Ketoacidosis at the diagnosis of type 1 (insulin dependent) diabetes mellitus is related to poor residual beta cell function. Childhood Diabetes in Finland Study Group.

The determinants of the degree of metabolic decompensation at the diagnosis of type 1 (insulin dependent) diabetes mellitus (IDDM) and the possible role of diabetic ketoacidosis in the preservation and recovery of residual beta cell function were examined in 745 Finnish children and adolescents. Children younger than 2 years or older than 10 years of age were found to be more susceptible to diabetic ketoacidosis than children between 2 and 10 years of age (< 2 years: 53.3%; 2-10 years: 16.9%; > 10 years: 33.3%). Children from families with poor parental educational level had ketoacidosis more often than those from families with high parental educational level (24.4% v 16.9%). A serum C peptide concentration of 0.10 nmol/l or more was associated with a favourable metabolic situation. Low serum C peptide concentrations, high requirement of exogenous insulin, low prevalence of remission, and high glycated haemoglobin concentrations were observed during the follow up in the group of probands having diabetic ketoacidosis at the diagnosis of IDDM. Thus diabetic ketoacidosis at diagnosis is related to a decreased capacity for beta cell recovery after the clinical manifestation of IDDM in children.     

Family characteristics and insulin dependent diabetes.

During the calendar year of 1988 a survey of new cases of insulin dependent diabetes mellitus (IDDM) in children under the age of 15 years in the British Isles was conducted. After cases had been confirmed and permission obtained to contact the families, postal questionnaires were sent to the parents of children diagnosed in England, Wales, Northern Ireland, and the Republic of Ireland. Children who developed diabetes were significantly more likely to be heavier at birth in comparison with national reference data. The diabetic children were less likely to have been breast fed, and there were more children than expected whose fathers were in nonmanual occupations. Where there was a first degree relative with IDDM there were positive correlations between the age at diagnosis of the index cases and that of their diabetic fathers and their diabetic siblings, but not their diabetic mothers. A higher proportion of children than expected who were diagnosed under the age of 5 years had fathers with IDDM. Characteristics of family members associated with IDDM in children that might provide pointers to the aetiology of the disease include heavier birth weight, method of infant feeding, the age at onset of IDDM in affected fathers and affected siblings, and the family lifestyle as defined by social class of the father.     

The severity of clinical presentation of type 1 diabetes in children does not significantly influence the pattern of residual beta-cell function and long-term metabolic control

Aim: The purpose of the present study was to compare relationships between the clinical presentation of type 1 diabetes in children and residual β‐cell secretion and long‐term metabolic control. Methods: This retrospective study was conducted in 66 diabetic children with age at diagnosis ranging from 0.7 to 14.8 yr. The patients showed contrasting characteristics at diagnosis: either diabetic ketoacidosis (DKA) (group 1, n = 29) or absence of metabolic derangement (group 2, n = 37) associated with marked (group 2A, n = 12) or mild hyperglycemia (group 2B, n = 25). A regular follow‐up was available for at least 10 yr (10–32 yr) in all cases and for 20 yr in 23 cases. C‐peptide levels were measured from diagnosis and thereafter at intervals for the first years of disease until becoming permanently undetectable. Results: C‐peptide levels at diagnosis were undetectable in about 20% of the cases both with and without DKA. C‐peptide levels at diagnosis, the duration of measurable C‐peptide levels and the maximum value found during follow‐up were not significantly different in the three groups and were not correlated with glycated hemoglobin (GHb) calculated throughout the whole period. No differences were found between the groups of patients concerning GHb values and insulin dose at 10, 15 and 20 yr of disease. The patients of group 2A, characterized by an extremely high glycemic level without ketoacidosis, had a significantly higher prevalence of HLA DR3/4 heterozygosity. Conclusions: The severity of clinical presentation at diagnosis does not significantly influence residual β‐cell function, and long‐term metabolic control.     

Descriptive Epidemiology of Non-Insulin Dependent Diabetes Mellitus Detected by Urine Glucose Screening in School Children in Japan

During the period from 1974 through 1988, we annually examined approximately 225,000 to 386,400 school children residing in Tokyo for glycosuria to detect juvenile diabetes. If the first test was positive for glucose, glycosuria was confirmed by a second test. In children who gave a positive result in both the first and second tests 0-GTT were performed. All 124 patients were diagnosed as NIDDM according to the criteria of the WHO Report on Diabetes of 1985. The incidence of NIDDM in children in Japan has increased in recent years and from 1984 to 1986 was approximately 3.8 per 100,000 per year. The frequency of NIDDM increases with age up to 14 years. In about 84% of cases, the body weight at diagnosis is more than 20% above the ideal weight and the height is often above average. There is a high frequency in families with a history of diabetes. Diet and exercise therapy in newly diagnosed patients irrespective of the presence or absence of obesity may result in remission, but some cases may require insulin therapy or oral administration of a hypogly cemic drug to obtain a better glycemic control. Children with NIDDM are more likely to be complicated by incipient retinopathy within two years after diagnosis than those with IDDM. Therefore, it is important to keep strict glycemic control to prevent diabetic complications in NIDDM children just as in juvenile onset IDDM.     

����֢ͪ���ж�������ˮ���ٴ�����

目的探讨儿童糖尿病酮症酸中毒并发脑水肿的临床特征及危险因素。方法对重庆医科大学附属儿童医院1993—2005年住院治疗的糖尿病酮症酸中毒并发脑水肿患儿的临床特征及病因进行分析,并与未并发者进行对照比较。结果在71例酮症酸中毒患儿中,有6例临床表现符合脑水肿的诊断标准,临床确定为并发脑水肿,并发率为8·4%。6例均为重型酮症酸中毒。与未发生脑水肿同等程度的重型酮症酸中毒患儿相比较,并发脑水肿患儿酸中毒更为严重,在治疗期间血钠上升缓慢及持续低钠血症,尿素氮水平升高。6例患儿中有5例应用碳酸氢盐治疗,用量大于未并发者。结论糖尿病儿童并发重型酮症酸中毒易发生脑水肿。严重酸中毒、血钠上升缓慢或持续低钠血症、血尿素氮升高及碳酸氢盐的使用有可能增加脑水肿发生的危险性。     

儿童1型糖尿病的10年回顾及白介素-10在糖尿病酮症酸中毒中的临床意义

目的：回顾浙江大学医学院附属儿童医院10年来住院儿童 1 型糖尿病的发病状况并探讨白介素-10（IL-10）在儿童 1 型糖尿病酮症酸中毒（DKA）中的临床意义。方法：对1999年1月至2009年2月在该院住院的263例334例次1型糖尿病患儿的临床资料进行回顾性分析;并对其中48例1型糖尿病患儿进行血脂、细胞因子等检查,根据有无酮症酸中毒分为 DKA组和非DKA组,24例正常健康儿童作为对照组,比较各组间血脂、细胞因子等参数的差异。结果：儿童1型糖尿病患儿中,女性多见（56.3%）,发病年龄以6~11.9岁多见。32.7% 的患儿以酮症酸中毒为就诊表现。DKA组血脂、血糖及糖化血红蛋白均高于非DKA组,二分类logistic 回归分析示上述指标水平的升高均为酮症酸中毒的危险因素。IL-10水平在DKA组明显升高,余细胞因子在DKA组和非DKA组无明显差异。糖尿病组各细胞因子水平明显高于正常对照组。结论：1型糖尿病患儿酮症酸中毒发生率较高,糖、脂代谢紊乱是酮症酸中毒的危险因素。IL-10可能为酮症酸中毒的敏感指标。［中国当代儿科杂志,2010,12（11）：849-854］     

Partial remission phase of diabetes in children younger than age 10 years

There is renewed interest in the phase of partial remission in recently diagnosed diabetes because of the potential for pharmacological and immune intervention to preserve beta cell function. 95 children younger than 10 years were investigated to assess the influence of age, sex, diabetic ketoacidosis (DKA), admission at diagnosis, and ethnicity on the frequency of remission and insulin requirements during the first two years after diagnosis. Partial remission was defined as a requirement of insulin < 0.5 U/kg body weight/day. There was partial remission in 41 patients, with no differences for children aged 2-4 years and those aged 5-9 years. None of the five children aged < 2 years remitted. Forty five of 95 children were admitted to hospital at diagnosis, of whom 26 of 45 had DKA (blood pH < 7.25). In this number of children we were unable to show a statistical difference in the rate of remission with respect to DKA, admission to hospital at diagnosis, sex, or South Asian ethnic background. There were no differences in insulin requirements between the different groups by the end of two years and at that time seven of the children required insulin < 0.5 U/kg/day. The results suggest that even in preschool children there is potential for attempting to preserve beta cell function.     

Partial remission phase of diabetes in children younger than age 10 years.

There is renewed interest in the phase of partial remission in recently diagnosed diabetes because of the potential for pharmacological and immune intervention to preserve beta cell function. 95 children younger than 10 years were investigated to assess the influence of age, sex, diabetic ketoacidosis (DKA), admission at diagnosis, and ethnicity on the frequency of remission and insulin requirements during the first two years after diagnosis. Partial remission was defined as a requirement of insulin < 0.5 U/kg body weight/day. There was partial remission in 41 patients, with no differences for children aged 2-4 years and those aged 5-9 years. None of the five children aged < 2 years remitted. Forty five of 95 children were admitted to hospital at diagnosis, of whom 26 of 45 had DKA (blood pH < 7.25). In this number of children we were unable to show a statistical difference in the rate of remission with respect to DKA, admission to hospital at diagnosis, sex, or South Asian ethnic background. There were no differences in insulin requirements between the different groups by the end of two years and at that time seven of the children required insulin < 0.5 U/kg/day. The results suggest that even in preschool children there is potential for attempting to preserve beta cell function.     

Prepubertal girls with insulin-dependent diabetes mellitus have higher exogenous insulin requirement than boys

In a population based study, the prescribed insulin dose of 348 prepubertal children with insulin-dependent diabetes mellitus (IDDM) was analysed 2 years after the diagnosis of diabetes. Girls had an insulin dose 13.6% higher than that in boys. When children younger than 5 years of age at diagnosis were analysed separately, the difference in insulin dose between boys and girls remained. The increased insulin dose in girls was not explained by possible differences in endogenous insulin secretion, body mass index, metabolic control or the number of daily insulin injections. Our observations indicate that prepubertal girls with IDDM have a poorer insulin sensitivity than boys. Received: 26 May 1997 / Accepted in revised form: 26 January 1998     

The association of poor adherence and acute metabolic disorders with low levels of cohesion and adaptability in families with diabetic children

A cross-sectional multicentre study was conducted in 165 French diabetic children aged 7-13 y and their parents. A standardized scale (FACES III) was used to determine if family cohesion and adaptability (i) differed in diabetic children's families, as compared to other families; (ii) were related to an adherence measure; or (iii) were related to metabolic control. More diabetic families than comparison families fell into the categories of disengaged with low levels of cohesion, and rigid with low levels of adaptability. Scores of cohesion and adaptability were significantly and positively correlated with both children's and parents' adherence scores, but not with HbA1c levels. Children whose families were characterized as rigidly disengaged had a significantly greater number of hypoglycaemias and six times as many episodes of ketoacidosis than the other diabetic children. Not only in adolescents, but also in families of diabetic children, family-centred interventions are needed to improve compliance and to prevent acute metabolic complications.     

Permanent neonatal diabetes mellitus: Epidemiology, mode of presentation, pathogenesis and growth

Permanent neonatal diabetes mellitus (PNIDDM) is a rare form of IDDM with unclear etiology and pathogenesis. We determined the incidence and prevalence rates and studied the clinical and biochemical features of PNIDDM in the Sultanate of Oman. The mean incidence rate during the study period from January 1989 to December 1994 was 1.788±0.82 per 100,000 live births per year. At the end of December 1994 the prevalence rate was 2.4 per 100,000 children below the age of 5 years. They constituted 41.6% of all cases of IDDM in this age group. Diarrhoea, fever, lethargy, poor feeding and failure to thrive were the most common presenting symptoms. Dehydration and tachypnoea were the most common signs. All patients who developed IDDM during the neonatal period had intrauterine growth retardation and 4.5 presented with diabetic ketoacidosis (plasma glucose 37±9 mmol/L, pH 7.12±0.1). Hypertriglyceridemia was a constant feature (19.4±4.8 mmol/L). They were products of consanguineous marriage with significantly high prevalence of IDDM and NIDDM in their family members. None of the infants had clinical or immunological evidence of congenital viral infection. Three of the five children had HLA-DR2, the diabetes resistance alleles. C-peptide secretion was absent during and after metabolic control of hyperglycemia in all the studied infants and none had circulating islet cell antibody at presentation or during the first year after diagnosis. Despite marked growth retardation at birth, there was a significant improvement of growth after initiating insulin therapy. Four of the 5 patients had normal developmental milestones, one had mild developmental delay following a severe and prolonged attack of hypoglycemia. None of the patients had exocrine pancreatic deficiency. In summary, the very high rate of parental consanguinity, occurrence in both sexes and in two siblings in the same family, absence of islet cell antibodies and the presence of HLA-DR2 loci in 3/5 of patients suggest that PNIDDM is a different disease process to standard IDDM in childhood and an autosomal recessive mode of transmission.     

Factors associated with childhood diabetes manifesting as ketoacidosis and its severity]

INTRODUCTION: Type 1 diabetes in children in France is frequently diagnosed at the stage of ketoacidosis (DKA). PATIENTS AND METHODS: A prospective study was performed in a group of 72 children (mean age = 9.4 years) at onset of diabetes, in order to determine which factors were associated to DKA and to the severity of DKA (pH < 7.10) at diagnosis. RESULTS: Younger age was related to DKA (p = 0.03), but not to its severity. A lesser frequency of DKA was found in children with a family history of insulin-treated diabetes ( p = 0.04). Misdiagnosis was more frequently observed in children with DKA than in children without DKA (p = 0.02) and in case of severe DKA at admission by comparison with non severe cases (76 vs 23%; p = 0.002). Children in low economic intake families exhibited more frequently a severe DKA (77 vs 23%; p = 0.002) and were more frequently misdiagnosed before admission (48% vs 10%; p < 0.01). Urine strips for glucose and ketone determinations were underused for diagnosis before admission (15% only). CONCLUSION: Those results underline the need to both inform physicians and ameliorate the access to health care for low social class families, in order to take up the challenge of reducing the incidence of DKA at diagnosis in diabetic children in our country.     

Partial remission phase and metabolic control in type 1 diabetes mellitus in children and adolescents

A better understanding of the remission phase, while residual beta-cell function is still present in recently diagnosed type 1 (insulin dependent) diabetes mellitus (IDDM), is very important because of the potential for pharmacological intervention to preserve this function. To evaluate the natural course and characteristics of the remission phase in children and adolescents with IDDM, a retrospective study was performed on patients diagnosed with IDDM under the age of 18 years during the years 1991-1998. Sixty-two patients whose medical records were available were included in the study. Data were collected by reviewing the hospital records of patients from the time of diagnosis through the first 24 months after diagnosis. The duration of symptoms and history of infection prior to presentation, diabetic ketoacidosis (DKA) at diagnosis, length of hospitalization, initial glucose level, basal C-peptide levels at diagnosis, daily insulin requirements per kg body weight and HbA1c at diagnosis and at each visit were recorded. Thirty-five patients (56.5%) entered partial remission. We observed similar remission rates in those aged <10 and > or =10 years at diagnosis and in boys and girls. History of infection and presentation with DKA were associated with a lower rate of remission (p<0.001, p<0.0001, respectively) and were more commonly observed under the age of 10 years (p<0.0001, p<0.0001, respectively). The average insulin requirements per kg body weight calculated at diagnosis decreased with increasing age (r = -0.31, p = 0.012). The length of time until remission was 1.36+/-1.03 (mean +/- SD) months and positively correlated with insulin requirements at discharge from the hospital (r = 0.63, p<0.0001). Mean duration of remission was 11.67+/-5.82 months and was much longer in boys than girls (p<0.05). Six patients, all boys, entered total remission for 3.80+/-3.73 months. HbA1c concentrations in the first year of the disease were significantly lower in patients who underwent a remission phase (7.31+/-1.24% vs. 8.24+/-1.47%, p <0.05). However, this difference was not observed during the second year of the disease. In conclusion, history of infection prior to presentation and DKA at diagnosis were associated with young age and were the most important factors negatively influencing the remission rate in newly diagnosed IDDM patients.     

Diabetic ketoacidosis in children

Although diabetic ketoacidosis should, theoretically, be largely preventable in patients with established diabetes, a recent report from a major US childhood diabetes center showed that children with type 1 diabetes remain at high risk for diabetic ketoacidosis, with an incidence of 8 per 100 patient-years. Children who are uninsured or underinsured, have psychiatric disorders, have poorly controlled diabetes, and live in dysfunctional families are most vulnerable. The efficacy and cost effectiveness of strategies to reduce the incidence of diabetic ketoacidosis-before diagnosis and in patients with established diabetes-are important issues for future investigation.     

A 5‐yr follow‐up nerve conduction study for the detection of subclinical diabetic neuropathy in children with newly diagnosed insulin‐dependent diabetes mellitus

Lee S‐S, Han H‐H, Kim H. A 5‐yr follow‐up nerve conduction study for the detection of subclinical diabetic neuropathy in children with newly diagnosed insulin‐dependent diabetes mellitus. Pediatric Diabetes 20XX: 00: 000–000 Objective: To investigate the changes of peripheral nerve conduction in children with insulin‐dependent diabetes mellitus (IDDM) prospectively from diagnosis and to know how those results were related to clinical risk factors. Methods: A total of 37 patients (14 males and 23 females) aged 3–19 yr (mean 12.0 ± 3.7) with newly diagnosed IDDM underwent bilateral nerve conduction studies (NCS) of median, ulnar, posterior tibial, peroneal, and sural nerves annually for 5 yr. Results: In all, 12 patients (32.4%) showed electrophysiological evidence of polyneuropathy in at least two different nerves including the sural nerve at the diagnosis of IDDM; 20 patients (54%) had multiple (≥2) abnormalities in parameters of NCS. The most common abnormal parameters at the diagnosis were conduction velocities of peroneal motor and sural nerves. In sequential NCS over 5 yr, the percentage of abnormal nerve conduction velocities rose except within the sural nerve. Poor metabolic control, height, duration of diabetes, and older age of onset were related to the changes of parameters of NCS over 5 yr. Among those risk factors, the duration of diabetes and sustained hyperglycemia affected the parameters of NCS more frequently than others. Conclusions: Children with IDDM frequently have nerve conduction abnormalities without clinical neuropathy at initial diagnosis. The frequency of abnormalities of any attribute of nerve conduction increased over the 5 yr follow‐up. The duration of diabetes and poor glycemic control proved to be more important risk factors over 5 yr as related to the development of subclinical neuropathy.     

Ketoacidosis occurring in newly diagnosed and established diabetic children

A 6-y retrospective case note review was performed to determine the causes of ketoacidosis. 135 patients and 463 diabetic years were involved. Fifty-two ketoacidosis episodes occurred: 19 episodes in new patients and 33 episodes in 19 patients with established diabetes. 27% of newly diagnosed patients presented in ketoacidosis. They were similar in terms of age, sex and proportion living in single parent families to those presenting without ketoacidosis. The 33 ketoacidosis episodes occurring in established patients included 12 episodes in 3 children who were transferred to our care because of uncontrolled diabetes. Insulin omission was the cause of ketoacidosis in 9/19 (47%) patients, and was suspected in a further 5/19 (26%). Family and school problems were common and 14/19 patients came from single parent families. Established patients aged ≥11 y were predominantly female (10F, 2M), whereas patients aged ≥10y were predominantly male (6M, 1F). 7 patients with multiple ketoacidosis episodes were all ≥11 y and 6 were female. Families with ≥2 diabetic children appeared vulnerable, 4 cases coming from 3/7 such families.     

Type 2 diabetes mellitus: incidence,management and prognosis

Type 2 diabetes mellitus in childhood emerged in the UK in about 2000, and now affects about 250 children or 1% of all childhood diabetes in the UK. It characteristically presents in an obese child, around the time of puberty, with osmotic symptoms of thirst and polyuria. Some children are identified co-incidentally, and a small proportion may present with decompensation and diabetic ketoacidosis. Acanthosis nigricans is a common feature. There is a significant female preponderance, with children from ethnic minorities disproportionately represented, and usually a history of type 2 diabetes mellitus in first-degree relatives. In contrast to type 1 diabetes, children with type 2 may have already have microvascular complications by the time they are diagnosed. The differential diagnosis clearly includes type 1 diabetes, usually distinguished by the presence of autoantibodies GAD65, ICA and IAA; but also diabetes secondary to monogenic causes, transplant and immunosuppression, and rarer syndromes. Management includes confirming the diagnosis of diabetes according to World Health Organisation criteria; screening for both microvascular complications and complications of metabolic syndrome; and initiating lifestyle, dietary and exercise advice to decrease calorie intake and increase energy expenditure. Children with osmotic symptoms or HbA1c greater than 8.5% should be commenced on insulin therapy then weaned off over 1–3 months. Metformin should also be instituted from diagnosis provided there is no ketoacidosis, and the dose increased to the maximum tolerated. Insulin is currently the only second line treatment licensed for use in the UK. A pragmatic approach is to offer once a day long acting insulin; and add in mealtime short acting insulin if insufficient response. The glycated haemoglobin target for optimal glycaemic control is less than 7.0%. Future treatments under investigation for paediatric use include the GLP-1 antagonists and DPPV4 inhibitors. It is hoped that the current batch of phase III studies will lead to an evidence base for better treatments in future.     

A randomized, controlled study of insulin pump therapy in diabetic preschoolers

OBJECTIVE: To compare glycemic control, safety, and parental satisfaction in preschool-aged diabetic children randomized to treatment either with continuous subcutaneous insulin infusion (CSII) or intensive insulin injection therapy. STUDY DESIGN: This clinical trial enrolled 42 patients <5 years of age who had been diagnosed with diabetes for at least 12 months. Children were randomly assigned to CSII (n = 21) or intensive insulin injection therapy (n = 21). Hemoglobin A1c (HbA1c) level was measured at baseline, 3, and 6 months. Secondary outcomes included severe hypoglycemic events, meter-detected hypoglycemia, blood sugar variability, body mass index (BMI), and satisfaction with therapy. RESULTS: Thirty-seven patients completed 6 months of therapy. There was a significant decrease in HbA1c during the study period for both groups (from 8.9% +/- 0.6% to 8.6% +/- 0.6% at 3- and 6-month visits). At 3 months, children using pumps had a significantly lower HbA1c than the injection group (8.4% vs 8.8%); however, by 6 months the two groups were similar (8.5% vs 8.7%). No differences in pre-meal blood sugar variabilities were seen between groups. Children on pumps had increases in the number of meter-detected episodes of hypoglycemia. Pump therapy was safe and well tolerated. No episodes of ketoacidosis occurred in either group, whereas one hypoglycemic seizure occurred in each group. Parents reported satisfaction with CSII, with 95% of families continuing on CSII beyond the 6-month study period. CONCLUSION: Pump therapy in preschool-aged children was not associated with clinically significant differences in glycemic control as compared with intensive injection therapy. The rationale for initiating CSII in this age group should be based on patient selection and lifestyle preference.     

Insulin-dependent diabetes mellitus in southern Chinese children: An overview

Thirty-three southern Chinese children with insulin-dependent diabetes mellitus (IDDM) were studied. The patients had a mean follow-up of 5.2 years (range 1.2–8.2). The mean age of onset was 8.3 years (range 1.7–13.5). The mean duration of symptoms before diagnosis was 22 days. Ten patients (30%) presented with diabetic ketoacidosis at diagnosis. Only one patient (3%) was found to have thyroid microsomal antibodies and none was found to have hypothyroidism or hyperthyroidism. Incidence and prevalence were calculated from data recorded retrospectively and prospectively in a population-based registry of IDDM. The prevalence in 1991 and 1992 was 8.3 per 100000 children under 15 years of age. The age-standardized incidence of IDDM was 1.7/100 000 per year with the 95% confidence interval of 1.2–2.4/100 000 per year for children under 15 years of age during the years 1986–93. The incidence for males was 1.1/100 000 per year and for females, 2.4/100 000 per year.     

Hyperinsulinuria in diabetic ketoacidosis

Urinary insulin excretion has been measured in diabetic children. In nine well controlled diabetics urinary insulin excretion (mean and 2SD) was significantly greater than that found in healthy controls (8.8; 4.4-17.5 versus 3.8; 2.3-6.4 microU/min/1.73 m2). In ten ketoacidotic diabetic children a 50 fold increase of the urinary insulin excretion has been observed (464.8; 158.2-1363 microU/min/1.73 m2). The hyperinsulinuria proved to be reversible: significantly decreased excretion rate (12.9; 9.7-12.3) could be measured 8-10 days after recovery from coma. The hyperinsulinuria during ketoacidosis indicates severe reversible tubular dysfunction. Recurrent ketoacidotic episodes might play a role in the development of diabetic nephropathy.