Original Article
Hypoglycemia during Diarrhea in Childhood — Prevalence, Pathophysiology, and Outcome
N Engl J Med 1990; 322:1357-1363May 10, 1990
- Abstract
- Abstract
To determine the frequency and outcome of hypoglycemia during diarrhea in childhood, we screened 2003 consecutive patients less than 15 years of age who were admitted to a diarrhea treatment center in Dhaka, Bangladesh. Hypoglycemia, defined as a blood glucose concentration <2.2 mmol per liter, was found in 91 patients (4.5 percent), 39 (42.9 percent) of whom died. We also measured the plasma concentrations of glucoregulatory hormones and gluconeogenetic substrates in 46 of the patients with hypoglycemia who were 2 to 15 years old and in 25 normoglycemic patients matched with them for age and weight.
The patients with hypoglycemia had had diarrhea for less time than the normoglycemic patients (median, 12 vs. 72 hours; P<0.05), and their last feeding had been 18 hours before admission, as compared with 9 hours for the normoglycemic patients (P<0.05). The groups were similar in terms of nutritional status, the proportion of patients who had fever, and the types of pathogens recovered from stool samples. The plasma C-peptide concentrations were low (<0.30 nmol per liter) in all the hypoglycemic patients. As compared with the normoglycemic patients, the patients with hypoglycemia had elevated median plasma concentrations of glucagon (44 vs. 11 pmol per liter; P = 0.001), epinephrine (3400 vs. 1500 pmol per liter; P = 0.012), norepinephrine (7500 vs. 2900 pmol per liter; P = 0.002), and lactate (3.5 vs. 2.1 mmol per liter; P = 0.020) and similar alanine and β-hydroxybutyrate concentrations. Eighteen hypoglycemic patients with severe malnutrition had been ill longer than 26 better-nourished patients with hypoglycemia (median duration of illness, 18 vs. 10 hours; P = 0.023) and had lower median plasma concentrations of lactate (1.9 vs. 3.9 mmol per liter; P = 0.021) and alanine (173 vs. 293 μmol per liter; P = 0.040).
We conclude that hypoglycemia is a major cause of death in association with diarrhea. Because the glucose counterregulatory hormones were appropriately elevated in the children with diarrhea and hypoglycemia, whereas the gluconeogenetic substrates were inappropriately low, we further conclude that the hypoglycemia observed in such patients is most often due to the failure of gluconeogenesis. (N Engl J Med 1990; 322:1357–63.)
Media in This Article
Figure 1
Scattergram of the Plasma Glucose and C-Peptide Concentrations in 44 Patients at the Time Hypoglycemia Was First Detected.Table 1
Frequency and Outcome of Hypoglycemia in 2003 Children Less than 15 Years of Age Admitted to the Diarrhoea Treatment Centre of the International Centre for Diarrhoeal Disease Research, Bangladesh, According to the Enteric Pathogen Identified in Stool Samples.Article Activity
- Article
IN many developing countries, diarrhea is the major cause of death in children one to five years of age.1 This has remained true even in areas where oral rehydration therapy has been widely used.2 , 3 The lethal complications of diarrhea, other than dehydration, are for the most part poorly understood. Hypoglycemia has been found to be a potentially fatal complication of infectious diarrhea in both well-nourished and poorly nourished children.4 5 6 In those studies, which involved limited numbers of patients, hyperinsulinemia was not considered responsible for the hypoglycemia.4 Three of four patients examined at autopsy had adrenal hemorrhage or necrosis, suggesting that deficiencies in epinephrine and cortisol secretion were responsible for the hypoglycemia; except for insulin, however, circulating concentrations of glucoregulatory hormones were not measured.4 Hence, the mechanism of the hypoglycemia remained obscure.
To understand better the hypoglycemia that occurs during diarrhea, we determined the frequency and outcome of hypoglycemia in a prospective study of 2003 severely ill children with diarrhea who were admitted to a treatment center during an eight-month period; we also studied the biochemical and hormonal features of the illness in 46 of the hypoglycemic patients and in 25 patients without hypoglycemia.
Methods
Selection of Patients
This study was conducted at the Diarrhoea Treatment Centre of the International Centre for Diarrhoeal Disease Research in Dhaka, Bangladesh, during the period from August 1986 through March 1987. Each year this treatment center provides care free of charge to approximately 70,000 patients with diarrhea, 65 percent of whom are children less than 15 years of age. After examination by a physician, approximately 7 percent of these patients, mostly children, are admitted to an inpatient unit. Although there are no rigid criteria for admission to the inpatient unit, those who are admitted are generally patients at the extremes of age, severely malnourished patients, patients with very high purging rates or severe dehydration, or patients with complications of diarrhea.
To determine the incidence of hypoglycemia in children with diarrhea, we used a reflectometer (Glucochek SC, Medistron, Horsham, England) and glucose oxidase—peroxidase test strips (Haemo-Glukotest 20–800R, Boehringer–Mannheim, Mannheim, Federal Republic of Germany) to measure the blood glucose concentration in all children less than 15 years of age who were admitted to the inpatient unit. In order to determine the biochemical and hormonal features of the hypoglycemia, we also undertook a more intensive study of a subset of those children, 2 to 15 years of age, who had a capillary blood glucose concentration of less than 2.2 mmol per liter at admission. Patients less than two years old were excluded from this part of the study because of the amount of blood required for study purposes. For the first four months, all patients were tested for hypoglycemia; because only one patient identified as having hypoglycemia during this period did not have altered consciousness, only patients with altered consciousness were tested for hypoglycemia during the second four months.
For each of the first 25 patients with hypoglycemia, a patient who was acutely ill with diarrhea and had a blood glucose concentration between 3.3 and 9.0 mmol per liter at the time of admission was selected. The hypoglycemic (case) and normoglycemic (control) patients were matched for age and nutritional status, with use of the following categories: age – 24 to 47 months, 48 to 71 months, 72 to 119 months, and 120 months to 15 years; weight – <60 percent of the National Center for Health Statistics (NCHS)7 median for age, 60 to 74 percent of the median for age, and ≥ 75 percent of the median for age. The first patient identified after the admission of the case patient who met the matching criteria was selected to serve as the control. The number of control patients was limited to 25 because of constraints on the number of assays that could be performed. Moreover, these 25 patients were broadly representative of the general inpatient population in terms of nutritional status and the enteric pathogens identified in their stools and were adequate for comparison with the hypoglycemic group.
In the calculation of weight for age, the weight after the correction of dehydration (if present) was used. Patients were considered marasmic if their weight for age was less than 60 percent of the NCHS median and if they had clinical features typical of marasmus, such as sparse, brittle hair, apathy, wasting, and characteristic skin changes; they were considered to have kwashiorkor if nutritional edema was present.
Study Protocol
When hypoglycemia was diagnosed or a suitable control patient was identified, informed consent for participation in the study was requested from the patient's parents or next of kin. If consent was provided, a Teflon catheter was inserted into an antecubital vein and blood was drawn into a tube containing potassium oxalate and sodium fluoride for the determination of the plasma glucose level; into lithium—heparin tubes for the determination of the plasma C-peptide, β-hydroxybutyrate, lactate, alanine, Cortisol, epinephrine, norepinephrine, and growth hormone levels; into a chilled lithium—heparin tube containing aprotinin for the determination of the plasma concentration of pancreatic glucagon; and into an untreated tube for the determination of the serum albumin, bilirubin, and aspartate aminotransferase levels. Blood was also obtained for a complete blood count, bacteriologic culture, and measurements of plasma electrolytes and creatinine. While blood was being drawn, another member of the study team obtained a history and carried out a physical examination. Patients who did not respond to verbal commands, who could not be aroused, or who had convulsions were considered unconscious.
Immediately after blood was obtained from the patients with hypoglycemia, 5.6 mmol (1 g) of dextrose per kilogram of body weight was infused over a period of 5 to 10 minutes into a different vein from the one used for blood sampling. The patients then received intravenous fluids containing 278 mmol (50 g) of dextrose per liter at rates determined by their fluid requirements (the median infusion rate during the first 24 hours of hospitalization was 77 ml per kilogram of body weight as measured at admission; range, 11 to 265 ml per kilogram). Blood sampling for the determination of plasma glucose and C-peptide concentrations was repeated 15 minutes and 1, 2, 4, and 24 hours after the initial sampling; sampling for the determination of β-hydroxybutyrate, lactate, alanine, Cortisol, epinephrine, norepinephrine, growth hormone, and glucagon levels was repeated after 4 and 24 hours. Plasma glucose concentrations were also determined whenever there was a deterioration in the patient's level of consciousness.
Laboratory Methods
Plasma was separated within 15 minutes of blood sampling and then stored at −70°C until it was transported, packed in dry ice, to collaborating laboratories in England. Hemolysis in some samples precluded the determination of insulin levels; therefore, plasma C-peptide concentrations, which are not affected by hemolysis, were determined instead in all samples. Plasma glucose, glucoregulatory hormones, β-hydroxybutyrate, alanine, epinephrine, and norepinephrine were measured by Standard methods.8 Plasma Cortisol and growth hormone concentrations were determined by radioimmunoassay with commercial kits (DPC/ERI, Witney, Oxfordshire, England). A single stool sample was obtained from each patient for the identification of shigella species, salmonella species, Campylobacter jejuni, Vibrio cholerae, Escherichia coli producing heat-stable or heat-labile toxin, and rotavirus, according to previously described methods.9
Statistical Analysis
The two-tailed Mann—Whitney U test was used to test the significance of differences in continuous variables, most of which had a non-normal distribution. When more than two groups were compared, the Mann—Whitney U test was used only if P was less than 0.05 for the overall group comparison according to the Kruskal—Wallis test. The chi-square statistic was used to test the significance of differences in proportions. Entry of data into the computer and all statistical testing were performed with Stat-Pac Gold version 3.0 software (Walonick Associates, Minneapolis). Ninety-five percent confidence intervals for relative risks were calculated according to the method of Katz et al.10
This study was approved by the Ethical and Research Review Committees of the International Centre for Diarrhoeal Disease Research, Bangladesh.
Results
During the eight-month study period, 2003 children less than 15 years old were admitted to the inpatient unit. Hypoglycemia was detected in 91 (4.5 percent) of these patients. Thirty-nine (42.9 percent) of the 91 hypoglycemic patients died while in the hospital, as compared with 122 (6.4 percent) of 1912 patients without hypoglycemia (P<0.001; relative risk, 6.7; 95 percent confidence interval, 5.0 to 9.0) (Table 1Table 1
Frequency and Outcome of Hypoglycemia in 2003 Children Less than 15 Years of Age Admitted to the Diarrhoea Treatment Centre of the International Centre for Diarrhoeal Disease Research, Bangladesh, According to the Enteric Pathogen Identified in Stool Samples.). Thus, hypoglycemia occurred in 24.2 percent (39 of 161) of the patients who died. The median age of the patients with hypoglycemia was 36 months, as compared with 13 months for the patients without hypoglycemia (P<0.001). Seventy-three (80.2 percent) of the hypoglycemic patients were five years of age or less. Forty-five (49.5 percent) were boys, as compared with 38.9 percent of patients without hypoglycemia (P = 0.034).Enteric pathogens were identified in stool samples from 59 (64.8 percent) of the hypoglycemic patients. Shigella species (identified in samples from 35 patients) and V. cholerae 01 (12 patients) were the two most commonly identified pathogens. Other enteric pathogens identified were V. cholerae non-01 (in samples from three patients), Entamoeba histolytica (three patients), E. coli producing heat-stable toxin (two patients), rotavirus (two patients), and C. jejuni and salmonella group C (one patient each). Hypoglycemia was significantly more common among the patients with shigellosis than among the patients infected with other enteric pathogens or those in whom no pathogen could be identified (Table 1). Among the patients with shigellosis, hypoglycemia was most common in those infected with Shigella flexneri (Table 1).
Clinical Findings
More detailed clinical observations were made and the biochemical and hormonal features of the hypoglycemia were studied in 46 (50.5 percent) of the 91 hypoglycemic patients. Of the 45 hypoglycemic patients who were not included in this part of the study, 24 were excluded because they were less than 24 months old, and 21 either because informed consent was not obtained or because a study physician was not available when the hypoglycemia was detected. The results in these 46 patients were compared with those in 25 normoglycemic (control) patients selected as previously described (Table 2Table 2
Clinical Characteristics at Admission of 46 Hypoglycemic Patients and 25 Normoglycemic Patients with Diarrhea.*). The two groups of patients were comparable in terms of age, nutritional status, and pathogens identified in the stool. The duration of diarrhea was significantly shorter for the hypoglycemic patients than for the normoglycemic patients, and the length of time since the last feeding was significantly longer. Even among the hypoglycemic patients, however, the time since the last feeding was relatively short; 11 (23.9 percent) had eaten in the 12 hours before admission, and 40 (87.0 percent) within 24 hours of admission. Before admission, 14 patients (30.4 percent) had received allopathic medications (Western pharmaceutical agents), 2 (4.3 percent) had received homeopathic medications (agents compounded by practitioners of homeopathic medicine, usually prepared from ingredients imported from homeopathic pharmaceutical companies in the United States), and 8 (17.4 percent) had received oral rehydration therapy. No patient had received herbal remedies.Seven (15.2 percent) of the patients with hypoglycemia had marasmus, and 1 (2.2 percent) had kwashiorkor. Eighteen of the 46 patients with hypoglycemia (39.1 percent) were severely malnourished when weight for age was used as a criterion (<60 percent of the NCHS median). During the last half of the study we also determined weight for height, a better measure of acute wasting than weight for age. Two of 22 patients (9.1 percent) were severely malnourished (<70 percent of the median) according to this criterion.
Sixteen of the hypoglycemic patients (34.8 percent) had convulsions at the time of admission. An additional 24 (52.2 percent) were unconscious but had no convulsions. Five of the remaining six patients, classified as conscious, were either lethargic or irritable. The median duration of altered consciousness before admission was five hours. One hour after the initial dextrose infusion, 18 (45.0 percent) of the 40 patients who had convulsions or were unconscious at the time of admission had regained consciousness. Samples of spinal fluid were obtained from 28 patients, and all the samples were sterile. No sample contained more than 4 white cells per cubic millimeter.
Sixteen (34.8 percent) of the 46 hypoglycemic patients in the study group died, 10 within 24 hours of admission, as compared with 3 (12.0 percent) of the 25 normoglycemic patients (P = 0.073). Two of the 30 surviving hypoglycemic patients had neurologic deficits; one patient was blind, and a second had recurrent focal seizures.
Biochemical and Hormonal Findings
The biochemical and hormonal findings in the hypoglycemic and normoglycemic patients are summarized in Table 3Table 3
Biochemical and Hormonal Findings at Admission in 46 Hypoglycemic Patients and 25 Normoglycemic Patients with Diarrhea.*.Plasma Glucose and C-Peptide Concentrations
The initial plasma glucose concentrations in the 46 hypoglycemic patients ranged from 0.05 to 2.2 mmol per liter (median, 0.7 mmol per liter). All the patients who were found to be hypoglycemic on screening with the reflectometer had plasma glucose concentrations <2.2 mmol per liter when the samples were assayed later in England. As an additional check on the accuracy of the diagnosis of hypoglycemia, plasma glucose concentrations were determined in duplicate samples obtained at admission from 28 hypoglycemic patients and submitted to the laboratory at the International Centre for Diarrhoeal Disease Research, Bangladesh. The glucose concentrations measured in Dhaka and England were similar; the median concentration measured in England (0.83 mmol per liter) was slightly higher than that measured in Dhaka (0.74 mmol per liter). Three patients (none of whom was severely malnourished) had recurrent hypoglycemia 10, 20, and 48 hours after admission. In all three patients, the hypoglycemia was relatively mild (plasma glucose concentrations, 2.1, 1.9, and 1.6 mmol per liter) and only one had symptoms of hypoglycemia. Two of the three patients were still receiving intravenous solutions containing dextrose at the time of the recurrent episode.
In the 44 patients in whom it was measured, the plasma C-peptide concentration at the time hypoglycemia was first detected was appropriately low (median, 0.03 nmol per liter; range, 0 to 0.29 mmol per liter). Only three patients had concentrations ≥0.20 nmol per liter (Fig. 1Figure 1
Scattergram of the Plasma Glucose and C-Peptide Concentrations in 44 Patients at the Time Hypoglycemia Was First Detected.).Counterregulatory Hormone Responses
In general, all counterregulatory hormones were appropriately elevated when hypoglycemia was detected; except for plasma cortisol, the median concentration was significantly higher than that in the normoglycemic patients. Two patients had plasma epinephrine concentrations within the normal range for supine children; in both, the plasma glucagon concentration was elevated. Three patients with hypoglycemia had plasma cortisol concentrations of 180, 312, and 484 nmol per liter when their plasma glucose concentrations were 1.7, 1.5, and 2.2 mmol per liter, respectively. The plasma cortisol concentration subsequently exceeded 550 nmol per liter in two of these patients but remained low in the other. Plasma growth hormone concentrations at admission were less than 20 mU per liter in 8 of 45 patients but subsequently exceeded this value in 5 of the 8 patients.
Plasma β-Hydroxybutyrate, Lactate, and Alanine Concentrations
The plasma β-hydroxybutyrate concentrations in the hypoglycemic patients were generally elevated and correlated with the interval between the last feeding and the initial detection of hypoglycemia (r = 0.628, 44 df; P<0.001). The median value was not, however, significantly higher than that in the normoglycemic patients. The median lactate concentration in the hypoglycemic patients was significantly higher than that in the normoglycemic patients (P = 0.021). There was a significant correlation between plasma lactate and alanine levels (r = 0.899, 44 df; P<0.001). Plasma alanine concentrations were similar in the hypoglycemic and normoglycemic patients.
Serum Bilirubin and Aspartate Aminotransferase Concentrations
Five (13.5 percent) of the 37 hypoglycemic patients tested and none of the 17 normoglycemic patients tested had serum bilirubin concentrations higher than 17 μmol per liter (P not significant). The median bilirubin concentration, however, was significantly higher in the hypoglycemic patients than that in the normoglycemic patients. Serum aspartate aminotransferase concentrations greater than 100 U per liter were found in 20 (48.8 percent) of the 41 hypoglycemic patients tested and 1 (4.8 percent) of the 21 normoglycemic patients tested (P = 0.001).
Response to the Infusion of Dextrose
The infusion of 5.6 mmol of dextrose per kilogram resulted in a prompt and sustained increase in plasma glucose concentrations. The median levels at 15 minutes and at 1, 2, 4, and 24 hours were, respectively, 23.5, 19.8, 17.3, 9.3, and 4.9 mmol per liter. The plasma C-peptide concentrations rose appropriately but lagged behind the increase in glucose; the C-peptide concentrations at the corresponding times were 0.27, 0.29, 0.49, 0.59, and 0.75 nmol per liter. These glucose and C-peptide concentrations were all significantly higher than those measured at admission (P<0.001). The median concentrations of all counterregulatory hormones and of β-hydroxybutyrate were significantly lower at 4 and 24 hours than at the time of admission (P<0.05). The median plasma alanine and lactate concentrations did not change significantly.
Differences among Subgroups of Patients with Hypoglycemia
We analyzed separately five subgroups of patients with hypoglycemia: severely malnourished patients (<60 percent of the NCHS median weight for age); patients who died; patients with shigella infection; patients who had convulsions on admission; and patients with moderate or severe dehydration.
The severely malnourished patients with hypoglycemia had a distinct clinical and biochemical profile as compared with better-nourished patients with hypoglycemia (Table 4Table 4
Clinical Characteristics and Biochemical and Hormonal Findings at Admission in Patients with Hypoglycemia and Diarrhea, According to nutritional Status.*). They had been ill longer, were less likely to have fever, had lower median plasma alanine and serum albumin concentrations, and had a higher median plasma cortisol concentration. For all the hypoglycemic patients, the plasma lactate concentration and weight for age were positively correlated (r = 0.425, 42 df; P = 0.005), and the plasma cortisol concentration and weight for age were inversely correlated (r = −0.635, 38 df; P<0.001).The patients who died, when compared with those who survived, had a significantly lower median weight for age (56 vs. 63 percent of the NCHS median; P = 0.044), a higher median serum bilirubin concentration (8 vs. 5 μmol per liter; P = 0.027), and higher median plasma norepinephrine concentrations at 4 hours (8400 vs. 3600 pmol per liter; P = 0.002) and 24 hours (4500 vs. 1800 pmol per liter; P = 0.050). The patients with shigellosis differed from the other patients only in having greater numbers of white and red cells in their stools.
The patients who had convulsions at the time of admission had a significantly lower median plasma glucose concentration than the patients without convulsions (0.4 vs. 1.1 mmol per liter; P = 0.001) and a higher median plasma lactate concentration (3.9 vs. 2.9 mmol per liter; P = 0.078). The median lactate levels did not differ significantly between the patients who died and those who survived, nor between the patients who were moderately or severely dehydrated and those with mild or no dehydration.
Discussion
We identified hypoglycemia in 91 (4.5 percent) of 2003 children less than 15 years of age who had diarrhea and who were admitted to our treatment center during an eight-month period. Because only patients with altered consciousness were screened for hypoglycemia during the second half of the study, the true incidence of hypoglycemia may be slightly higher than 4.5 percent. The hypoglycemia was often profound; the initial plasma glucose concentrations were <1.0 mmol per liter in 61 percent of the 46 hypoglycemic patients studied further. Thirty-nine hypoglycemic patients (43 percent) died, accounting for 24 percent of all inpatient deaths during this period.
When they were hypoglycemic, all patients who were studied in more detail had appropriately low plasma C-peptide concentrations. The median plasma concentrations of glucagon and epinephrine, the major counterregulatory hormones,15 were increased. Thus, the failure of glucose counterregulation — a mechanism suggested on the basis of previous findings of adrenal hemorrhage or necrosis at autopsy in a few patients who had had hypoglycemia during diarrhea4 — was not a major cause of hypoglycemia in the patients in this study. The hypoglycemia occurred a median of 18 hours after the last feeding, when glycogen reserves in these patients would have been exhausted. This fact, along with the appropriate glucoregulatory hormone response, suggests that in most patients defective gluconeogenesis was the cause of the hypoglycemia.
The failure of gluconeogenesis may occur either because of defects in the enzymatic pathways and transport systems required for the process or because of deficiencies in protein- and fat-derived gluconeogenetic substrates.16 Although the plasma β-hydroxybutyrate concentrations in the hypoglycemic patients were higher than those that have been reported for normal, nonfasting children, they were lower than the values reported in some studies of fasting hypoglycemic patients.13 , 17 In addition, the β-hydroxybutyrate concentrations in the hypoglycemic patients were not significantly higher than those in the normoglycemic patients. Because the hypoglycemic patients had fasted longer and had lower C-peptide concentrations, fatty acid oxidation would probably have been occurring at a faster rate than in the normoglycemic patients, resulting in concomitant elevations in the plasma β-hydroxybutyrate concentrations. That these concentrations were not higher suggests there was deficient generation of ketones, either because fat stores were diminished as a result of malnutrition or because the oxidation of fat was impaired.
The 18 severely malnourished hypoglycemic patients had plasma concentrations of alanine, the major protein-derived substrate for gluconeogenesis,18 that were significantly lower than those in the 26 hypoglycemic patients who were not severely malnourished. The values in the severely malnourished group were similar to those in children with ketotic hypoglycemia,19 which is thought by some to result from insufficient mobilization of alanine for gluconeogenesis during periods of fasting.14 In the 26 better-nourished hypoglycemic patients, the median alanine concentration was normal but the median lactate concentration was higher than has been reported for normal children. These findings are consistent with impaired activity of hepatic enzymes involved in gluconeogenesis. Sepsis and endotoxemia are known to impair the activity of phosphoenolpyruvate carboxylase, an enzyme essential for gluconeogenesis.20 , 21 Shigellosis, which was diagnosed in 35 of the hypoglycemic patients, can produce endotoxemia even in the absence of bacteremia.22 However, a number of other hypoglycemic patients were infected with enteric pathogens, such as V. cholerae, that remain confined to the intestinal lumen and are unlikely to result in endotoxemia. In addition, the elevations in the plasma lactate concentrations may in part be attributable to excess production, since 17 patients with hypoglycemia had convulsions at the time of admission.
What are the implications of these findings for the management of children with diarrhea in developing countries? First, continued feeding during diarrhea is indicated not only because it lessens weight loss,23 , 24 but also because it may prevent fatal hypoglycemia. Although it has been well established that severely malnourished children may have mild hypoglycemia as a result of substrate deficiencies,25 26 27 28 an important conclusion of this study is that even children with no overt signs of malnutrition may have insufficient substrate available to sustain gluconeogenesis when the added stress of infection is present.29
Second, the predominance of patients with shigellosis among the patients with hypoglycemia emphasizes the need to give special attention to patients with this disease. Third, our findings point out the importance of complications other than dehydration in causing death in children with diarrhea; these complications may not be preventable simply by administering oral rehydration therapy.30 Fourth, the initial dose of dextrose administered to the children in this study (5.6 mmol per kilogram of body weight) most often resulted in prolonged hyperglycemia. Although the consequences of the hyperglycemia are uncertain, it is possible that it aggravated fluid derangements that were already occurring as a result of the underlying diarrhea. A better approach might be to use dextrose-containing intravenous solutions routinely in patients with diarrhea who require parenteral therapy; we have found this approach to be safe and effective.9
Finally, hypoglycemia has now been found to be a serious complication of both malaria31 , 32 and diarrhea, two major killers of children in the tropics. This was a somewhat unexpected finding, since in both diseases neurologic abnormalities had often been attributed to the direct effects of the infection on the central nervous system. The extent to which hypoglycemia occurs in children with diarrhea in developing countries other than Bangladesh is uncertain. Until more information is available, however, children with diarrhea and altered consciousness must be assumed to be at risk for hypoglycemia, and they should be treated accordingly.
Supported by the United Nations Children's Fund. Dr. Bennish's work was supported by grants from the Danish International Development Agency and the Applied Diarrheal Disease Research Project of the United States Agency for International Development. Dr. Phillips' work was supported by a grant from the Wellcome Trust of Great Britain.
We are indebted to Dr. R.C. Turner for performing the glucose and C-peptide assays; to Dr. D.H. Williamson and Mrs. Vera Ilic for excellent advice and for performing the β-hydroxybutyrate, lactate, and alanine assays; to Dr. J. Kaye for performing the bilirubin, aspartate aminotransferase, and albumin assays; to Dr. R.A. Moore for performing the cortisol and growth hormone assays; to Dr. G. Inglis for performing the epinephrine and norepinephrine assays; to Dr. S.R. Bloom and Dr. M. Ghatei for performing the glucagon assays; to Medistron Ltd., Horsham, England, for kindly providing the reflectometers used in screening patients for hypoglycemia; to Boehringer–Mannheim, Mannheim, Federal Republic of Germany, for providing the glucose oxidase test strips; to Dr. T. Butler for initial suggestions; to Dr. G. Keusch and Dr. J. Menken for their review of the manuscript; to Mr. H. Kabir for data entry and Mr. M.A. Khan for secretarial assistance; and especially to Dr. D.A. Warrell and Dr. R. Eeckels for their assistance in initiating the study and for their continuing advice and encouragement and to Dr. A. Sadeghi-Nejad for many helpful discussions.
Source Information
From the International Centre for Diarrhoeal Disease Research, Bangladesh, Dhaka, Bangladesh (M.L.B., A.K.A.); the Departments of Pediatrics and Medicine, Tufts—New England Medical Center, Boston (M.L.B.); the Department of Epidemiology, Harvard School of Public Health, Boston (O.R.); and the Nuffield Department of Clinical Medicine and Institute of Molecular Medicine, University of Oxford, England (R.E.P.). Address reprint requests to Dr. Bennish at the New England Medical Center, 750 Washington St., Box 041, Boston, MA 02111.
- References
References
1
The burden of disease resulting from diarrhea. In: Division of Health Promotion and Disease Prevention, Division of International Health, Institute of Medicine. New vaccine development: establishing priorities. Vol. 2. Diseases of importance in developing countries. Washington, D.C.: National Academy Press, 1986:159–69.
2
Chen
CrossRef | Web of Science | Medline3
Ronsmans
CrossRef | Web of Science | Medline4
Hirschhorn
CrossRef | Web of Science | Medline5
Jones
CrossRef | Web of Science | Medline6
Molla
Medline7
Measuring change in nutritional status. Geneva: World Health Organization, 1983:87–101.
8
White
Full Text | Web of Science | Medline9
Rahman
CrossRef | Web of Science | Medline10
Kahn HA, Sempos CT. Statistical methods in epidemiology. New York: Oxford University Press, 1989:62–4.
11
Behrman RE, Vaughn VC III, Nelson WE, eds. Nelson textbook of pediatrics. 13th ed. Philadelphia: W.B. Saunders, 1987:1536–58.
12
Adrian
Web of Science | Medline13
Wolfsdorf
CrossRef | Web of Science | Medline14
Pagliara
CrossRef | Web of Science | Medline15
Tse
CrossRef | Web of Science | Medline16
Senior
CrossRef | Medline17
Lamers
CrossRef | Web of Science | Medline18
Felig
CrossRef | Web of Science | Medline19
Pagliara
CrossRef | Web of Science | Medline20
McCallum
Web of Science | Medline21
Knowles
Web of Science | Medline22
Koster
Full Text | Web of Science | Medline23
Black
Web of Science | Medline24
Brown
CrossRef | Web of Science | Medline25
Slone
CrossRef | Web of Science | Medline26
Hadden
CrossRef | Web of Science | Medline27
Wharton
CrossRef | Web of Science | Medline28
Rao
Web of Science | Medline29
Kerr
CrossRef | Web of Science | Medline30
Bennish
CrossRef | Web of Science | Medline31
White
CrossRef | Web of Science | Medline32
Taylor
Full Text | Web of Science | Medline
- Citing Articles (18)
Citing Articles
1
Michael L. Bennish, M. John Albert. 2011. Shigellosis. , 187-204.
CrossRef2
James E. Colletti, Kathleen M. Brown, Ghazala Q. Sharieff, Isabel A. Barata, Paul Ishimine. (2010) The Management of Children with Gastroenteritis and Dehydration in the Emergency Department. The Journal of Emergency Medicine 38:5, 686-698
CrossRef3
R. Ganeshalingam, M. O'Connor. (2009) Evidence Behind the WHO Guidelines: Hospital Care for Children: What is the Efficacy of Sublingual, Oral and Intravenous Glucose in the Treatment of Hypoglycaemia?. Journal of Tropical Pediatrics 55:5, 287-289
CrossRef4
Wilco C.W.R. Zijlmans, Anne A.M.W. van Kempen, Mireille J. Serlie, Hans P. Sauerwein. (2009) Glucose metabolism in children: influence of age, fasting, and infectious diseases. Metabolism 58:9, 1356-1365
CrossRef5
DAVID R. LANGDON, CHARLES A. STANLEY, MARK A. SPERLING. 2008. Hypoglycemia in the Infant and Child. , 422-443.
CrossRef6
D. Carré. (2004) Conduite à tenir devant une diarrhée aiguë. Étiologies. EMC - Chirurgie 1:5, 493-532
CrossRef7
Richard Hackett, Thomas Iype. (2001) Malnutrition and childhood epilepsy in developing countries. Seizure 10:8, 554-558
CrossRef8
Aida N. Lteif, W.F. Schwenk. (1999) HYPOGLYCEMIA IN INFANTS AND CHILDREN. Endocrinology & Metabolism Clinics of North America 28:3, 619-646
CrossRef9
S. N. Heyman, Y. Ginosar, M. Shapiro, Y. Kluger, N. Marx, S. Maayan. (1997) Diarrheal Epidemics Among Rwandan Refugees in 1994. Journal of Clinical Gastroenterology 25:4, 595-601
CrossRef10
Evelien Dekker, Johannes A. Romijn, Catherine Waruiru, Mariëtte T. Ackermans, Gerrit J. Weverling, Robert W. Sauerwein, Erik Endert, Norbert Peshu, Kevin Marsh, Hans P. Sauerwein. (1996) The relationship between glucose production and plasma glucose concentration in children with falciparum malaria. Transactions of the Royal Society of Tropical Medicine and Hygiene 90:6, 654-657
CrossRef11
T Teka, ASG Faruque, GJ Fuchs. (1996) Risk factors for deaths in under-age-five children attending a diarrhoea treatment centre. Acta Paediatrica 85:9, 1070-1075
CrossRef12
T. Solomon, R.E. Phillips, J.M. Felix, M. Samuel, G.A. Dengo, A. Schapira, R.A. Saldanha. (1994) Hypoglycaemia in paediatric admissions in Mozambique. The Lancet 343:8890, 149-150
CrossRef13
Guerrant, Richard L., Bobak, David A., . (1991) Bacterial and Protozoal Gastroenteritis. New England Journal of Medicine 325:5, 327-340
Full Text14
R.E. Phillips, T. Solomon. (1990) Cerebral malaria in children. The Lancet 336:8727, 1355-1360
CrossRef15
(1990) Hypoglycemia during Diarrhea in Childhood. New England Journal of Medicine 323:14, 1000-1001
Full Text16
Keusch, Gerald T., . (1990) Vitamin A Supplements — Too Good Not to Be True. New England Journal of Medicine 323:14, 985-987
Full Text17
N.G. Kawo, A.E. Msengi, A.B.M. Swai, D.G. McLarty, L.M. Chuwa, K.G.M.M. Alberti. (1990) Specificity of hypoglycaemia for cerebral malaria in children. The Lancet 336:8713, 454-457
CrossRef18
Haymond, Morey W., . (1990) Diarrhea, Malnutrition, Euglycemia, and Fuel for Thought. New England Journal of Medicine 322:19, 1390-1391
Full Text
- Letters
