Keywords
Mexican children, diabetes mellitus, normoglycemia, recommended glycemic goals, hyperglycemia, hypoglycemia

Abbreviations
DM = Diabetes mellitus 
DM1 = Diabetes mellitus, type 1A 
ADA = American Diabetes Association 
BG = Blood glucose 
BMI = Body Mass Index

Palabras clave
Niños mexicanos, diabetes mellitus, metas glucémicas recomendadas, hiperglucemia, hipoglucemia, normoglucemia

Abreviaturas
DM = Diabetes mellitus
DM1 = Diabetes mellitus, tipo 1A
ADA = American Diabetes Association
BG = Glucosa sanguínea
BMI = Indice de Masa Corporal

To determine fasting glycemic levels presented by Mexican children from 1–5 years of age, who are neither overweight nor obese and who do not have DM1. 

The need to investigate and determine normoglycemia in Mexican children under the age of 6 begins with a lack of relevant published data.  Another motive for reviewing the currently recommended glycemic goals for children and adolescents with DM1 derives from the observed and well-known fact that children and adolescents who do not have DM1 do not develop microvascular diabetic complications.  Today, thanks to insulin analogues and the basal/bolus therapy regimens that they facilitate, children with DM1 have the option of achieving true euglycemia and of potentially benefitting from its advantages.

Clearly diabetic complications are related to the principal marker of DM1:  hyperglycemia, especially chronic hyperglycemia⁽¹⁾⁽²⁾.  In order to recommend or suggest safe glycemic levels for children with DM1, it is important to know what constitutes a true “normal” glycemic range among similarly-aged children who do not have DM.  Any reasonable clinical plan for the glycemic management and self-management of children and adolescents with DM1 depends on knowledge of the typical or “normal” nondiabetic values for the general infantile population of the same body weight.  These “normal” glycemic levels are not well defined. 

The information gained from the present study may help to define a clinical goal of glycemic values equivalent or at least similar to “normal” childhood values and permit a comparison between the glycemic ranges currently recommended by the American Diabetes Association (ADA) and a schema of hyperglycemic control that may be more salutary and preventive than the therapeutic recommendations currently in international use⁽¹⁾⁽²⁾, which are at any rate rarely achieved consistently^(3)(4)(5)(6). 

It is not known to what degree the current glycemic goals for Mexican children who have DM1, which are basically identical to those defined by the ADA, reflect the true glycemic values of Mexican children without DM1 and with appropriate body weight.  Precisely because these children do not exhibit chronic or near chronic diabetic hyperglycemia, they will not suffer microvascular diabetic complications unless at some point they should develop DM1 or DM2, followed by similar chronic or near chronic diabetic hyperglycemia. 

The lack of published fasting glycemic data for children who do not have DM1 and the naturally occurring wide variability in their glycemic levels have been key factors in the longstanding need, acceptance, and implementation of current official recommendations, which are liberally pragmatic in nature.  Documentation of glycemic values for Mexican children without DM1 or obesity, ideally in both the fasting and postprandial states, may clarify the glycemic levels that constitute childhood normoglycemia.  Moreover, either it could confirm the current recommendations for children who have DM1 and provide justification for the recommendations or support revision of the recommended glycemic levels in order to improve conformity to reality. 

The normal, natural glycemic range for young children without DM1 can suggest the criteria for the ranges to be indicated or suggested for children who have DM1.  For these children, the maintenance of approximate normoglycemia serves an undoubtedly preventive purpose, inasmuch as the child who does not have DM1 does not and probably will not show any of the physiological diabetic complications.  For the child with DM1, these complications are attributed to chronic hyperglycemia or recurrent hypoglycemia. 

Participants and methods

Subjects were children who attend two kindergartens administered by the Fundación para la Asistencia Educativa Familiar and located in the State of México:  one in Ciudad Nezahualcóyotl (27 September 2007) and the other in Chimalhuacán (16 October 2007). 

The criteria for inclusion in the study were:

• The mother of the subject was not born in the Republic of México. 
• The subject had not completed an 8-hour fast immediately prior to the BG determination. 
• The subject showed visual evidence of body weight superior or inferior to the age-adjusted normal body weight, that is, at the moment of the BG determination, the subject had a Body Mass Index (BMI) ³ 5th percentile or £ 85th percentile, adjusted for gender and chronological age.
• The subject had received a diagnosis of DM1 or DM2, according to the information given by parents or teachers. 
• The subject verbally refused to participate or did not wish to participate, as evidenced by tears or screams, in the BG monitoring, even if the investigators had previously received the parents’ written permission.

The monitoring took place in the open air of the patio at each kindergarten, between 7 a.m. and 10 a.m., on two weekdays.  The technical equipment used consisted of the Optium XCEED (serial number XC0 511 2108, Abbott Laboratories, México, D. F.) home BG monitor and the proprietary reactive strips (Lot #41420).  Before beginning measurement of the children’s BG, the recommended calibration of the equipment was carried out with both high and low control solutions (Abbott Laboratories, México, D. F.).  All glycemic monitoring assessed capillary blood drawn by the same person (the author) from the left thumb of each subject. 

The unpaired Student’s t test is a statistical tool that makes possible the evaluation of significant differences between the means of groups of different sizes.  This test was employed for analysis of all fasting BG value means. 

Results

A total of 438 BG measurements were made, of which 396 were those of children.  As a courtesy, the BG of the 42 adult staff members of the two kindergartens was also measured, although the results were not included in the statistical analyses for the determination of typical or “normal” BG levels in non-obese children without DM. 

In order to further optimize the determination of typical or “normal” fasting infantile glycemic values and in accord with the aforementioned criteria for exclusion from the study, the results of 93 children (24%) and of 5 adults (12%) were eliminated before statistical analysis.  The data from the 303 remaining children were analyzed (Tables 1—5). 

The glycemic values of the adults who were monitored are presented in Table 6, solely to permit gross comparison with those of the children.

Six highly significant statistical differences were found:

1.  Between the glycemic means of the children in the two kindergartens (Tables 2 and 3, P = .0001).  The glycemic mean of the children in Nezahualcóyotl was higher (81.52 mg/dL vs. 70.99 mg/dL).  The children with the higher mean showed the least dispersion.  Dispersion or variability of the results, which is quantified by the standard deviation, was 67% greater among the children from Chimalhuacán.  The amplitude of the glycemic range among the children from Chimalhuacán (46–91 mg/dL) also was greater than among the children in the other kindergarten (67–91 mg/dL).  The cause and meaning of these differences are unknown. 

2.  Between the genders in the Nezahualcóyotl kindergarten (Table 3, P = .0237).  The girls had levels of BG lower than those of the boys.  This difference between the genders did not appear in the children from Chimalhuacán. 

3.  Between the genders in the total sample of all children (Table 4, P = .02).  The girls tended to have BG levels consistently lower than those of the boys.

4.  Between all the children, compared by gender at 48–59 months of age.  The fasting BG levels of the girls were ~ 5.22 mg/dL lower than those of the boys (Table 4, P = .0028).

5.  In the distribution and frequency of the individual glycemic values (Table 5).  The majority (204 of 303 or 67.3%) of the children’s fasting glycemic values were “low” (< 80 mg/dL).  Fully 35.6% of the children monitored had therapeutic hypoglycemia, that is, their BG values were < 71 mg/dL.  Therapeutic hypoglycemia is the level at which caloric intake is indicated in order to maintain BG levels somewhat above 70 mg/dL.  No BG value > 91 mg/dL was registered among these non-obese children who did not have DM1.

6.  Between the global fasting glycemic means of the children and of the adults (Table 6, P = .0001).  This result suggests the presence of commonly reported glycemic increases with greater age. 

The results of this normative study indicate that for Mexican children from 1 to 5 years of age, who have normal body weight and who do not have DM, the mean level of fasting BG is ~ 74.1 mg/dL.  This glycemic value is assumed to represent an approximate normal fasting BG for the children studied.  According to the official recommendations for children of the same age who have DM1, this “normal” level of BG is considered practically equivalent to therapeutic hypoglycemia (£ 70 mg/dL).  But, at this level and even at lower levels, the children monitored in this study had no evident, self-reported, or other-reported hypoglycemic signs or symptoms.  Thus, it is probable that asymptomatic hypoglycemic levels in these children is physiological, not pathological. 

The children’s global mean BG in this study is almost exactly the glycemic mean (74.7 mg/dL) naturally maintained among pregnant women who do not have DM and who are not obese, during the third trimester of pregnancy⁽⁷⁾.  After birth and for at least the first 5 years of extrauterine life, it appears that healthy Mexican children tend to maintain a glycemic state equivalent to that of their last 3 months of intrauterine life.

Of the 303 children included in this study, 67.3% had BG levels £ 79 mg/dL.  The accepted definition of therapeutic hypoglycemia is any glycemic value < 70 mg/dL⁽⁸⁾.  In this sample, 35.6% of children presented at least this degree of therapeutic hypoglycemia; 9% had hypoglycemia of £ 60 mg/dL and 5% had biochemical hypoglycemia (< 50 mg/dL)⁽⁸⁾⁽⁹⁾. 

The presence of therapeutic and/or biochemical hypoglycemia was more common than expected.  The range of fasting BG values detected among these “normal” prepubertal children is ample (46–91 mg/dL) and weighted toward the lower BG levels.  Yet no child was using exogenous insulin injections or oral hypoglycemic agents.  No child reported symptoms of hypoglycemia and no sign of such was observed by the investigators.  The teachers did not comment on the presence of hypoglycemic signs or symptoms amongst their students.  In contrast, not a single hyperglycemic value (³ 100 mg/dL) was found.  In fact, no fasting BG value > 91 mg/dL was found.  It is unlikely that these children possess immediate risk for developing DM2. 

Therapeutic or biochemical hypoglycemia caused by excess or inexpert use of exogenous insulin routinely causes terror and panic among parents of children whose DM1 is treated with insulin and often among the children themselves.  But, although almost 70% of the children monitored in this study had BG levels < 80 mg/dL and although they were fasting and were not allowed to eat breakfast until after the BG monitoring of all the children in their kindergarten (89 and 214, respectively) had been completed, in this lengthy interval no hint of the anxiety or urgency that often characterize children who have DM1 and their parents when confronted with a hypoglycemic threat or episode was seen among the teachers or among these “normal” children.  In contrast, young people with DM1 who employ daily insulin therapy may be conditioned by the unpredictability and risk of additional or rapid reduction in BG levels to below 71 mg/dL, which may be dangerous.  These “normal” children’s response to real but asymptomatic hypoglycemia was not so conditioned.

As the present study shows, the symptoms of hypoglycemia and of the physical and mental uneasiness that ordinarily accompany abnormally low BG levels can be absent in the child who does not have DM, even when the glycemic concentrations are extremely low (46–60 mg/dL).  For this reason, in children with DM1 and in those without DM, the diagnosis of hypoglycemia solely by either a specific glycemic value or the presence of known symptoms is not tenable.  Habitual BG monitoring of children with DM is necessary and prudent, in order to assure timely discovery of the presence of infantile hypoglycemia. 

Other investigators^(10)(11) report that among young, healthy men from 26–45 years of age, elevated fasting BG levels (85–99 mg/dL), even when they are within the theoretical “normal” range (< 100 mg/dL)^(3)(12), are strongly associated with medium- and long-term development of DM2.  While fasting, each mg/dL of BG above 80 mg/dL⁽¹⁰⁾ increased their risk of DM2 by 6%, independently of other risk factors (for example, BMI, dyslipidemia, hypertension, cardiovascular disease, smoking).  Simply smoking tobacco increases the risk of DM2 by 36%, according to the study of 46,578 subjects whose fasting BG was < 100 mg/dL⁽¹¹⁾.  On account of their age, children have fewer risk factors than adults for the hyperglycemia that indicates the presence of untreated or inadequately treated DM.

Like the children in the present study, neonates normally have levels of BG much lower than adults.  After birth, the BG levels of the infant tend to rise at a still undefined pace⁽¹³⁾.  The results of the present study suggest that at least for the first 5 years in the life of Mexican children who do not have DM1, the pace of glycemic elevation is gradual and practically unmeasurable, especially among the girls.

The ADA, in its glycemic recommendations for persons who have DM⁽³⁾, recognizes that even adults and children who do not have DM tend to differ in their fasting glycemic values.  In spite of the fact that various normative studies, including the present one, confirm lower glycemic levels in “normal” children than in adults without DM, the official recommendations champion glycemic levels illogically higher for children who have DM than for adults who have DM.

The significant glycemic differences between children and adults found in this study highlight the large discrepancy between the glycemic values of these children who do not have DM1 and those recommended (110 – 200 mg/dL)⁽³⁾ as optimal during the night for children of the same age who have DM1.  These findings further suggest that “normal” Mexican children’s fasting BG levels are also far from the fasting values of 100–180 mg/dL currently recommended by the ADA for children from 0–12 years of age who have DM1.  It has been shown that the appearance of diabetic complications in the childhood, adolescence, and young adulthood of those who have had DM1 from an early age are related to high levels of BG, including levels found within the officially recommended ranges, which, in spite of being elevated above the normal ranges found in the current study, are frequently surpassed in everyday practice⁽¹⁾. 

Because young Mexican girls tended to have BG levels consistently lower than those of their male counterparts, the inevitable question is whether the glycemic levels officially recommended as optimal for children with DM1 should be identical for young boys and girls, as has long or perhaps always been the case. 

From the results reported here, complementary questions arise, still unanswered by the data:  Is there a positive or negative effect of BG levels below 80 mg/dL on the ability to learn in children who do not have DM1?  Supposing that the fasting BG level represents the lowest daily BG level, what is the highest typical or “normal” glycemic level, reached in the absorptive or postprandial state?  The response to this question should inform definition of the “normal” postprandial glycemic mean to be recommended and, together with the data from the present study, may help define the upper and lower limits of the normoglycemia typical of Mexican children without DM from 1—5 years of age. 

Acknowledgments

The author sincerely thanks Lic. Julio López of the Fundación para la Asistencia Educativa Familiar for coordinating and facilitating the programming and realization of the BG monitoring among the students and staff of the two institutions in his charge, and to the children and staff who with courage and good humor participated in this temporarily painful study.  Likewise, I am grateful to Lic. Sergio Martínez of Abbott Laboratories of México, for the timely provision of the Optium XCEED BG meter and the many reactive strips required for this investigation.  I thank my colleague, Lic. en Nutr. María Elena Morales, Certified Diabetes Educator, who donated her time and professional skills to the planning of this study and to the early morning blood collections.  Lastly, I express my gratitude to Lic. en Psic. Abimael Aguilar, Berlin, Germany, for his advice and orientation in the elaboration of this article. 

Bibliographic references
 

(1)	.	DCCT Research Group.  The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus.  New England Journal of Medicine, 1993, 329(14):977–986.
.	.	.
(2)	.	Khan F, Green FC, Forsyth JS, Greene SA, Morris AD, Belch JF.  Impaired microvascular function in normal children:  Effects of adiposity and poor glucose handling.  Journal of Physiology, 2003, 551(2):705–711. http://jp.physoc.org/cgi/content/full/551/2/705.  Accessed 25 September 2008.
	.
(3).	.	American Diabetes Association.  Standards of Medical Care in Diabetes—2008. Diabetes Care, 2008, 31:S12-S54.  http://care.diabetesjournals.org/cgi/content/full/31/Supplement_1/S12.  Accessed 25 September 2008.
.	.	.
(4)	.	American Diabetes Association.  Standards of medical care in diabetes:  Clinical Practice Recommendations 2007.  Diabetes Care, 2007, 30:S4–S41.
.	.	.
(5)	.	Silverstein J, Klingensmith G, Copeland K, Plotnick L, Kaufman F, Laffel L, Deeb L, Grey M, Anderson B, Holzmeister LA, Clark N.  Care of children and adolescents with type 1 diabetes.  Diabetes Care, 2005, 28(1):186–212.
.	.	.
(6)	.	www.guideline.gov/summary/summary.aspx?ss=15&doc_id=6826&nbr=4193.  Accessed 25 September 2008.
.	.	.
(7)	.	Parretti E, Mecacci F, Papini M, Cioni R, Carignani L, Mignosa M, La Torre P, Mello G.  Third-trimester maternal glucose levels from diurnal profiles in nondiabetic pregnancies:  Correlation with sonographic parameters of fetal growth.  Diabetes Care, 2001, 24(8):1319-1323.
.	.	.
(8)	.	Pérez Pastén E.  Guía para el educador en diabetes.  Soluciones Gráficas, México, DF, 1997, pág. 137.
.	.	.
(9)	.	Stobo JD, Traill TA, Hellmann DB, Ladenson PW, Petty BG.  Principles and practice of medicine.  McGraw-Hill Professional, New York , 1996, pág. 332.
.	.	.
(10)	.	Tirosh A, Shai I, Tekes-Manova D, Israeli E, Pereg D, Shochat T, Kochba I, Rudich A.  Normal fasting plasma glucose levels and type 2 diabetes in young men.  New England Journal of Medicine, 2005, 353:1454–1462.
.	.	.
(11)	.	Nichols GA, Hillier TA, Brown JB.  Normal fasting plasma glucose and risk of type 2 diabetes diagnosis.  American Journal of Medicine, 2008, 121(6):519–524.  www.amjmed.com/article/S0002-9343(08)00231-3/fulltext.  Accessed 25 September 2008.
.	.	.
(12)	.	Bernstein RK.  Diabetes solution:  The complete guide to achieving normal blood sugars.  Little, Brown, Boston, 2003.
.	.	.
(13)	.	LeRoith D, Taylor SI, Olefsky JM.  Diabetes mellitus:  A fundamental and clinical text.  3ra edición.  Lippincott Williams & Wilkins, Philadelphia, 2004, pág. 1258.
.	.	.