Early signs of microvascular complications in pediatric patients with short duration of type 1 diabetes mellitus seen in southeast Nigeria.

Background: Clinically evident microvascular complications are rarely seen among children and adolescents with type 1 diabetes mellitus (T1DM), although early signs develop during childhood and accelerate during puberty. Aim: The aim of this study was to determine the prevalence of early signs of microvascular complications in children and adolescents aged 9-19 years with a short duration of T1DM by screening for retinopathy and nephropathy.
Methods: A cross-sectional study and participants were consecutively enrolled from the Endocrinology Clinic at Federal Teaching Hospital, Abakaliki. Physical examination and mydriatic ophthalmoscopy were conducted. Three early morning spot urine specimens for albumin/creatinine ratio were estimated 3 months apart. Serum creatinine levels were estimated, and the glomerular filtration rate was calculated. Glycosylated hemoglobin (HbA1c) was determined.
Results: Twenty-four individuals participated, 15 (62.5%) were male and the mean age at diagnosis was 12.4 ± 2.3 years. The mean duration of diabetes was 23.8 ± 20.6 months. The mean HbA1c was 11.4%. Retinopathy was seen in 16.7%, whereas 33.3% had microalbuminuria. Blood pressure range was within the 50th-90th percentile for all the participants.
Conclusion: The study outcome demonstrated a high prevalence of early signs of microvascular complications such as retinopathy and nephropathy among youths with short duration of T1DM. Poor glycemic control, if not halted, is associated with early signs of microvascular complications which may become clinically evident; contrary to the belief that they are rare in childhood.

INTRODUCTION

Diabetes mellitus is associated with a significant disease burden mainly related to the development of microvascular complications.[1] Early stages of microvascular complications such as nephropathy and retinopathy are often asymptomatic with duration and magnitude of hyperglycemia strongly correlating with both the extent and rate of disease progression.[2]

Although clinically evident microvascular complications are rarely seen among children and adolescents with type 1 diabetes mellitus (TIDM), there is clear evidence that their pathogenesis and early signs develop during childhood as early as 9 years with 5 years diabetes duration or 11 years with 2 years diabetes duration and accelerate during puberty.[3]

Nephropathy is the leading cause of end-stage renal disease often preceded by the development of microalbuminuria as its first clinical sign.[4] There is widespread evidence of advanced glomerular structural changes concomitant with increased systolic and diastolic blood pressure.[5] Renal function during this phase may be increased, normal, or reduced.

Intensive insulin therapy designed to achieve improved glycemic control is able to attenuate the development of nephropathy, as assessed by urinary albumin excretion.[6] Persistent microalbuminuria, if left untreated, is often a reliable harbinger of overt nephropathy. Therefore, screening of patients with diabetes mellitus for nephropathy is now recommended to include at least twice per year measurements of urinary albumin concentrations in patients with T1DM.[7] The best approach to screen for microalbuminuria remains controversial; however, a spot urine albumin/creatinine ratio (ACR) in an early morning urine specimen has been validated and appears to be a practical option for routine clinical practice.[8]

Retinopathy is now the leading cause of new blindness in people aged 20–74 years with diabetes mellitus.[4] It is known to afflict virtually all patients with diabetes mellitus and principally a sequel of chronic microvascular complication.[9] The severity of diabetic retinopathy is closely correlated with the onset and duration of diabetes. Although retinopathy is rare in prepubescent patients with T1DM, nearly all patients with T1DM and >60% of patients with T2DM develop some degree of retinopathy after diabetes duration of 20 years.[9]

Adolescents have a higher risk of progression to vision-threatening retinopathy compared to adult patients with diabetes mellitus.[4] Lack of appropriate glycemic control and puberty are significant risk factors for the onset and progression of diabetic retinopathy.[2] Diabetic retinopathy is broadly classified into nonproliferative diabetic retinopathy (NPDR) and proliferative diabetic retinopathy (PDR) categories.[10] In general, diabetic retinopathy progresses from no retinopathy through mild, moderate, severe, and very severe NPDR and eventually to PDR.[10]

Annual comprehensive eye examination by dilated ophthalmic examination is an essential guide for children and adolescents with TIDM.[8] Until a cure for diabetes is discovered, the primary clinical care emphasis for the prevention of vision loss or renal impairment is appropriately directed at good glycemic control, early identification, accurate classification, and timely treatment of these complications.

Many patients with diabetes mellitus in Southeastern Nigeria have poor glycemic control and do not receive adequate eye care at an appropriate stage in their disease. For these reasons, this study was undertaken to answer the research question: Are microvascular complications found in Nigerian children aged 9–19 years with short duration of diabetes?

METHODS

This was a cross-sectional study carried out at the Federal Teaching Hospital Abakaliki (FETHA), Ebonyi State, Southeastern Nigeria. FETHA is a tertiary health facility that serves the state and towns in the neighboring states. The Ebonyians are mostly Agrarians. However, people from all works of life reside in Abakaliki, the state capital and location of the hospital. All the children and adolescents (36 individuals) between the ages of 9 and 19 years, followed up for T1DM in the endocrinology unit of the hospital were enrolled for the study. Individuals were recruited consecutively during routine follow-up visits.

Sociodemographic data

A questionnaire was completed recording the patient demographic data which included: name, age, gender, age at onset (diagnosis), date of diagnosis of diabetes, duration of diabetes, current treatment modality/dose, parents' educational level, and occupation (parent/child) of all the participants. All the participants were grouped into their respective social class using the Oyedeji classification.[11] This classification is based on parents/caregivers educational level and occupation.

Clinical assessment

The participants had routine clinical examination. Weight was measured without shoes or heavy clothing with a SECA® scale and recorded to the nearest 0.1 kg.[12] Height was determined to the nearest 0.1 cm with a SECA stadiometer following the protocol for height measurement.[12] Body mass index (BMI) was obtained using the formula weight (kg)/(height [m2]) and values plotted on the Centers for Disease Control Growth Charts of age-/sex-specific BMI percentiles.[13] The sexual maturity rating of the children was determined using the Tanner scale for girls and the Tanner scale for boys.[14] The Prader orchidometer was used to estimate the testicular volume. Blood pressure was measured using a mercury sphygmomanometer with an appropriate cuff. After a 10 min rest with the patient in the sitting position, blood pressure was measured two times at 5 min interval. The first and the fifth Korotkoff sounds were used to determine the systolic blood pressure (SBP) and diastolic blood pressure (DBP) measurements, respectively. The second blood pressure measurement was used as the blood pressure for the individual. Readings were compared with a nomogram for age, sex, and height. Hypertension was defined as average SBP and/or DBP that was ≥95th percentile for gender, age, and height on three or more occasions.[15]

Assessment of retinopathy

This was assessed using mydriatic ophthalmoscopy by an ophthalmologist dedicated to the study. Eyes were dilated with Mydriacyl (tropicamide) after which ophthalmic examination was carried out. Clinical findings of diabetic retinopathy were reported as no retinopathy, mild, moderate, severe, very severe NPDR, or PDR.[10]

Laboratory evaluation

Glycosylated hemoglobin

The glycosylated hemoglobin (HbA1c) was measured immunochemically in the clinic using DCA 2000®+ (Bayer Corporation). The instrument is standardized against the Diabetes Control and Complication Trial method.[16] The recommended target range for all age groups was <7.5%.[6]

Albumin/creatinine ratio

This was estimated using early morning spot urine sample. Microalbuminuria was defined as ACR of 2.5–25 mg/mmol or 30–300 μg/ml (spot urine) in males and 3.5–25 mg/mmol in females (because of lower creatinine excretion). Three early morning spot urine ACRs were estimated. Microalbuminuria was confirmed by finding two abnormal samples over a 3–4 months' period.[3]

Serum creatinine estimation

This was done using the Jaffe–Slot alkaline picrate method. Normal range was 53–115 μmol/L.[6] The estimated glomerular filtration rate was calculated using the Schwartz equation (normal range between the 5th and 95th percentile: 140 and 97 ml/min/1.73 m2).[17]

Ethical consideration

Approval was obtained from the Ethics Review Committee of the FETHA before conducting the study. Consent was obtained from caregivers of participants as well as assent from the children ≥7 years in addition to their parental consent before undertaking the study.

Statistics

Information obtained was transferred to electronic database prepared using Microsoft Office Excel 2007. All statistical analyses were performed using the Statistical Package for the Social Sciences software for Windows, version 19.0 (IBM Corp.Q, Armonk, Newyork: USA). Data are presented in tables and graphs. All distributions within groups were tested for statistically significant differences. Parametric and nonparametric tests were applied as appropriate. P < 0.05 was considered statistically significant.

RESULTS

Sociodemographic characteristics of the study population

Twenty-four of 36 children and adolescents aged 9–19 years followed up in the clinic were consecutively enrolled; 6 were excluded from the analysis because of incomplete data. Of these 6 patients excluded, 1 died before the end of the study, 1 relocated, whereas 4 were lost to follow-up. Therefore, 18 patients, 10 males, and 8 females, completed the study. All the participants were of the Black African race and Igbo ethnic tribe. The participants were aged 9–19 years (mean 14.2 ± 2.5 standard deviation [SD]). The age at diagnosis of diabetes ranged from 8 to 16 years (mean 12.4 ± 2.3 SD). Thirteen (72.2%) of the participants were in puberty, whereas 5 (27.8%) were prepubertal. Diabetes duration in months ranged from 3 to 70 months (mean 23.8 ± 20.6 SD). All were receiving twice-daily premixed insulin (humulin 70/30) at a mean dose of 1.1 unit/kg/day (range: 0.7–1.4 IU/kg/day). Blood pressure range (both systolic and diastolic) was within the 50th–90th percentile for all the participants. Sixteen (89%) were of the lower socioeconomic class and 2 (11%) were of the higher socioeconomic class. Sociodemographic and clinical characteristics of the participants are shown in Table 1.

Table 1

Sociodemographic and clinical characteristics of participants (n=18)

	n (%)
Age of participants (years)
9-<10	2 (11.1
10-14	7 (38.9
15-17	8 (44.4
18-19	1 (5.6
Mean age±SD (years), range	14.2±2.5, 9-19
Age (years) at diagnosis
<9	2 (11.1)
9-<13	6 (33.3)
13-19	10 (55.6)
Mean age at diagnosis (years), range	12.4±2.3, 9-16
Gender (male/female)	10/8 (55.6/44.4)
Parents’ education
No formal education	10 (55.6)
Primary education	4 (22.2)
Secondary education	2 (11.1)
Tertiary education	2 (11.1)
Parents’ occupation
Farming	10 (55.6)
Self-employed (carpenter, mechanic, and trading)	6 (33.3)
Civil servants (teaching/pharmacist)	2 (11.1)
Socioeconomic status
Upper socioeconomic class	2 (11)
Lower socioeconomic class	16 (89)
BMI
>+2SD	1 (5.56)
±2SD	14 (77.78)
<−2SD	3 (16.67)
HbA1c (%)
<7.5	3 (16.67)
7.5-<9.5	2 (11.11)
>9.5	13 (72.22)

SD: Standard deviation, HbA1c: Glycosylated hemoglobin, BMI: Body mass index

Microvascular complications: Retinopathy and microalbuminuria

The prevalence of diabetic retinopathy noted was 16.7% (3/18). Two males and one female were affected. Fifteen patients (83.33%) had no apparent retinopathy. All the three patients with retinopathy had mild nonproliferative retinopathy. Other ocular manifestations were also noted. They include 1 patient with cataract and 2 who had refractive error.

The prevalence of microalbuminuria was 33.3% (6/18). Macroalbuminuria was found in 1 patient. Microalbuminuria occurred in 4/11 (36%) patients with diabetes duration <2 years and 2/6 patients with HbA1c <9.5%. All the patients (3) with retinopathy also had microalbuminuria.

Linear regression analysis, with microalbuminuria and retinopathy as the dependent variable and gender, duration of diabetes, HbA1c, pubertal development, BMI SD, and socioeconomic class as covariates, is shown in Table 2.

Table 2

Linear regression analysis with retinopathy/ microalbuminuria as the dependent variable and gender, duration of diabetes, glycosylated hemoglobin, sexual maturity rating, body mass index, and socioeconomic class as covariates

Covariates	Retinopathy		P	Microalbuminuria		P
	Yes	No		Yes	No
Gender
Male	2	8	0.671	3	7	0.757
Female	1	7		3	5
Duration of diabete (years)
<2	0	10	0.034	4	7	0.387
>2	3	5		3	4
HbA1c (%)
<7.5	0	3	0.500	1	2	0.866
7.5-9.49	0	2		1	2
>9.5	3	10		4	9
Sexual maturity rate
Tanner Stage 1	0	5	0.308	0	5	0.017
Tanner Stage 2	0	3		0	3
Tanner Stage 3	2	4		4	2
Tanner Stage 4	1	1		2	0
Tanner Stage 5	0	2		0	2
BMI SD
>+2SD	0	1	0.598	0	1	0.765
−2-+2SD	3	11		5	9
<−2SD	0	3		1	2
Socioeconomic class
Lower (s)	3	13	0.50	6	10	0.289
Upper (s)	0	2		0	2

Number of participants with missing data: HbA1c (4), complete albumin/ creatinine ratio (6), ophthalmic examination (5). SD: Standard deviation, HbA1c: Glycosylated hemoglobin, BMI: Body mass index

Renal function evaluation

Serum creatinine levels of the six participants with microalbuminuria were estimated and the glomerular filtration rate calculated. Two of these participants had out of range estimated glomerular filtration rate (eGFR) of 6.152 mL/min/1.73 m2 and 48.108 mL/min/1.73 m2, whereas the other four participants had normal eGFR.

DISCUSSION

In the present study, a high prevalence of early signs of microvascular complications such as retinopathy (16.7%) and nephropathy (33.3%) was demonstrated among children and adolescents with short duration of T1DM and mean HbA1c of 11.4%. The mean duration of diabetes in this study was 23.8 months which is short compared to the mean duration of diabetes seen in most studies on the prevalence of microvascular complications among children and adolescents with T1DM. Downie et al.[18] evaluated children with diabetes duration of ≥5 years, whereas Kubin et al.[19] evaluated children with mean diabetes duration ≥4.6 years for microvascular complications. The duration of diabetes is closely associated with the onset and severity of diabetic retinopathy. The duration of diabetes is probably the strongest predictor for the development and progression of retinopathy. This is supported by the report of the Wisconsin Epidemiologic Study of Diabetic Retinopathy which identified diabetes duration as a predictor for the development of retinopathy.[9] Besides diabetes duration, the magnitude of hyperglycemia is also a strong risk factor for the development and progression of diabetic microvascular diseases. Majority (72.2%) of our participants had very poor glycemic control (HbA1c >9.5%). This may in part explain the early onset of microvascular complications in this cohort.

The mean age at the onset of diabetes mellitus was 12.4 ± 2.3 years (range: 9–16.9 years). Our observations were consistent with the findings of Swai et al.[20] in Dar es Salaam who recorded a peak age at the onset of diabetes between 10 and 19 years of age in a longitudinal incidence study lasting over a 10-year period and a similar report by Afoke et al.[21] in Nigeria. In Europe, two peaks of T1DM presentation have been observed; the childhood-onset T1DM – one occurring between 5 and 7 years of age and the other at or near puberty.[1] The peak in Sub-Saharan Africa tends to coincide with that occurring during puberty in Europe. This might be because of underdiagnosis during the childhood period in Africa. The development of nephropathy has been found to be influenced by the age at the onset of diabetes. Kofoed-Enevoldsen et al.[22] reported that nephropathy develops later in persons with the onset of T1DM before the age of 10 years than in those with onset after puberty, and the risk of persistent proteinuria declines after the age of 35 years regardless of the duration of diabetes. However, few reports exist on children with diabetes who develop diabetic nephropathy in the prepubertal period.[23]

The prevalence of retinopathy found in this study was 16.7% which is higher compared with the prevalence found among Caucasians children and adolescents with T1DM. Kubin et al.[19] obtained a prevalence of 11.8% in a population-based retrospective study which was carried out to evaluate the prevalence and risk factors of diabetic retinopathy in children with T1DM in Finland. In France, a prevalence of 4.6% was reported by Massin et al.[24] The observed difference may in part be due to the smaller population size used in this study or due to a difference in the technique used to determine retinopathy. Whereas 7-field stereoscopic fundus photography which provides greater sensitivity for detecting both background and proliferative retinopathies was used by the two studies, in this study, mydriatic direct ophthalmoscopy was used.

Our study showed a statistically significant association between retinopathy and duration of diabetes. This finding is consistent with those of Massin et al.[24] who found an association between retinopathy with only longer duration of diabetes after adjustment for age.

Higher HbA1c and puberty (Tanner Stages 3 and 4) were also associated with retinopathy in this study but did not reach statistical significance (HbA1c: P =0.500; SMR: P =0.308). The fact that 72.2% (13/18) of the patients had very poor glycemic control (HbA1c >9.5%) and were also in puberty, might explain this, at least in part. All patients with retinopathy had a duration of diabetes >2 years, HbA1c >9.5%, and were in puberty. Poor glycemic control and the period of pubertal development are significant risk factors for the onset and progression of diabetic retinopathy.[225] We found no association between retinopathy and gender (P = 0.671), BMI SD (P = 0.598), and socioeconomic class (P = 0.502). Downie et al.[18] observed, however, an association between disadvantaged socioeconomic class and retinopathy. In our study, 89% (16/18) of the patients were from disadvantaged (lower) socioeconomic class.

The type of retinopathy observed in this study was mild nonproliferating diabetic retinopathy, which is consistent with findings among children and adolescents with T1DM by other researchers.[25]

The prevalence of microalbuminuria observed was 33.3% (6/12) which is comparable with a prevalence of 38% seen by Gupta et al.[25] in India and 21.9% in Congo by Rissassi et al.[26] This is relatively higher than the Caucasian value of 12% reported by Downie et al.[18] who also demonstrated a significant association with glycemic control and a continued reduction in the prevalence of microalbuminuria in a cohort followed up longitudinally for over 8.6 years.

Statistically significant association was found between microalbuminuria and pubertal development. This finding is supported by the findings of Schultz et al.[27] who found that puberty conferred a three-fold increased risk of microalbuminuria, independent of poor glycemic control. This may, at least in part, be related to pubertal hormonal variables. Amin et al.[28] suggested that microalbuminuria risk in this age group is associated with growth hormone hypersecretion and insulin resistance, particularly in females. This study failed to demonstrate any significant relationship between glycemic control and microalbuminuria, so do most other available studies from Sub-Saharan Africa. Lutale et al.[29] and Goldschmid et al.[30] demonstrated a high prevalence of microalbuminuria and nephropathy among Tanzanians and African-American patients with diabetes mellitus. In their study, there was no association between microalbuminuria and metabolic control. In this study, microalbuminuria was found in patients with a short duration of diabetes (<2 years) in contrast to the finding of Downie et al.[18] who found an association between microalbuminuria with long duration of diabetes and high BMI among other variables. In this study, 77.8% of the participants had BMI within ± 2 SD and this BMI range has not been associated with development of microalbuminuria. Microalbuminuria is the first clinical sign of diabetic nephropathy.[7] It is associated with widespread evidence of advanced glomerular structural changes.[3] Concomitant with these changes is usually increased SBP and DBP. However, all our patients had both SBP and DBP between the 50th and 90th percentile for age, sex, and height, which is within normal values. Association with microalbuminuria was not significant (P = 0.635).

Our study demonstrated a significant association between microalbuminuria and retinopathy (P = 0.007). Parving et al.[31] and Michael et al.[32] showed that nephropathy and retinopathy almost always coexisted. The presence of microalbuminuria and renal disease is an excellent predictor of the presence of retinopathy.[24] Renal disease, as manifested by microalbuminuria and proteinuria, is another significant risk factor for the onset and progression of diabetic retinopathy.[7]

CONCLUSION

The signs of microvascular complication manifest early in children and adolescents with TIDM in Southeastern Nigeria. A significant proportion had a short duration of diabetes mellitus. Poor glycemic control and puberty are significant risk factors for the prevalence of early signs of microvascular complications in this cohort. The proportion of patients with poor glycemic control was high, and therefore, focusing on glycemic control would be of great benefit.

Study limitations

The study limitations include small sample size which makes it difficult for strong generalizing conclusions to be made and poor turnout for follow-up visits with a lack of means of communication.

Financial support and sponsorship

This study was funded by the International Society for Pediatric and Adolescent Diabetes through the Pediatric Endocrinology Training Center for Africa project.

Conflicts of interest

There are no conflicts of interest.