such diseases are characterized by diverse and heterogeneous feature manifestations (7, 20, 21). Even a rough estimate of the long-term developmental trajectory of children with MD could provide important information for diagnosis and development of possible intervention plans. In future study, we hope that we could overcome this limitation on the heterogeneity of MD with a larger number of patients and include prospective data as well based on the understanding of developmental course of pediatric MD from current preliminary data.

Despite the study limitations, our findings highlight the importance of DQ in the prediction and rating of the long-term functional course of children with MD. There has been no other established predictor in terms of patient’s functional measure in pediatric MD because of its difficulties of expecting the clinical course. In the perspective of this limitation, development quotient might be a candidate as a predictor or a measure in pediatric MD by giving the functional level of patients, which is a very meaningful value in pediatrics and could be obtained even from pediatric MD patients with much delayed developmental level. Our observations suggest that, upon confirmation of the diagnosis by the physician, measures, such as symptomatic treatment, general supportive care, and early administration of medicine, are encouraged to shorten the lead time to diagnosis and thereby minimize the patients’ developmental regression.

All procedures were approved by the institutional review board of Gangnam Severance Hospital in Seoul, Korea.

SE conceptualized and designed the study, carried out the data analysis, drafted the initial manuscript, reviewed and revised the manuscript, and approved the final manuscript as submitted. Y-ML conceptualized and designed the study, critically reviewed and revised the manuscript, and approved the final manuscript as submitted. Both listed authors meet the appropriate authorship criteria. No one who qualifies for authorship has been omitted from the list. Contributors and their funding sources have been properly acknowledged, and authors and contributors have approved the acknowledgment of their contributions.

The authors declare that they have no financial relationships relevant to this article. The funders had no role in study design, data collection, data analysis, interpretation of results, manuscript preparation, or decision to submit the article for publication.