Introduction

The prevalence of diabetes, and its antecedent insulin-resistant phenotype, continues to rise in children and adolescents, with an annual percentage change of 1.9% per year for type 1 diabetes (T1D), and 4.8% per year for type 2 diabetes (T2D) from 2002 to 2015; this rise is even more pronounced among racial and ethnic minorities [1, 2] and, for T2D, parallels the increase in the prevalence of obesity. In addition, compared to adults, diabetes has a more aggressive clinical course in children and adolescents, marked by a muted response to current interventions [3, 4], as well as faster loss of β-cell function, progression of insulin resistance, and development of end-organ complications [5, 6]. As a consequence, the prevalence of diabetic kidney disease (DKD) is increasing in children and adolescents, rising from 1.16 to 3.44% between 2002 and 2013 [7], particularly in youth with T2D who, compared to age-matched controls, have been noted to have a 23-fold increased risk of kidney failure and a 39-fold increased risk of dialysis [8]. Of equal concern is the observation that DKD marks a subset of people at especially high risk of cardiovascular disease and premature mortality [9, 10]. As such, diabetes and its complications, particularly DKD, continue to gain more importance for pediatric nephrologists. Here, we review recent updates to the diagnosis and management of DKD in children and adolescents.

Diagnosis

Current clinical markers

As with many other chronic kidney diseases (CKD), the diagnosis of DKD hinges on changes in urinary albumin excretion rate (AER) and glomerular filtration rate (GFR). Structural changes characteristic of DKD are observed in kidney biopsies as early as the first few years after the onset of diabetes [11], but the disease remains clinically silent for at least the first 10–15 years.

Hyperfiltration

Defined as a GFR of 120–150 mL/min/1.73m2 or > 2 standard deviations above the mean normal GFR [12], hyperfiltration is reported in 25–40% of youth with diabetes [6]. Hyperfiltration was initially attributed to early increases in kidney plasma flow and intraglomerular pressure and is believed to be a strong predictor of subsequent GFR loss and DKD progression [13]. However, the evidence supporting the association between hyperfiltration and albuminuria has been called into question more recently, in both T1D and T2D, having been observed in some studies (T1D, T2D [14]) but not in others (T1D [15], T2D [16]). Hyperfiltration has been associated with GFR loss in both T1D [17] and T2D [14], but only one study documented a decline to a GFR < 60 mL/min/1.73m2 [18], leaving open the possibility that the observed rapid GFR drop was either reversal of the early glomerular hyperfiltration or regression to the mean. Importantly, early hyperfiltration was not associated with a greater risk of subsequent progression to an estimated GFR of < 60 mL/min/1.73m2 in people with T1D in the seminal Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) study [19]. As the authors pointed out, this observation does not counter the experimental data showing a deleterious effect of single-nephron hyperfiltration and the resulting glomerular hypertension. As discussed later, current recommendations do not support angiotensin blockade purely for the management of hyperfiltration.

Albuminuria

Increases in urinary AER above normal were classically described as the earliest sign of DKD. Microalbuminuria, defined as a urinary AER 30–299 mg/day (or urine albumin-to-creatinine ratio 30–299 mg/g in spot urine specimens), occurs in 26% and 51% of children and adolescents after 10 and 19 years of diabetes, respectively [20]. Macroalbuminuria, defined as a urinary AER ≥ 300 mg/day (or urine albumin-to-creatinine ratio ≥ 300 mg/g in spot urine specimens), was reported in 14% of children with T1D after a median diabetes duration of 10 years [20]. Evidence suggests that albuminuria occurs earlier and progresses more rapidly in children and adolescents with T2D [21]. However, use of micro- or macroalbuminuria as surrogate markers for DKD has been challenged. The DCCT/EDIC participants with microalbuminuria had a much higher chance of regression to normoalbuminuria (10-year cumulative incidence of 40%) than of progression to macroalbuminuria (28%), CKD stage 3 or higher (15%), or kidney failure (4%) [22]. Macroalbuminuria was associated with a greater risk of DKD progression, but even this higher level of albuminuria led more often to normoalbuminuria (48% 10-year cumulative risk), than of progression to CKD stage 3 or higher (32%) or kidney failure (16%) [23]. In addition, while most cases of GFR loss are preceded by albuminuria, in a sizable minority (24%), GFR loss occurs in the absence of albuminuria [24], further challenging the utility of albuminuria as a key predictor of DKD progression. Despite these caveats, pediatric guidelines recommend annual screening for detection of albuminuria, using a urine sample collected in the morning, starting at puberty or 10 years of age (whichever is earlier) beginning 5 years after T1D diagnosis, and upon T2D diagnosis [25, 26] (see Table 1).

Table 1 Guidelines for screening and key interventions in children and adolescents with diabetes
Full size table

GFR loss

Also recommended is monitoring GFR, calculated from serum creatinine using estimating equations, at baseline and intermittently thereafter, based on clinical status, age, and the duration and treatment of diabetes [27]. All existing GFR estimating equations are imperfect, particularly in children and adolescents with T1D and those with normal or higher than normal GFRs [28], necessitating serial eGFR determinations using a consistent, age-appropriate and validated eGFR calculator; for youth > 18 years of age, an adult eGFR calculator should be utilized. In children with GFR < 60 mL/min/1.73m2, a 4-variable kidney failure risk equation (including age, sex, AER, and eGFR) was able to predict 1–5-year kidney failure risk, supporting the importance of eGFR and albuminuria as the predominant determinants of kidney failure risk in children with advanced CKD [29]. However, this level of CKD has thus far been less commonly seen in pediatric diabetes, though its prevalence may be rising with the increase in T2D.

In summary, our current understanding of the trajectory of DKD in children and adolescents suggests that advanced CKD and kidney failure take decades to develop after the onset/diagnosis of diabetes, which means that the data on the prevalence and time course of these outcomes in childhood-onset diabetes is largely derived from adult studies [30]. This presents a dilemma for any rigorous study of DKD in children and adolescents because understanding any aspect of DKD (biomarkers, risk factors for progression, response to interventions, etc.) has had to rely on intermediate outcomes (e.g., albuminuria, hyperfiltration) [30], which are not reliable surrogates for DKD progression, or else on extrapolation from studies in adult patients with diabetes.

Novel biomarkers

The lack of reliable surrogate markers for DKD progression during childhood and adolescence makes identification of novel markers of early disease in youth even more critical than it is in adults. Most published studies report cross-sectional associations between various urinary/serum protein biomarkers and intermediate outcomes such as albuminuria, with a smaller number of studies examining these associations using longitudinal data [31,32,33]. Rare studies are notable in bypassing the reliance on these flawed surrogate markers and examining the association between putative biomarkers such as plasma advanced glycation end-products [34] or plasma bradykinin [35] with early kidney structural changes in youth with T1D. In adults, serum tumor necrosis factor receptor 1 (TNFR1) and TNFR2 have been found to be associated with early structural changes of DKD as well as with DKD progression [36], highlighting the contribution of inflammatory pathways to the disease process. Other potential biomarkers for DKD, based on adult studies, are urinary neutrophil gelatinase–associated lipocalin (NGAL), kidney injury molecule-1 (KIM-1), N-acetyl-β-d-glucosaminidase (NAG), and liver fatty acid–binding protein (LFABP). Sometimes referred to as “kidney tubular injury markers,” these urinary proteins have been associated with DKD progression in some [37] but not in other [38] studies. None of these putative markers is currently a part of routine clinical care in adult or pediatric DKD.

Monogenic forms of diabetes

While a comprehensive discussion of this subject [39] is beyond the scope of our manuscript, it is worth mentioning that monogenic forms of diabetes, which represent an uncommon group of single-gene disorders characterized by functional defects of pancreatic β-cells, are being increasingly recognized as a result of advancements in genomic medicine. Monogenic forms of diabetes include neonatal diabetes mellitus, maturity-onset diabetes of the young, and mitochondrial diabetes, to name a few. Mutations in over 30 different genes are associated with monogenic diabetes; some of these have concomitant kidney anomalies. Awareness of monogenic forms of diabetes among pediatricians is important because of their screening and therapeutic implications, allowing for personalized medicine.

Diabetic ketoacidosis-associated acute kidney injury

A more recently noted aspect of the DKD trajectory is the recurrent episodes of acute kidney injury (AKI) and their contribution to DKD progression. In particular, AKI is a common consequence of diabetic ketoacidosis (DKA), occurring in 43–64% of pediatric patients with DKA [40]. Risk factors for DKA-associated AKI include older age, higher heart rate, higher blood urea nitrogen (BUN), higher glucose-corrected serum sodium concentration, higher glucose concentration, and lower pH, all on initial presentation [41]. In addition, a prior episode of DKA-associated AKI substantially increases the risk of subsequent similar episodes, suggesting that either children vary in their risk for AKI or that previous episodes of AKI increase kidney susceptibility to subsequent injury [41]. DKA frequently predisposes to AKI by volume depletion and renal hypoperfusion due to osmotic diuresis and occasionally gastrointestinal losses, contributing to a pre-renal picture. However, the greater severity of AKI in some children (stage 2 or 3) suggests the possibility that kidney injury has progressed from merely pre-renal to acute tubular injury [42].

In adults with diabetes, AKI is associated with a greater risk of developing DKD, and this risk increases with each AKI episode [43]. Compared to adults with DKA who do not experience AKI, those who do have a more rapid progression of DKD as well as a higher long-term mortality [44]. Furthermore, the rate of CKD progression is proportionate to the severity of the DKA-associated AKI [44]. Mechanisms driving the transition from AKI to CKD may include the development of tubulointerstitial fibrosis following proximal tubular injury, glomerular endothelial dysfunction, oxidative stress, ongoing inflammation, and cell cycle arrest [45]. In children, data regarding the association between DKA-associated AKI and DKD progression are currently lacking.

Management

Addressing risk factors that promote the development of DKD

The seminal DCCT/EDIC study highlighted hyperglycemia as the primary causal mechanism for diabetes-related complications in children and adolescents based on the reduction of microvascular complications by intensive diabetes treatment in 195 DCCT participants, who were 13–17 years old, similar to what was noted in the entire cohort [46]. Based on this study, the HbA1c target has been set at < 7.0–7.5% for patients with T1D and T2D by national and international societies [25, 26]. However, in determining glycemic targets for diabetes treatment, one consideration is worth noting. Several large observational studies have found that the glycemic threshold with the lowest risk of end-organ complications was at an HbA1c nadir of 5.0–5.5% [47, 48]. However, while lowering HbA1c to 7% led to a reduction of complications [49], no interventional trial has shown improvement in clinical outcomes by lowering HbA1c below 7 (e.g., VADT [50], ADVANCE [51]) and some have shown harm (ACCORD [52]). An explanation for this apparent dichotomy between observational and interventional studies may be that the optimal glycemic target is determined by a balance of the benefits vs. risks for any given intervention. While a naturally occurring HbA1c of 5.0–5.5% may be associated with the lowest risk of complications, when HbA1c is pushed from 7 to 5.5% using more intensive diabetes treatments, these higher risks (e.g., hypoglycemia, weight gain) may negate any clinical benefits from the lower HbA1c threshold. On the other hand, with the use of treatments that have lower risks such as hypoglycemia (e.g., metformin or SGLT2 inhibitors) targeting a lower HbA1c threshold may be desirable to achieve improved outcomes. As such, the optimal glycemic target is likely not a fixed threshold, but one that varies based on the balance of the risk vs. benefit of each specific intervention used to achieve that glycemic threshold. Table 1 outlines recommended guidelines in the management of children and adolescents with T1D and T2D.

Previously a feature of T2D, but becoming more prevalent in both T1D and T2D, insulin resistance (IR) has been strongly linked to the development of DKD. IR can initiate kidney injury independently of [53], and even prior to progression to [54] frank hyperglycemia. IR is well-known to be augmented by the growth and hormonal changes that occur during puberty. For example, IR is prominent in adolescents with T1D even in the absence of adiposity [55]. The rise of obesity in youth has further added to the puberty-induced increase in IR resulting in rapid progression of IR to diabetes in this population, as shown in the TODAY [56] and RISE [57] studies.

Hypertension, a well-established risk factor for the development of DKD [58], increases in incidence rapidly in youth with diabetes, particularly T2D [56]. This rise is unaffected by glycemic control and requires multiple medications in a sizable fraction of patients, suggesting that obesity and diabetes contribute significantly to treatment-resistant hypertension [56]. Current guidelines [59] recommend that pharmacotherapy be instituted promptly, along with lifestyle measures, for treatment of hypertension (BP > 95th percentile). Recommendations are also that first-line agents in people with diabetes are angiotensin-converting enzyme inhibitors (ACEis) or angiotensin receptor blockers (ARBs), following the provision of reproductive counseling to women of childbearing age. The target BP in all patients should be less than the 90th percentile. However, in general, hypertension often goes underdiagnosed in children, making this a critical target for research and innovations to improve recognition of this important risk factor for DKD.

Serum uric acid (SUA) has long been considered a likely causal risk factor for DKD in both adults and adolescents [60]. Furthermore, SUA reduction in small clinical trials was associated with a slower decline in GFR in patients with CKD [61], raising hopes that targeting SUA may yield a novel strategy for slowing DKD progression. Unfortunately, two recent clinical trials targeting SUA reduction with allopurinol (PERL [62]) and febuxostat (FEATHER [63]) showed no impact on DKD/CKD progression.

Dietary and lifestyle interventions

Current standards of care include dietary education, e.g., carbohydrate counting, calorie tracking and consumption of low glycemic index foods [25], and promotion of increased physical activity, all of which are associated with improved glycemic control [64]. However, these strategies continue to have limited success. For example, only 20% and 30% of adolescents with T2D limit high-fat foods and use carbohydrate counting, respectively [64]. Dietary recommendations in youth must also consider the high prevalence of disordered eating behavior in youth with diabetes [65], itself associated with worse glycemic control and greater adverse outcomes. The American Diabetes Association (ADA) guidelines also recommend weight control, particularly for youth with T2D [25]. However, intensive lifestyle intervention programs that combine diet and exercise goals have thus far been unsuccessful in promoting significant weight loss or improvement in glycemic control in adolescents with T2D [66].

More recently, the notable success of DiRECT (Diabetes REmission Clinical Trial) has re-ignited interest in the dietary approach to diabetes management, particularly with use of very low-energy diets (VLED). Obese adults with recent-onset T2D, fed a very low calorie (825–835 kcal/day) meal replacement diet for 3–5 months, followed by structured re-introduction of food and monthly support, achieved an astounding 46% diabetes remission rate at 12 months, with a decline in remission rate to 36% at 24 months [67]. While these results have generated significant excitement as well as a recommendation for VLED utilization in some patient populations, their safety and efficacy in adolescents have not been well-studied to date. A recent systematic review and meta-analysis of VLEDs in children and adolescents demonstrated weight loss and improved cardio-metabolic outcomes, with several caveats [68]: VLED safety could not be assessed because adverse events were inadequately described in most of the analyzed studies. Furthermore, of the 24 identified studies, only four were published since the year 2000, only two were randomized controlled trials, and only two focused on patients with a diagnosis of T2D [68]. As such, despite their potential long-term benefits, inclusion of VLEDs in the current management of diabetes in children and adolescents awaits support from larger and controlled trials and cannot be recommended at this time due to the potential for adverse effects in this highly vulnerable population where growth and neurodevelopmental considerations are of utmost importance.

Pharmacological management

There have been significant advances in pharmacologic treatments for adults with diabetes over the past few decades [69, 70]. Several glucose-lowering medications from three novel classes have been approved in the USA for use in adult patients with diabetes: glucagon-like peptide-1 (GLP1) receptor agonists, dipeptidyl peptidase-4 (DPP4) inhibitors, and sodium glucose cotransporter-2 (SGLT2) inhibitors [71], many of which have beneficial effects on kidney outcomes [69]. However, children and adolescents with diabetes are yet to benefit from these advances with primary diabetes treatments in children being limited to metformin and insulin [72]. While metformin is an effective glucose-lowering agent without the risk of hypoglycemia, it does not substantially preserve β-cell function such that over time glycemic function tends to worsen in patients. In spite of early fears related to the association between metformin and lactic acidosis in patients with CKD, data to support this complication have been inconsistent such that the FDA revised its warning regarding metformin use in patients with CKD who have an eGFR > 45 mL/min per 1.73 m2; dose adjustments are required with a decline in eGFR and there are currently no safety data for metformin use in patients with an eGFR < 30 mL/min per 1.73 m2 or in those on dialysis [73].

To date, only one novel drug, a GLP1 receptor agonist, has been approved for children and adolescents. Given the rising prevalence of diabetes as well as DKD [7, 8] in children and adolescents, the increase in lifetime adverse outcomes and premature mortality in this population, and the salutary effects of some of the novel diabetes medications on adverse cardiorenal outcomes in adults, it becomes critical to prioritize studies needed to furnish data on the safety and efficacy of these agents in children and adolescents. The current limited and small pediatric trials of these agents have been conducted only pursuant to the pediatric rule, an FDA mandate that all medications approved for use in adults must also undergo safety and efficacy testing in pediatric patients. As such, pediatric participant numbers are miniscule (≤ 300) (Tables 2 and 3), in stark contrast to the respective adult trials that have enrolled thousands of participants. Furthermore, few results from these pediatric trials have been published in peer-reviewed journals.

Table 2 Clinical trials of GLP1 receptor agonists, DPP4 inhibitors, and SGLT2 inhibitors in children and adolescents with type 1 diabetes. List of all ongoing (in italics) and completed trials of GLP1 receptor agonists, DPP4 inhibitors, and SGLT2 inhibitors in children and adolescents < 18 years in the USA. Participant numbers listed in parentheses indicate target enrollment numbers for ongoing trials
Full size table
Table 3 Clinical trials of GLP1 receptor agonists, DPP4 inhibitors, and SGLT2 inhibitors in children and adolescents with type 2 diabetes. List of all ongoing (in italics) and completed trials of GLP1 receptor agonists, DPP4 inhibitors, and SGLT2 inhibitors in children and adolescents < 18 years in the USA. Participant numbers listed in parentheses indicate target enrollment numbers for ongoing trials
Full size table

GLP1 receptor agonists for T2D

GLP1 receptor agonists mimic the incretin system by stimulating glucose-dependent insulin secretion, inhibiting glucagon secretion and gastric emptying, and suppressing appetite, leading to a reduction in blood glucose levels [85] and some weight loss. The use of GLP1 receptor agonists is associated with a 17% reduction in risk of composite kidney outcomes including incident macroalbuminuria, reduced eGFR, progression to dialysis, or death from kidney causes, without increasing the incidence of hypoglycemia [85]. Six GLP1 receptor agonists have been approved in the USA for the treatment of T2D in adults in the past 5–10 years; only one (liraglutide) has been approved for use in patients age ≥ 10 years with T2D. A small 5-week dose escalation study in youth showed that liraglutide had a similar pharmacokinetic profile as that in adults with no serious adverse events in that short follow-up period [77]. It was also shown to improve glycemic control in youth with T2D, in conjunction with metformin [76], leading to FDA approval in June 2019. The Evaluation of Liraglutide in Pediatrics with Diabetes (ELLIPSE) trial in predominantly white, obese adolescents with T2D showed that the addition of liraglutide to metformin improved glycemic control [76]. A 3-month study examining liraglutide efficacy in 100 black youth and young adults with T2D (NCT02960659) is to be completed by the end of 2022. It is worth noting that FDA approval of liraglutide in children and adolescents is based on two trials involving a total of 156 participants, in contrast to adult trials with thousands of participants (e.g., n = 9,340 for liraglutide in LEADER).

GLP1 receptor agonists for T1D

No GLP1 receptor agonists are approved for use in children with T1D. The studies that have been completed thus far, and the one that is ongoing, are all small (< 40 participants) and have largely remained unpublished (Table 2).

DPP4 inhibitors for T2D

Like GLP1 receptor agonists, DPP4 inhibitors act on the incretin system. However, DPP4 inhibitors act to prevent the degradation of incretins by inhibiting the DPP4 enzyme [86]. This leads to an increase in postprandial GLP1 which stimulates insulin secretion, decreasing plasma glucose concentration. Unlike GLP1 receptor agonists, DPP4 inhibitors do not affect appetite or gastric emptying and are administered orally, rather than via the subcutaneous route [86]. DPP4 inhibitors provide effective glycemic control in patients with T2D but do not confer cardiovascular protection and potential benefits on kidney outcomes are controversial [87]. None has yet been approved for use in children or adolescents.

A dose-dependent reduction in mean HbA1c was noted in a small 12-week study of linagliptin, a DPP4 inhibitor, in 37 adolescents with T2D compared with placebo, but this difference was not statistically significant. There were no serious adverse events in the intervention group [80]. The ongoing DINAMO study (NCT03429543) is currently assessing the longer-term safety and efficacy of once daily linagliptin over 52 weeks and is estimated to be completed in 2023. The study aims to enroll 186 participants aged 10–17 years with T2D and BMI ≥ 85th percentile for age and sex who will be randomized 1:1:1 to placebo, linagliptin, or empagliflozin (discussed in more detail below).

Similarly, trials with other DPP4 inhibitors have not shown statistically significant improvements in HbA1c. None of these trials has made it to publication, likely due to the very small population size. Additional larger trials are underway (Table 3).

DPP4 inhibitors for T1D

Even in adults, DPP4 inhibitors are currently only approved to treat T2D, while their efficacy in adults with T1D is being investigated [88]. In adolescents with T1D, the DPP4 inhibitor sitagliptin did not lead to significant improvement in 2-h c-peptide response following a meal challenge [75]. Another study (NCT01718093) in 21 adolescents investigating the glucose-lowering effects of adding sitagliptin, metformin, or both, to an insulin regimen, was completed in 2015 but results have not yet been reported.

SGLT2 inhibitors for T2D

SGLT2 inhibitors block sodium and glucose reuptake in the proximal tubules of the kidney, resulting in increased urinary sodium and glucose excretion leading to improvements in metabolic and hemodynamic parameters, including hyperglycemia, body weight, adiposity, and BP [89]. In addition, SGLT2 inhibitors have been shown to attenuate renal hyperfiltration in subjects with diabetes, by affecting tubular-glomerular feedback mechanisms. The cardiorenal benefits of SGLT2 inhibitors in adults have been well-established in dedicated cardiovascular outcome trials. There is also data showing SGLT2 inhibitors significantly lower the risk of kidney adverse outcomes, such as sustained decline in eGFR, need for dialysis, or death from kidney causes in adults [90, 91]. SGLT2 inhibitors are associated with an increased risk of genital infections [90, 91]. In the past decade, four SGLT2 inhibitors have been approved in the USA for adults with T2D and are highly recommended for patients with T2D and kidney disease to slow progression to dialysis [71]; however, no SGLT2 inhibitor has been approved for use in adolescents.

One study (NCT02725593) assessed the efficacy of dapagliflozin, an SGLT2 inhibitor, vs. placebo in reducing mean HbA1c in 72 adolescents and young adults. To our knowledge, trial results have not been published in a peer-reviewed journal. However, data submitted to clinicaltrials.gov indicate that there was no significant difference in the change in HbA1c or fasting plasma glucose from baseline to week 24 between dapagliflozin and placebo. Another study assessed the pharmacokinetic and pharmacodynamic profiles of various doses of empagliflozin, another SGLT2 inhibitor, in 27 adolescents (mean age 14.1 years) and found a mean decrease in fasting plasma glucose, a dose-dependent increase in urinary glucose excretion, and no serious adverse events [83].

Phase 3 trials for dapagliflozin (NCT03199053) and empagliflozin (NCT03429543) in children and adolescents are ongoing and estimated to be completed in 2023. Both trials are following a similar design, randomizing an estimated 243 and 186 participants, respectively, 1:1:1 to placebo, an SGLT2 inhibitor (dapagliflozin or empagliflozin), or a DPP4 inhibitor (saxagliptin or linagliptin). Both trials will assess efficacy in change in HbA1c from baseline at week 26 and will continue to follow participants to week 52. Several other SGLT2 inhibitors are currently in various phases of clinical trials in adults and adolescents with T2D (Table 3).

SGLT2 inhibitors for T1D

At present, SGLT2 inhibitors are only approved for use in adults with T2D but their efficacy in treating T1D is currently being investigated [88]. Only one trial is currently investigating the use of an SGLT2 inhibitor in adolescents (NCT04333823) and, notably, it is the only trial of a new diabetes agent in adolescents with a kidney-related primary outcome. The Adolescent Type 1 Diabetes Treatment with SGLT2i for hyperglycEMia and hyPerfiltration Trial (ATTEMPT) plans to recruit 100 adolescents (age 12–18 years) with T1D for at least 1 year; participants will be randomized to receive dapagliflozin or placebo once daily for 16 weeks. The primary outcome measure of the trial is change in measured GFR from baseline to week 16, with the goal of understanding the physiological effects of SGLT-2 inhibition on early-onset diabetes complications in youth. This study is estimated to be completed in 2023.

Surgical intervention

Bariatric surgery has profound, rapid, and long-lasting effects on improving weight loss, cardiovascular risk factors, and glycemic control in adults with T2D, leading to remission of diabetes in over half of all patients after 2 years [92]; surgery has also been shown to reduce the incidence of albuminuria and slow DKD progression [93].

Bariatric surgery is becoming a more accepted treatment option for morbidly obese youth. The Teen-Longitudinal Assessment of Bariatric Surgery (Teen-LABS) study enrolled 242 adolescents undergoing bariatric surgery into a prospective observational study [94]. The most common surgical procedure was Roux-en-Y gastric bypass (66.5%), followed by sleeve gastrectomy (27.7%) and adjustable gastric band (5.8%). Teen-LABS demonstrated that bariatric surgery led to a significant decrease in BMI within 6 months, and that this weight loss was maintained at 3 years. This was associated with significant improvements in eGFR and albuminuria at 3 years [94]. A secondary analysis found that adolescents were 27% more likely than adults to achieve remission of T2D, and 51% more likely than adults to achieve remission of hypertension 5 years after bariatric surgery [95].

Further evidence suggests that bariatric surgery has greater advantages over current standard-of-care lifestyle and pharmacologic options for adolescents with T2D. A secondary analysis compared surgical vs. medical management of T2D in adolescents by matching 30 Teen-LAB participants at the time of surgery with 63 adolescents from the TODAY study. Participants were matched for age, sex, race/ethnicity, and baseline BMI, irrespective of the TODAY treatment group. After 5 years of follow-up, Teen-LABS participants had lower BMI, mean HbA1c, mean triglycerides, and improved insulin sensitivity, whereas TODAY participants had higher BMI, HbA1c, and triglycerides, and worsened insulin sensitivity [96]. Medical management of T2D in adolescents was also associated with poorer kidney outcomes. After 5 years of follow-up, TODAY participants had 27-fold greater odds of elevated urinary AER compared with Teen-LABS participants. However, no randomized controlled trials have done a head-to-head comparison of the effectiveness and safety of surgery to non-surgical conventional treatment options, and therefore, the decision to pursue surgical intervention needs to be individualized and tailored to the unique needs of the patient and the available expertise at the treating institution. In addition, surgical interventions in this young population would require a thorough evaluation of their risks vs. benefits over a much longer follow-up. In general, the guidelines used as an indication for metabolic surgery in adolescents include BMI > 35 kg/m2 with comorbidities or BMI > 40 kg/m2 with or without comorbidities.

Transition of care

The period of transition from pediatric to adult care is one associated with deterioration in glycemic control; increased occurrence of acute complications; psychosocial, emotional, and behavioral challenges; and the emergence of chronic complications [97, 98]. Moreover, the process of transition from pediatric to adult care is prone to lead to fragmentation in health care delivery, which may adversely impact health care quality, cost, and outcomes [99].

ADA guidelines recommend that providers prepare youth for transition to adult health care in early adolescence and, at the latest, at least 1 year before transition occurs [25, 98]. During this time, both pediatric and adult diabetes care providers should provide support and resources for transitioning youth. Given the variability among youth in gaining independence and managing both the logistical and medical aspects of diabetes care, the formal transfer of care should only occur when it is deemed appropriate by the patient and the provider.

Strategies to better prepare youth for transition include a directed focus on diabetes self-management skills for the emerging adult with a gradual transfer of diabetes care responsibilities to the teen from the parent or guardian, sharing of information about the differences between pediatric and adult providers in their approaches to care, and education regarding health insurance options and how to maintain coverage. Ideally, the patient and future diabetes provider should also receive a written summarization of all factors related to the patient’s diabetes management [98]. For more details, the reader is referred to the ADA Transitions Workgroup report [98]. While currently there are limited evidence-based strategies for optimal transition of care for youth with diabetes, this is an area of great importance and one that is receiving increased attention; we are hopeful that results of clinical trials such as the “Evaluating Innovations in Transition From Pediatric to Adult Care—The Transition Navigator Trial” (NCT03342495) will help provide a framework to allow the provision of more seamless and high-quality care to such youth.

Conclusions

In summary, diabetes and DKD are increasingly encountered in children and adolescents. At the same time, our understanding of the pathophysiology of the disease has also increased, as has the armamentarium of therapeutic strategies available to us. Most promising are the three novel classes of mediations which have thus far been largely studied in adult patients: GLP1 receptor agonists, DPP4 inhibitors, and SGLT2 inhibitors, as well as dietary approaches to reverse diabetes and its associated complications (e.g., VLEDs). Many of these interventions have the potential to reduce end-organ damage. Considering the increasing prevalence of diabetes in children and young adults and the more aggressive phenotype of diabetes in this highly vulnerable population, it is time to refocus our efforts and resources on this area by conducting larger and more rigorous clinical trials of these novel therapeutic agents in children and youth with diabetes.

Key summary points

  1. (1)

    The incidence of diabetes, particularly type 2 diabetes, and its complications, are on the rise in children and adolescents, disproportionately affecting racial-ethnic minorities.

  2. (2)

    The cornerstone in the prevention of diabetic kidney disease is optimal glycemic control, along with screening for and management of hypertension and albuminuria.

  3. (3)

    GLP1 receptor agonists, in conjunction with metformin, have been shown to have a beneficial effect in reducing the incidence of adverse kidney outcomes, and are now approved for use in older children with type 2 diabetes.

  4. (4)

    While many new therapies have been studied and approved for use in adults with diabetes and diabetic kidney disease, insufficient progress has been made in performing clinical trials in children and young adults.

Multiple choice questions

  1. 1.

    Which of the following statements accurately describes the relationship between albuminuria and progression of DKD?

    1. a)

      Progression of DKD is universally preceded by the development of albuminuria.

    2. b)

      Most pediatric patients with microalbuminuria progress to macroalbuminuria.

    3. c)

      Macroalbuminuria is associated with a higher risk of progressive CKD.

    4. d)

      Macroalbuminuria invariably progresses to advanced CKD.

  2. 2.

    Which of the following reflects the cumulative incidence of CKD stage 5 in pediatric patients with T1D, 30 years after the onset of disease?

    1. a)

       < 10%

    2. b)

      10–25%

    3. c)

      25–30%

    4. d)

       > 30%

  3. 3.

    When should screening for albuminuria start in children and adolescents with T2D?

    1. a)

      At the time of diagnosis

    2. b)

      After 1 year of diagnosis

    3. c)

      After 5 years of diagnosis

    4. d)

      When diabetic retinopathy is first diagnosed

  4. 4.

    Which of the following statements regarding the development of AKI during DKA is true?

    1. a)

      AKI is a rare occurrence in children with DKA.

    2. b)

      The most common cause of AKI in DKA is nephrotoxic ATN.

    3. c)

      Adults with DKA who develop AKI have a more rapid progression of DKD.

    4. d)

      Younger age is associated with a higher risk for children with DKA to develop AKI.

  5. 5.

    Which of the following is the most important risk factor associated with the development of DKD in patients with diabetes?

    1. a)

      Hypertension

    2. b)

      Insulin resistance

    3. c)

      Serum uric acid

    4. d)

      Hyperglycemia