Hypoglycemia in children with type 1 diabetes: unawareness is a concrete risk

The incidence of type 1 diabetes (T1D) is increasing rapidly, especially in younger age groups. The data from Western European Diabetes Mellitus Centers suggest an annual rate increase of 3–4% in children and adolescents¹. The latest worldwide estimates show 415 million of patients with diabetes overall, that would become 642 million around 2040; there are 86,000 new cases a year of T1D among children and 542,000 patients worldwide². The incidence varies approximately 400-fold between nations, with wide variations among regions of the same state. For example, the incidence of T1D in Italy is significantly different from the observed rate in Sardinia³ (epidemiologic data in 2011: 45/100,000 new cases per year between 0 and 14 years of age), that has a trend of increase second only to Finland^2,4.

International consensus and guidelines have underlined the relevance of multiple dose insulin therapy to obtain a near physiological glycosylated hemoglobin (HbA1c) that may prevent or slow the progression of chronic complications of T1D⁵. The Diabetes Control and Complications Trial (DCCT) has demonstrated that intensive therapy delays microvascular complications in adolescents and adults and euglycemia has a great impact on the management of diabetes in the young⁶. However, a stricter control of glycemic values also increases the risk of severe hypoglycemia⁷.

Severe hypoglycemia is one of the most dangerous acute complications during insulin treatment in patients with T1D. Hypoglycemia is defined as an abnormally low plasma glucose concentration that may expose the individual to potential harm⁸. Metabolic impairment is a basic aspect in diabetes and may have severe outcomes on multiple physiologic pathways; glycemic control has a relevant role in this metabolic change and it is a daily challenge due to a great variability of glycemic profile. The brain is a glucose-dependent tissue for development, growth and remodeling: inadequate glycemic control becomes a metabolic insult that may compromise long-term cognitive functions in children, especially when severe hypoglycemia, recurrent hypoglycemia or chronic hyperglycemia occur. Although some data derives from cross-sectional observational epidemiological studies, minimizing glucose variability should be a major goal for therapy⁹.

Shaefer et al.¹⁰ have reviewed the relevant problem of hypoglycemia and awareness. They have underlined how a consistent definition of hypoglycemia is needed, and the lack of an unambiguous definition makes recognition very difficult and the risk of admission in critical health condition very high. A recent observational study¹¹ investigated the frequency of hypoglycemia unawareness via a questionnaire in a free website. The patients were divided into three categories: normal awareness, uncertain awareness and unawareness. According to the questionnaire, 23.4% of patients with T1D had hypoglycemia unawareness, 15.3% uncertain awareness and 14.1% experienced at least one episode of severe hypoglycemia. Although hypoglycemia is a critical acute complication, severe episodes and unawareness are not associated with an increased risk of all-cause mortality or cardiovascular death in patients with T1D¹².

The analysis of predisposing factors for severe hypoglycemia in T1D in pediatrics has led to controversial results¹³. The experience of hypoglycemia is very uncomfortable and the fear of hypoglycemia, mainly in adolescents and in parents of young children, can limit the achievement of good metabolic control^14,15. The identification of variables associated with severe hypoglycemia became essential. New insulin analogues and multiple injection treatment have presumably modified the epidemiology of severe hypoglycemia in pediatrics. Nevertheless, the main topic is still represented by the question: is an increased incidence of hypoglycemia an unavoidable effect of improved metabolic control? There is conflicting data on the risk factors associated with hypoglycemia: preceding episodes, newborns/toddlers or advanced age, chronic comorbidities, diet habits, glycemic monitoring and treatment regimens have been investigated¹⁰. To evaluate these aspects, our study group analyzed 195 patients in a 7.5 year prospective study referring to the T1D population-based register of Abruzzo, Italy¹⁶. We highlighted an increased risk of nocturnal hypoglycemic episodes and the lowest risk after breakfast. Moreover, we found an association between hypoglycemia and mistakes in insulin doses, strenuous physical activity, low-carbohydrate diet, and disease duration >10 years. Our overall incidence of severe hypoglycemia was lower (9.4/100 patients-year) compared to previously published data (Wagner et al.¹⁷: 31.2/100 patient-years in young children between 0 and <5 years; 19.7/100 patient-years between 5 and <7 years; 21.7/100 patient-years when >7 years; p < 0.05). A homogeneous therapy regimen with both a multiple daily insulin regimen and insulin analogues seems to reduce hypoglycemia risk despite tighter control^18,19. We have not found any association between good glycemic control (low HbA1c) and the risk of severe hypoglycemia: this evidence is in contrast with data from the DCCT and some studies on the topic^6,14 but other experimental studies have confirmed our results^20,21. We have also found a direct association between symptoms of severe hypoglycemia (accompanied by seizures) and low HbA1c levels as compared to higher values. To our knowledge, this was the first report about this relationship and it may be speculated that a counter-regulatory response, as well as hypoglycemia awareness and glycemic threshold, may be blunted in such patients.

As Shaefer and colleagues reviewed in their article, hypoglycemia unawareness needs special attention: the lack of neurologic symptoms or the inability to be aware of symptoms related to severe low brain glucose concentration is associated with a three-fold increase in the risk of death¹⁰. The effects of severe hypoglycemia on the brain have been largely analyzed and proved²². Patients with T1D present cognitive decline and a direct impact on brain structure that may be derived from diabetes duration, early onset or related complications²³. Puczynski et al. studied cognitive function after mild hypoglycemia in 24 school-age children with T1D compared to a control group of euglycemic subjects: the test scores of cognitive functions showed significant differences between the recovery rate of physical symptoms and cognitive function in children following moderate hypoglycemia²⁴. The medial temporal lobe region of the brain (hippocampus) seems to be affected by severe hypoglycemia: despite conflicting evidence about the neuropsychological consequences of this damage in T1D, severe hypoglycemia clearly impairs declarative and non-declarative memory functioning²⁵. Sachon and colleagues evaluated children with T1D and hypoglycemia using a psychometric test to find differences between those with hypoglycemia unawareness and those with awareness. Hypoglycemic episodes were associated with lower scores in psychometric tests and had deleterious effects on cognitive performances (psychomotor speed, selective attention, lexical fluency, permanent memory impairment)²⁶. Deficits of cognitive function are more probable during hypoglycemia irrespective of prior glycemic control compared to patients with hyperglycemia²⁷. Some authors have highlighted that hypoglycemic events may influence the quality of nocturnal sleep: the amount of deep sleep during nocturnal hypoglycemia was reduced and replaced by superficial sleep and arousals. However, neuropsychological test scores showed no differences between the euglycemic and hypoglycemic nights, hence the characteristics of nocturnal sleep may not have effects on neurological functions²⁸. Our study group underlined in a systematic review the significant association among nocturnal hypoglycemia and decreasing age, increasing insulin dose, insulin regimen, and higher weight standard deviation score²⁹.

The DCCT evaluated 1441 patients with T1D between 1983 and 1989, divided into intensive therapy and conventional therapy groups. In 2004, 1144 participants of DCCT were re-evaluated for a cognitive test battery through 18 year follow-up. Despite a total of 1355 episodes of coma and hypoglycemia, the authors demonstrated there was not any decline in cognitive function and performance in children and adults with T1D³⁰. Perhaps the small sample size and restrictive criteria may hamper any inference to the general population of patients with T1D. Moreover, it is difficult to distinguish negative effects of both hyper- and hypoglycemia on cognitive function, as mentioned in other studies³¹.

Shaefer et al.¹⁰ reviewed the relevance of hypoglycemia prevention and treatment: prevention helps to control the effects of hypoglycemia and several methods may contribute to prevention. Exercise is part of the lifestyle changes for patients with T1D but it is frequently associated with impairment of glycemic control. At the same time, therapy regimens with insulin or anti-hyperglycemic drugs are inevitable risk factors for hypoglycemia. New insulins such as glargine 300 (Gla-300), degludec, and insulin peg-lispro have been recently introduced and they might become a new tool to reach better glycemic control, reducing the risk of severe hypoglycemia. However, new studies should continue to strengthen the latest evidence^32–34. Indeed, there is conflicting data about their ability to reduce the risk of severe hypoglycemia or recurrence, except for limited evidence on Gla-300 (reduced frequency of episodes at any time and nocturnal hypoglycemia)³⁵. Furthermore, the current studies do not involve children and adolescence and we do not know the direct role in prevention of hypoglycemia awareness/unawareness^32,35.

Shaefer et al.¹⁰ underlined the importance of self-management, self-monitoring, appropriate selection of diabetes therapy schemes, avoiding of risk factors and a strict education program to reduce and prevent hypoglycemia¹⁰. They also highlighted the role of individualized glycemic goals, evaluation of knowledge about the use of devices, insulin types, symptoms and complications, consistent meals and a proper treatment of acute hypoglycemic events. Caregivers, teachers and co-workers should be educated about these daily problems. Some authors developed and showed in randomized controlled trials (RCTs) the efficacy of diabetes education programs; an example is a treatment and education program for people with T1D (PRIMAS), that shifts from knowledge-driven education to more self-management-oriented educational concepts. PRIMAS is also focused on empowering people to actively participate in treatment decisions and maintaining optimal compliance with recommendations. Fredheim and colleagues³⁶ demonstrated halved rates of severe hypoglycemia (overall incidence: 7.6 per 100 person-years) in their register-based population study on T1D, probably thanks to new diabetes cares and education programs. Grey and colleagues evaluated whether the internet may be a useful means to ameliorate outcomes in pre-pubertal and adolescent patients. They proposed two internet-based interventions: TeenCope, that evaluates coping skills; and Managing Diabetes, that is an education and problem-solving program. After 18 months, they found lower HbA1c values, better quality of life (QoL), social competence and self-efficacy in those patients that followed both interventions. Hence, internet programs may help to reach a larger number of youths with T1D³⁷. Sawtell et al. have analyzed psychological techniques as educational group programs and have shown positive effects of the Child and Adolescent Structured Competencies Approach to Diabetes Education (CASCADE) program for families and patients (better family relationships, knowledge and understanding of the disease, confidence and motivation to manage T1D). The training of educators and family involvement soon after diagnosis might lead to improved motivation³⁸.

New technologies, such as continuous glucose monitoring (CGM), new insulins and continuous subcutaneous insulin infusion (CSII) devices, are critical aspects of T1D management. They were also an important aspect of the abovementioned PRIMAS program that involves diabetologists and diabetes educators. The new educational materials have been evaluated for efficacy and outcomes in a RCT³⁹: participants with worse glycemic control, frequent hypoglycemic episodes or elevated diabetes distress had a better chance of clinical improvement; moreover, the PRIMAS education program preserved its efficacy in routine care compared to RCTs⁴⁰. New devices allow a tighter control of glycemic profile without an increased risk of hypoglycemia.

CGM has been a great innovation in daily management: CGM data can be used to evaluate glucose levels, the number and severity of glycemic excursions and variability. In a 6 month RCT of intensively treated individuals with T1D and HbA1c ≥7%, adults ≥25 years of age showed advantages from use of CGM compared with conventional glycemia monitoring. Moreover, patients with HbA1c <7% using a CGM had a reduction in hypoglycemia and HbA1c at the target range^41,42. The same population after 12 months of CGM presented comparable results, with a reduction in the incidence of severe hypoglycemia (from 21.8 to 7.1 events per 100 person-years)⁴³. CGM allows real-time hypoglycemia prediction: glycemic monitoring activates a predictive alarm for the risk of hypoglycemia in the following 35 minutes, causing the suspension of basal insulin delivery via automated insulin-pump. So, CGM is able to prevent from 82% to 91% of hypoglycemic events as severe immediate complications⁴⁴.

Real-time CGM may minimize glycemic variability in children with T1D: the STAR 3 study evaluated the effects of a sensor-augmented pump (SAP) on glycemic variability combined with CGM and compared with multiple daily injections. At HbA1c <8% SAP reduced glycemic variability and the frequency and severity of glycemic excursions. Moreover, CGM-SAP therapy reduced mean HbA1c in inadequately controlled T1D^45,46. In a pilot prospective study, Zalzali et al. assessed the value of SAP therapy and CGM in treatment of hypoglycemia unawareness. They observed an 80% reduction in hypoglycemia unawareness and an increase in hypoglycemia perception of 31 mg/dl as evaluated by blinded CGM, thanks to ease of use and no relevant disadvantages⁴⁷. However, even though CGM reduces the risk and frequency of severe hypoglycemia in hypoglycemia-unaware patients, it cannot restore awareness⁴⁸. Hypoglycemia and unawareness have a great impact on QoL.

Questionnaires and interviews may investigate whether patients impose changes in their lives, such as leaving employment or school, spending more time at home, being supervised by others, avoiding driving or all activities that increase risk of harm. By contrast, other patients downplay the relevance of hypoglycemic episodes increasing their risks. Frequently, caregivers and health professionals advise on clinical aspects but do not enquire about emotional and psychosocial aspects of unawareness. It has a profound impact on people’s confidence, careers and relationships: hence, healthcare professionals should pay more attention to these aspects of unawareness⁴⁹. It would appear that the patients’ education programs have provided a benefit in substantially reducing the overall frequency of severe hypoglycemia. However, if we define it as “inability to self-treat”, the adjusted incidence of severe hypoglycemia would be approximately equivalent to that observed in the intensively treated group in the DCCT study¹⁶. Intensive patient education avoids hypoglycemia, but the claim that it could eradicate severe hypoglycemia may be exaggerated by the use of a stricter definition of severe hypoglycemia and a selection bias in the recruitment of patients for studies⁵⁰. There are no RCTs in pediatrics or evidence-based therapy guidelines for a safe approach to accomplish prevention goals. Moreover, reported incidences of severe hypoglycemia vary between clinics of a multicenter study, suggesting the importance of treatment policy, process of care and patient education. Hence, large-scale surveys are needed for development of devices and treatment schemes that simulate the endocrine pancreas¹⁷.

RCTs allowed the introduction of CSII to perfect the therapeutic profile and insulin analogues: recent devices allow lower HbA1c without increased risk of severe hypoglycemia. However, only subjects who experienced benefits kept on using this approach, creating a bias⁵¹. The avoidance of hypoglycemia for several weeks reactivates autonomic responses to hypoglycemia (sympathoadrenal functions); this aim requires frequent or daily contact between healthcare professionals and patients, CGM and insulin adjustment to prevent episodes, particularly during the night⁵².

Finally, it is important to organize the daily management of diabetes-related needs in children of pre-school and school age. Edwards et al.⁵³ highlighted that health plans, school nurse support, telemedicine, training of school staff and the removal of structural, organizational, educational and attitudinal school barriers may ameliorate the impact of T1D in the short term, even though longer follow-up is required.

S. Franchini, L. Comegna, G. Prezioso and A. Blasetti
Department of Pediatrics, University of Chieti, Chieti, Italy
[email protected]

Transparency

Declaration of funding

This editorial was not funded.

Declaration of financial/other relationships

S.F., L.C., G.P., and A.P. have disclosed that they have no significant relationships with or financial interests in any commercial companies related to this study or article.

CMRO peer reviewer 1 has disclosed that he is the recipient of research/grant funding from GlaxoSmithKline, Novartis, Novo-Nordisk, Takeda, Astra-Zeneca, NIH, sanofi-aventis, Eli Lilly and Daiichi-Sankyo; is a consultant to, and lecturer for, GlaxoSmithKline, Novartis, Takeda, sanofi-aventis, Eli Lilly and Daiichi Sankyo; and is on the speakers bureau for Novo Nordisk and sanofi-aventis. Peer reviewer 2 has previously given talks and attended conferences sponsored by Astra-Zeneca, MSD, Novo-Nordisk, Angelini, Novartis, Pfizer, WinMedica and Libytec and also participated in trials sponsored by Amgen and Novartis.