Good diabetic control - a study in mass delusion

Safety and efficacy of blood glucose management practices at a diabetes camp

AIM

Camps are an important part of diabetic management in children yet data on the safety and efficacy of camps are limited. We assessed the safety and efficacy of blood glucose management guidelines at summer camps for diabetic children.

METHODS

Consistent management guidelines were implemented during 10 consecutive diabetes camps held in the same facility between 1998 and 2002. Using the entire sample of campers aged 9-13 years, we analysed insulin dosage alterations, the frequency of hypoglycaemia (<4 mmol/L), hyperglycaemia (>15 mmol/L) and ketosis and evaluated our overnight management guidelines. The effects of sex, year, age, insulin regimen and duration of diagnosis on hypoglycaemia frequency were determined.

RESULTS

Mean insulin doses decreased 19.2% (95% confidence interval 16.9-21.6%) by the last day of camp (day 6) relative to the day prior to camp. Mean blood glucose levels were 11.4 mmol/L before breakfast and the main evening meal, 11.3 mmol/L before bed, 10.8 mmol/L at midnight and 9.4 mmol/L at 3 am. Of the 10 839 readings analysed, 984 (9.1%) were below 4 mmol/L (0.5 per camper/day) with no clinical grade 3 (seizure or coma) hypoglycaemia. Hypoglycaemia frequency was independent of sex, year, age, insulin regimen and duration of diagnosis (all P > 0.05). There were 2570 (23.7%) readings above 15 mmol/L (1.4 per camper/day) but only 42 (0.4%) were associated with significant ketosis.

CONCLUSION

Children at diabetes camps experience considerable blood glucose variability; however, the careful application of monitoring and management guidelines can avoid serious adverse events.

Monitoring glycemia in diabetes. Short-term assessment

The monitoring of glycemic status is considered a cornerstone of diabetes care. This article reviews current recommendations for routine glycemia monitoring, with emphasis on practical applications. A description of the newly developed National Glycohemoglobin Standardization Program also is provided.

Compliance of young diabetics with health regimens

The compliance of young patients with diabetes regimens is explored by means of a questionnaire filled in by both the patients and their nurses and compared with the patients' glycosylated haemoglobin (HBA1) levels. The structured questionnaire was completed by 47 diabetics aged between 15 and 17 years. The adolescents had very similar views to the specialist nurse on their compliance with health regimens, and these assessments were also in line with the HBA1 values. One-third of the young patients showed a high degree of compliance with diabetes regimens, while just under a half showed average compliance; the remainder failed to comply adequately. Compliance was highest in the case of insulin treatment, self-care and co-operation with the nursing staff, and poorest in following dietary and home monitoring instructions.

Monitoring of diabetes in children

There is increasing evidence to show that a high degree of metabolic control in diabetes mellitus delays and reduces the severity and incidence of the microvascular complications of diabetes. The goals of appropriate therapy for diabetes should hence include an all-out effort to achieve as near normal metabolism as possible but it should be individualized accordingly. In actual practice it is difficult to achieve "good" control especially in diabetic children without attendant hypoglycaemia and its sequelae. Diabetic control is difficult to define in clinical terms, and the maintenance of metabolic control in diabetes involves a complex interaction of pathological, physiological, psychological, familial, social and environmental variables. The aims in insulin-dependent diabetic children is to ensure optimal emotional and physical health of the child or adolescent. The various modalities available for monitoring or assessment of diabetic control shall be discussed.

The effects of pancreas transplantation on the glomerular structure of renal allografts in patients with insulin-dependent diabetes

The microvascular complications of diabetes mellitus may be caused, in part, by poor glycemic control. Diabetic patients who have received renal allografts may have new glomerular lesions that are manifested structurally by increases in mesangial and glomerular volume. Successful pancreas transplantation produces long-term normoglycemia and provides a unique opportunity to evaluate the impact of the normalization of the blood glucose level on the development of the renal lesions typical of diabetes mellitus in transplanted kidneys. We obtained biopsy specimens from the functioning renal allografts of 12 patients with insulin-dependent (Type I) diabetes before successful pancreas transplantation (performed one to seven years after renal transplantation) and repeated the biopsy at least 1.9 years later. In renal biopsy specimens obtained before pancreas transplantation, the mesangial volume was normal or modestly increased and the glomerular basement membrane was moderately thickened. At follow-up, no progression could be detected in any structural measure in the glomerulus. Furthermore, the recipients of pancreas transplants had smaller glomerular volumes than 13 matched diabetic patients who were recipients of renal allografts but who did not undergo pancreas transplantation (mean +/- SD, 1.80 +/- 0.55 vs. 2.47 +/- 0.73 x 10(6) microns 3; P = 0.02) and showed markedly less mesangial expansion (mesangial-volume fraction, 0.19 +/- 0.07 vs. 0.31 +/- 0.10 microns 3 per cubic micrometer; P = 0.004). We conclude that successful pancreas transplantation is associated with significantly less severe diabetic glomerulopathy in kidneys previously transplanted into diabetic patients. These data support the hypothesis that normoglycemia can prevent the progression of diabetic glomerulopathy in humans.

Home blood glucose monitoring in diabetic children and adolescents. A 3-year feasibility study

In order to elucidate the question whether blood glucose monitoring should replace glucosuria testing in childhood diabetes 160 diabetic children and adolescents were invited to participate in a feasibility study on home blood glucose testing. Seventeen girls and 15 boys with an age of 4-21 years and duration of diabetes for 0.3-18.7 years accepted, thus a selection of motivated patients. They performed 20-22 diurnal blood glucose profiles, each consisting of 7 blood samples, during a 3 month period. Thereafter, all patients were encouraged to continue blood glucose self-control and the actual performance of the 32 patients was evaluated 3 years later. Daily glucosuria tests were also made and HbA1 was analysed. Patients' attitudes were evaluated through 2 questionnaires. The study shows that blood glucose monitoring is feasible in the actual age groups. Most patients were positive towards blood tests, particularly because it gave an immediate answer to an actual problem, but its introduction did not change the metabolic control. However, pain restricted its daily use and only 6.4% of the patients preferred blood testing to urinalysis for long term use. Furthermore, the correlation between home glucosuria and HbA1 was as good as between home blood glucose and HbA1. It is concluded, that blood glucose self-monitoring is a valuable tool in the management of childhood diabetes, but that it should be regarded as a complement to and not a substitute for daily home urinalysis.

Failure and efficacy of insulin therapy in insulin dependent (type I) diabetic patients

In order to determine the degree of metabolic control (HbA1c [normal less than 5.8%], mean blood glucose [MBG], glucosuria and lipids) and the prevalence of late diabetic complications in insulin-dependent diabetic patients treated by conventional insulin therapy both patients of a diabetes center (DC: n = 130; age 37.1 +/- 1.4 years) and a rural area (RA: n = 73; age 38.4 +/- 2.4 years) were examined within their local setting. Eighty such insulin-dependent diabetic patients were also taught a technique of near normal glycemic insulin substitution (NIS), which separates basal from prandial insulin replacement and instructs the patients to immediately correct self-controlled (3.8 +/- 0.1/day) aberrant blood glucose values. None of the groups on conventional insulin therapy was able to achieve satisfactory metabolic control or to avoid late diabetic complications, but rural patients were even worse off (BG 240 +/- 10 mg/dl; HbA1c 8.7 +/- 0.2% [normal: 3/73 = 4%]) than those of the DC (MBG 191 +/- 5 mg/dl; HbA1c 7.1 +/- 0.2% [normal: 27/130 = 21%]), while the prevalence of late diabetic complications was almost identical (RA/DC: neuropathy 22%/25%; retinopathy 41%/38%; macroangiopathy 15%/13%; but proteinuria 14%/5.4%). Metabolic control was improved by NIS with twice daily injections of basal (long acting) and separately of prandial (regular) insulin (total: 4.8 +/- 0.1 injections/day; MBG 130 +/- 2 mg/dl; HbA1c 5.8 +/- 0.1% [normal: 41/80 = 51%]. We conclude (1) that conventional insulin therapy just prevents metabolic catastrophe but in more than 79% of insulin-dependent diabetic patients lacks the ability to provide good metabolic control, while (2) NIS, a more physiological form of insulin therapy, improves this deplorable situation 5- to 12.4-fold.

Bedtime insulin injections: an alternative regimen

Sixteen children (aged 3 to 12 years) participated in a 12 month crossover study comparing bedtime with teatime insulin injections in an endeavour to reduce morning hyperglycaemia. Blood glucose values were lower at lunch and at teatime on the later injection, but higher at bedtime and midnight. There was no overall change in glycosylated haemoglobin. Despite more frequent mild hypoglycaemic attacks, parents preferred the convenience of the later injection. Analysis of individual children's glycosylated haemoglobin values showed that those whose metabolic control improved on the later injection were younger and went to bed earlier, indicating that this regimen may have a place in the management of younger children with diabetes mellitus.

Insulin infusion pump treatment of Type I diabetes

This article examines the background studies that led to the use of portable infusion pumps for continuous administration of insulin and considers the applicability of pump treatment in current pediatric practice.

Proteinuria in children with insulin-dependent diabetes: relationship to duration of disease, metabolic control, and retinal changes

The relationship of early retinal changes and subclinical proteinuria to duration and metabolic regulation of insulin-dependent diabetes was studied in 67 children. Retinopathy was found in 25 patients and occurred almost exclusively (96%) in those with duration of disease longer than five years. Glomerular filtration rate was normal or increased in all patients. Urinary excretion of beta 2-microglobulin, albumin, transferrin, and IgG was significantly increased in patients, as compared with controls, whereas serum concentrations of these proteins were generally normal. The mechanisms responsible for the hyperexcretion of both large and small proteins are unclear but probably involve both glomerular and tubular dysfunction. Increased urinary protein excretion occurred independently of duration of disease. Retinopathy but not microproteinuria was more common in patients with glycosylated hemoglobin greater than 11% and in those with duration of disease longer than five years. Although a significant association was found between retinopathy and the hyperexcretion of one or more of the large molecular weight proteins, the weight of the evidence suggests that these two sequelae of diabetes differ in their pathogenesis. Long-term follow-up of these patients may provide insight as to their risk of developing more serious retinopathy or nephropathy, and whether good glycemic control may protect against these complications of insulin-dependent diabetes.

Effects of dietary fibre and exercise on mid-morning diabetic control--a controlled trial

Twenty-one insulin-dependent diabetic children completed a trial of 4 different breakfasts, given in random order. Three diets differed in fibre content. The fourth diet contained soya beans as part (38%) of the dietary fibre source. Children collected capillary blood samples on to filter paper strips which were analysed for blood glucose content. Each morning the children were asked to exercise vigorously for an hour and to rest for an hour, resulting in comparable rest and exercise periods for each child. Mean initial blood glucose levels on the 4 diets were not significantly different. The low-fibre diet resulted in the highest blood glucose concentrations after breakfast. Blood glucose levels on the high-fibre diet did not differ from those on the medium-fibre diet. The bean diet produced the lowest mean blood glucose level and the smallest reduction in blood glucose level in the hour before lunch. All the children found the bean diet unacceptable but liked the high- and medium-fibre diets, which were as popular as the low-fibre diet. The level of prescribed exercise had no effect on the level of blood glucose. It appears that the potentially major benefits from beans are limited by their unpalatability. The more acceptable cereal fibre produces a smaller but important benefit on morning hyperglycaemia after breakfast.

Effect of intensive insulin treatment on linear growth in the young diabetic patient

Although impaired growth is a well-recognized complication of uncontrolled diabetes, it has not been established whether less severe metabolic derangements commonly seen with conventional treatment adversely affected growth potential. To examine this question, growth velocity was measured in nine type 1 diabetic patients (age 14 +/- 3 years) before and after six months of intensive insulin treatment either with the insulin pump or with multiple injections, which lowered mean plasma glucose concentration from 270 +/- 96 to 105 +/- 55 mg/dl and total glycosylated hemoglobin from 12.4 +/- 3.0 to 8.4 +/- 1.5% (mean +/- SD). During conventional treatment, growth velocity (5.3 +/- 2.2 cm/year) was within the range of normal despite elevations in plasma glucose concentrations. However, growth velocity increased sharply during intensive treatment (to 9.4 +/- 3.9 cm/year, P less than 0.005), reaching values in excess of normal in seven patients. The increase in growth velocity observed during intensive treatment was associated with a twofold rise in plasma somatomedin-C values. Skeletal maturation, previously normal or slightly delayed, did not advance excessively. These data indicate that the metabolic changes accompanying intensive treatment may enhance growth in diabetic children, even in those with apparently normal growth velocity during conventional therapy.

Influence of long-term insulin infusion pump treatment of type I diabetes on diabetic retinopathy

Isolated case reports have suggested short-term beneficial effects of subcutaneous insulin infusion pump treatment on background and proliferative diabetic retinopathy. To evaluate this question further, 30 eyes of 15 Type I diabetic patients were evaluated prospectively before and after 11--23 months (mean 18.1 months) of pump treatment. In each patient plasma glucose and total glycosylated hemoglobin fell to normal or near normal levels. The ten eyes without diabetic retinopathy at entry remained without. Four of 20 eyes with diabetic retinopathy at entry advanced by modified Early Treatment Diabetic Retinopathy Study (ETDRS) classification, including one eye that progressed from background to proliferative diabetic retinopathy. No eyes with diabetic retinopathy improved their modified ETDRS classification. One eye progressed to blindness; no other eye lost vision. Six eyes had laser treatment prior to insulin pump treatment; four of these and two more required laser during pump treatment. Two eyes had vitreous hemorrhages prior to pump treatment; one of these and four others hemorrhaged during pump treatment. No eyes with diabetic retinopathy showed regression of microvascular changes. The data suggest prolonged restoration of near normal glucose metabolism with the insulin pump does not reverse established diabetic retinopathy. Whether pump treatment slows the progression, or prevents the development, of diabetic retinopathy remains to be established.

Impact of home blood glucose monitoring on childhood diabetes

Ninety diabetic children each provided at least one 24-hour blood glucose profile at home using an impregnated filter paper strip. The mean 24-hour blood glucose level correlated significantly with urine control, height velocity, and Hb A1. The correlation coefficient for individual blood glucose values (r = 0.61) and for mean 24-hour blood glucose values (r = 0.73) repeated within 14 days showed an acceptable degree of reproducibility for the blood glucose profiles. Mean 24-hour blood glucose values fell significantly overall (11.4 to 9.8 mmol/l; 205 to 176 mg/100 ml) in 47 children who had repeated profiles more than 2 weeks apart. Unrecognised nocturnal hypoglycaemia (less than 3.0 mmol/l; 54 mg/100 ml) was found in 19% of children on twice-daily Semitard insulin. The study shows that children over age 7 years manage home blood glucose monitoring without difficulty. It shows that the results are reproducible and correlate with other indices of control, and that it provides a practical basis for the improvement of diabetic control.

Factors affecting glycosylated hemoglobin values in children with insulin-dependent diabetes

In 477 children with IDD treated by conventional methods, GHb (microcolumn chromatography) and a simultaneous random blood glucose concentration were measured over an 18-month period as indicators of metabolic control (once in 61 children, twice in 99, three or more times in 317). The data were analyzed to assess the effects of patient's age, sex, disease duration, and, in a random subgroup of 273, the number of daily insulin injections and insulin dose (U/kg). The mean +/- SEM percent GHb over this period was 11.8 +/- 0.2% and blood glucose concentration 237 +/- 9 mg/dl. Only seven children (1.4%) had a normal GHb value. There was a highly significant correlation between GHb and both age and blood glucose concentration but not with disease duration greater than one year. The correlation with age was present only in the girls. In 416 children evaluated more than once, with a mean duration between initial and most recent evaluations of 11.3 months, GHb remained within +/- 1% of the initial value in 40.5%, decreased in 32.3%, and increased in 24.2%. These data indicate a closer relationship between metabolic control in children with IDD and age of the child, particularly in females, than with disease duration. In our clinic, using conventional therapeutic methods, the ability to improve control over the short term as measured by changes in percent GHb has been quite limited. This study helps to target those IDD children, especially adolescent girls, requiring a more aggressive therapeutic approach.

Comparison of single- and split-dose insulin regimens with 24-hour monitoring

It has been asserted that twice daily injections of mixed insulin provide better blood glucose control than one. To compare the two regimens we conducted a random-order, double-crossover trial in ten diabetic children. Each regimen lasted for six weeks, concluding with a hospital evaluation. Control at home was assessed by a urine log and determination of glycosylated hemoglobin. Control in the hospital was assessed with measurements of quantitative urinary glucose, serum lipids, and by 24-hour blood sampling for glucose, C-peptide, and counterregulatory hormones. For the group as a whole, none of the indices of control demonstrated a significant advantage for either regimen. Individually, several children did appear to achieve better control on one regimen than the other. Indices of control at home did not consistently predict control in the hospital. In the hospital, the largest increases in glucose concentration followed breakfast (mean rise 148 mg/dl), and standardized exercise invariably reduced plasma glucose values (mean decrement 60 mg/dl). C-Peptide concentrations were low, but higher values were associated with better control. Although a split insulin regimen may improve metabolic control in some patients, this study did not demonstrate a substantial advantage for the majority of subjects over the short period of the trials.

Glycosylated hemoglobins and diabetic control assessment

This historical development, biosynthesis, analysis, interpretation, and utilization of glycosylated hemoglobin determinations (HbA1 or HbA1c alone) are presented. The relationship between glycosylated hemoglobin determinations and assessment of diabetic control is discussed. HbA1 and HbA1c levels are elevated approximately twofold in diabetic patients. A correlation exists between abnormal fluctuations in blood glucose levels and HbA1 concentrations. The results of glycosylated hemoglobin determinations are not influenced by recent meals, physical activity, emotional stress, or inherited abnormalities. Although glycosylated hemoglobin determinations are subject to fewer variables than blood and urine glucose assessments, the clinician must be aware of the indications and limitations for their use. These indications and limitations are discussed.

The effects of targeting improvements in urine glucose on metabolic control in children with insulin dependent diabetes

A treatment program designed to increase the percentage of negative urine tests was implemented using a multiple-baseline across groups design in a sample of 19 families of children with insulin dependent diabetes. The treatment involved instruction in insulin adjustment, decrease in intake of simple sugars and saturated fats, and increase in exercise, along with teaching the parents to support improvements in children's self-regulatory behaviors using a point economy and praise. New procedures designed to measure and reinforce adherence to the urine testing regimen were developed. Results showed significant increases in percentage of negative urines consistent with implementation of treatment across the three treatment groups, which were maintained over the follow-up period. Metabolic measures of control, including glycosylated hemoglobin and serum glucose did not show improvements even though the relationship between the percentage of negative urine tests and glycosylated hemoglobin was very high during treatment.

Comparison between multi-injection and continuous subcutaneous insulin therapy in insulin-dependent diabetic inpatients

Circadian blood glucose profiles have been evaluated in 8 insulin-dependent diabetic inpatients on their usual home insulin therapy, on a 3-injection regimen (ultralente in the morning plus 3 injections of regular insulin at meals), on continuous subcutaneous insulin infusion by portable micropumps (Mill Hill 1001) and, again, on a 3-injection regimen at the same insulin dose as during continuous subcutaneous insulin infusion. The 3-injection regimen achieved a mean daily blood glucose level comparable to that obtained by continuous subcutaneous insulin infusion, even if significantly more insulin was needed. At comparable insulin doses, continuous subcutaneous insulin infusion provided a significantly lower mean daily blood glucose. Glycemic control at 0600 and 0800 was better during continuous subcutaneous insulin infusion. Low acceptance by the patients of the home use of portable micropumps was evidenced because of the practical and psychological problems involved.

Control of blood sugar in insulin-dependent diabetes: comparison of an artificial endocrine pancreas, continuous subcutaneous insulin infusion, and intensified conventional insulin therapy

We compared the ability of closed-loop intravenous insulin infusion (i.e., an artificial "pancreas"), open-loop continuous subcutaneous insulin infusion, and intensified conventional insulin therapy (preprandial injections of regular insulin, with injection of long-acting zinc-suspension insulin before breakfast) to bring the hyperglycemia of insulin-dependent diabetic subjects to a level comparable to that of normal, nondiabetic subjects. The mean circadian levels of plasma glucose, mean amplitude of glycemic excursions, and M values (defined in Methods) did not significantly differ among the three regimens. Although these levels in the diabetic subjects approximated those in the normal subjects, the levels of plasma insulin, mean amplitude of glycemic excursions, and M values were significantly higher than those in normal subjects (P < 0.01). Therefore, at least on a short-term basis, all three regimens can produce comparable, nearly normal levels of blood sugar in such patients; moreover, closed-loop devices can be used to determine insulin requirements for conventional therapy.

Temperature dependence of the chromatographic assay of hemoglobin A1 and application of a temperature controlled assay to clinical evaluation of diabetic control

We investigated the temperature dependence of the cation-exchange chromatographic properties of hemoglobin A1 using the "Fast Hemoglobins Test System" obtained from Isolab, Inc., Akron, OH, U.S.A. and a temperature control chamber constructed in our laboratory. Six patient samples, two each from the "normal" range (5--8.5%, "transitional" range (8.5--12.0%) and "uncontrolled diabetic" range (12--20%) described by Isolab (as measured at 24 degrees C) were collected, hemolyzed and assayed according to product insert instructions. Each sample was assayed in triplicate in three separate runs at 20 degrees C, 22 degrees C, 24 degrees C, 26 degrees C and 28 degrees C in the temperature controlled (+/- 0.25 degrees C) chamber. Resultant mean increases in assay results of 0.8, 1.1 and 1.6% hemoglobin A1 per 2 degrees C increase in temperature for the three ranges respectively, indicate that the temperature dependence of the microcolumn assay for hemoglobin A1 is at least partially a function of the relative percentage of glycosylated hemoglobin in the sample. By controlling assay temperature we have reduced interassay variation throughout all concentration ranges by more than one-half to less than 5%. We applied this temperature controlled assay to measurement of hemoglobin A1 levels in 33 non-diabetic subjects and to the screening of hemoglobin A1 levels in 77 diabetic patients regularly attending the diabetic-endocrine outpatient clinic of this hospital. No non-diabetic subject had a HbA1 level above 8.5% and of the diabetic patients, 26% fell within the "normal" range, 43% within the "transitional" range and 31% within the "uncontrolled diabetic" range.

Self-control with urinalysis in juvenile diabetes

Urinary glucose excretion reflects the blood glucose levels and is therefore recommended and used as a relevant and practical method for self-control in juvenile diabetes. The purpose of this study was to estimate the attitudes of diabetic children and their parents towards such daily urinalysis. In 1975 69 juvenile diabetics 6--18 years old and their parents were studied and three years later another 69 patients were added. Standardized interviews and questionnaires were used. Only 3 out of 138 patients refused to test their urine regularly and to write down their results in the diary. The results indicate that a great majority of the patients and the parents easily accept the self-testing method and regard it as a valuable tool in the management of the disease. Almost nobody experienced the urine tests as a psychological problem.

Metabolic control in diabetic patients: assessment by hemoglobin A1 values

Hemoglobin A1 concentrations were measured in 230 patients with diabetes mellitus. Diabetic patients controlled by diet alone had significantly lower HbA1 levels compared to the insulin or oral hypoglycemic treated groups. Only 13% of insulin-treated patients had HbA1 levels below 10%, while 26% of the oral agent treated group and 70% of the diet treated group fell into this range. The HbA1 levels correlated with fasting and nonfasting plasma glucose concentrations. However, in certain patients, discrepant results between these two variables were found. Hemoglobin A1 was measured on three or more occasions in 60 patients over a 1-yr period. An improvement was noted in 40%, no change in 38%, and a deterioration in 22%. Hemoglobin A1 measurements have proved to be useful in the follow-up and treatment of diabetic patients.

HbAlc--an indicator of diabetic control

Hemoglobin Alc levels were determined on 16 nondiabetic and 115 diabetic subjects. Twelve of the diabetic patients were newly diagnosed and 12 were in the remission phase. The mean percentage of HbAlc in the nondiabetic subjects was 4.33 +/- 0.39 (SE), in those with long-standing diabetes 8.34 +/- 0.23 SE, in newly diagnosed diabetic patients 8.99 +/- 0.50 SE, and in those undergoing partial remission 5.16 +/- 0.18 SE. The mean HbAlc levels of these four groups differed significantly (0.001 less than P less than 0.025). Among the diabetic subjects there was a good correlation between HbAlc level and such measures as fasting blood sugar, urinary sugar, and degree of diabetic control, whereas the correlations between HbAlc and age, sex, or duration of the disease were not significant. The obtained data show that HbAlc measurement can serve as an objective measure of glucose control in diabetic subjects.

Clinical evaluation of somatostatin as a potential ajunct to insulin in the management of diabetes mellitus

To determine whether somatostatin, an inhibitor of glucagon and growth hormone secretion, might be useful as an adjunct to insulin the management of diabetic hyperglycaemia, seven insulin-requiring diabetic men were given somatostatin (100 microgram/h, IV) continuously for 3 days after their diabetes had been treated intensively by diet and insulin on a metabolic ward. During infusion of somatostatin and despite reduction in average insulin dose exceeding 50%, there was improvement in diabetic control as assessed by postprandial hyperglycaemia, 24-h glycosuria and the average daily serum glucose level and its fluctuation; when somatostatin was discontinued, but insulin doses held constant, diabetic control rapidly worsened. No adverse effects were observed. These results indicate that somatostatin plus insulin can be a more effective regimen than insulin alone in controlling diabetic hyperglycaemia. A longer acting and more selective somatostatin preparation may prove useful as an adjunct to insulin in the management of diabetes.