Vitamin A, retinol binding protein and lipids in type 1 diabetes mellitus


Objective: A case–control study was conducted to evaluate the effects of type 1 diabetes mellitus (IDDM) on plasma levels of vitamin A (retinol) and serum levels of retinol-binding protein (RBP) and their relationship with the atherogenic indicators.

Subjects: A total of 47 randomised IDDM children were recruited from those treated at the Endocrinology Unit of the University Hospital of Granada (Spain). They were matched for age and sex with 16 healthy children.

Methods: The following parameters were measured in all patients: serum concentrations of total cholesterol, triglycerides, high (HDL, spectrophotometry), very low (VLDL) and low (LDL) density lipoprotein cholesterol (Friedewald's formula); serum levels of RBP (kinetic nephelometry); plasma vitamin A and glycosilated haemoglobin (HbA1c; high performance chromatography).

Results: Higher RBP concentrations in IDDM children (P=0.05), lower retinol levels (P=0.05) and lower vitamin A/cholesterol ratio (P=0.02) than in the control group were found; no differences in the atherogenic indicators were observed. There was a correlation between RBP and vitamin A (P=0.0001). Relationships between retinol, RBP and atherogenic indicators were demonstrated in the IDDM group (A-LDLc/HDLc (P=0.01); A-(VLDL+LDL)c/HDLc (P=0.007); RBP-LDLc/HDLc (P=0.05); RBP-(VLDL+LDL)c/HDLc (P=0.02)), and an inverse relationship was found between the vitamin A/TG ratio and HbA1c (P=0.004). The children with HbA1c>8% showed increased atherogenic indicators and lower vitamin A/CHOL and vitamin A/TG ratios than those with good control of the illness.

Conclusions: The IDDM children with poor metabolic control face a higher atherogenic risk and vitamin A ‘relative deficiency’ risk than those with good metabolic control of their illness. Relationships between retinol and RBP with atherogenic indicators were found. The results suggest that vitamin A therapeutic supplements in IDDM children may reduce or prevent atherogenic risk.


Diabetes mellitus is a state of absolute or relative insulin deficiency that leads to hyperglycaemia and to profound alterations in the lipid and protein metabolism (Campoy et al, 1996; Bayés et al, 1998). The risk of cardiovascular disease is up to five times higher in diabetic patients than in the healthy population. Clinical symptoms are not usually apparent in children and young diabetics, but certain biochemical anomalies may be detected, especially in the markers of atherogenic risk (lipids and plasmatic lipoproteins), as may the presence of other signs, such as hypertension, which is conducive to the progression of vascular disease and should be studied and monitored from the onset of diabetes. Children with diabetes mellitus type 1 (IDDM) that is not under strict control tend to develop hypercholesterolemia, and micro- and macroangiopathy and thus are at great risk of cardiovascular disease (Romero et al, 1992); those patients with good metabolic control of the disease present normal levels of lipids and lipoproteins (Bayés et al, 1991; Laakso, 1995).

Children who present IDDM are classed in the group at risk of vitamin A and β-carotene deficiency. In cases of poorly controlled IDDM, there is a lower bioavailability of retinol (Basualdo et al, 1997). However, the high levels of β-carotene and vitamin E that have been observed in patients with IDDM could be beneficial as antioxidants, although they seem insufficient to prevent oxidative damage (Hozumi et al, 1998). This hypothesis is supported by the detection of an increase among children with IDDM in oxidative stress in plasma (Sato et al, 1979), in low-density lipoproteins (Tsai et al, 1994) and in erythrocytes (Jain et al, 1989).

Retinol is stored in the form of retinol esters in the liver. Before release, these esters are hydrolysed, and the free alcohol is linked to a specific protein, retinol-binding protein (RBP). The retinol–RBP compound is then secreted into the blood flow, where it is bonded to prealbumin by means of a non-covalent link. This macrocompound is transported in the blood to the target tissues that possess specific receptors for RBP and where the retinol is transferred into the cells. Recent in vivo studies have shown that retinol stimulates and regulates the release of its binding protein RBP into plasma (Thurnham & Northrop-Clewes, 1999; Biesalski et al, 1999).

It has been shown that plasma concentrations of vitamin A and RBP fall in adults with IDDM (Basu & Basualdo, 1997), although this is not the case for type 2 diabetics, who present normal concentrations of these substances (Abahusain et al, 1999). Nevertheless, another study has shown that totally different results are obtained with diabetic children; no differences in plasma concentrations of retinol were found between the diabetic children and a control group (Hozumi et al, 1998).

On the other hand, plasma concentrations of retinol have been related to those of cholesterol and triglycerides (Smith et al, 1992).

The aim of this study is to demonstrate the influence of IDDM and its metabolic control on plasma concentrations of vitamin A and serum levels of RBP and lipid atherogenic risk indicators, and to establish whether there exists any relationship between these factors.



We studied a total of 63 children divided into two groups. The first (G-I) comprised 47 randomised children with IDDM and a mean age of 11.91±1.60 (mean±s.d.) y, range 8.25–14.33; the development time of the disease was 5.99±2.66 y, range 1.50–13.16. Of these 47 patients, 23 were boys and 24 were girls. All were recruited from those attending the Paediatric Endocrinology Unit at the University Hospital of Granada and diagnosed as presenting IDDM, according to the current classification criteria of the Expert Committee for the Diagnosis and Classification of Diabetes Mellitus (The Expert Committee on the Diagnosis and Classification of Diabetes Mellitus, 1997). All the patients included in the study took part in a ‘home-based self-help optimisation programme’ (Bayés et al, 1991); in addition, all received calibrated doses of fast and slow insulin and an appropriate nutritional and physical exercise plan. Such a programme, particularly when excellent metabolic control is achieved (HbA1c≤7), has been described as exceedingly effective in preventing the onset of complications or in significantly retarding the rate of progression (Sochett & Daneman, 1999).

Following the American and British Associations for IDDM metabolic control (Butler et al, 1995), based on HbA1c levels, the IDDM children were classified as follows: (1) children with good control of their illness would be those with HbA1c≤7% (in our study, n=10); (2) IDDM children with moderate metabolic control would be those with HbA1c between 7.1 and 8 (in our study, n=12); and (3) IDDM with a bad metabolic control of the illness would be those with HbA1c>8% (in our study, n=25).

The second group (G-II) was made up of 16 healthy children (eight boys and eight girls) as a control group, matched for age (11.75±1.83 y; range 8.50–14.50 y) and sex with the study group, and were selected after screening 30 apparently healthy children. This group was selected from children who attended the University Hospital of Granada for minor surgery (strabismus, cryptorchidy, phimosis, growth studies, etc). When samples were obtained, none of them presented any disease that might interfere with the results of the protocolised study, and none were receiving any medical treatment.

The Ethical Committee of the ‘San Cecilio’ University Hospital of Granada approved the project. Informed consent was obtained from the parents or guardian of each child before inclusion in the study.


All samples were obtained by venous puncture coinciding with other biological examinations requested by the endocrinology unit or other paediatric units at the University Hospital. The samples were placed in three tubes, one containing sodium fluoride and potassium oxalate (to determine HbA1c, which was done immediately), another with EDTA as an anticoagulant (to measure vitamin A) and the other containing a coagulation activator (to determine the other biochemical parameters). These were then centrifuged at 3500 rpm for 5 min. In every case, the serum or the plasma obtained in each tube was separated and immediately frozen to −70°C until it was analysed.

The following techniques were used to determine the biochemical parameters: an enzymatic colorimetric test for total cholesterol (mg/dl; Boehringer Mannheim CHOD-PAP method) and serum triglycerides (mg/dl; Boehringer Mannheim GPO-PAP method; Kerscher et al, 1985); high-density lipoprotein cholesterol (HDLc; mg/dl) by an in vitro enzymatic test (Sugiuchi et al, 1995); low- and very-low-density lipoprotein-cholesterol (LDLc and VLDLc), were calculated according to Friedewald's formula (Friedewald et al, 1972); vitamin A was determined by high performance liquid chromatography (HPLC; Gonzalez-Corbellá et al, 1994); RBP was determined by kinetic nephelometry (Tomlinson et al, 1990). Glycosilate haemoglobin (HbAlc) was measured by HPLC (Sale & Moussa, 1985).

Statistical analysis

A test was performed to ensure rejection of extreme samples (α=5%); basic descriptive statistics were obtained, including the mean, standard deviation and maxima and minima for each variable analysed. The ANOVA test was carried out for each case to determine whether or not variances were equal. Comparison of the means for independent samples was done using the Student and/or Welch tests. Linear correlations were established and the curves fitted to the relations obtained for all parameters analysed and for each group. The Pearson ‘r’ and Fisher ‘Z’ transforms were used to compare correlation coefficients. The minimum level of significance considered was P<0.05. The biostatistical study was designed using the methodologies described in Martín-Andrés and Luna (1998) and the statistical program SPSS version 10.0 was used.


Table 1 shows the differences between the two groups from the comparison of means using the Student t/Welch test for independent variables. The serum concentration of RBP was higher in the group of diabetic children in our study. Plasma levels of vitamin A and the vitamin A/cholesterol ratio were lower in IDDM children than in the control group. No statistically significant differences were found between the remaining variables analysed for the two groups.

Table 1 Comparison of the mean values of the biochemical variables analysed (Student's/Welch t-test) in children with type 1 diabetes mellitus (G-I) and in healthy children (G-II; control group)

Vitamin A in plasma correlated in a simple, direct linear fashion with serum concentrations of RBP, both in the group of diabetic children (n=47, r=0.52, P<0.0001) and in the control group (n=16, r=0.81, P<0.0001). Furthermore, in the diabetic group (n=47), we found significant direct, linear correlations of vitamin A with the VLDLc (r=0.42, P=0.003), and with the markers for atherogenic risk (VLDLc+LDLc)/HDLc, r=0.39, P=0.007; LDLc/HDLc, r=0.37, P=0.01). Retinol plasma concentrations were also correlated with TG levels (r=0.42, P=0.003) and with lipids (TG+CHOL; r=0.36, P=0.02). These correlations were not significant among the control group, and after comparing the correlation coefficients by the Fisher Z transform test, no significant differences were found between the coefficients for the control group and those for the diabetic children. Vitamin A/TG correlated inversely with glycosilate haemoglobin (r=−0.41, P=0.004) and development time of the illness (r=−0.34, P=0.02) in G-I.

RBP correlated in a direct, simple, linear way with the markers for atherogenic risk (triglycerides (r=0.42, P=0.004), VLDLc (r=0.42, P=0.004), LDLc/HDLc (r=0.33, P=0.04) and (VLDL+LDL)c/HDLc (r=0.35, P=0.02)) in the group of children with IDDM. The Fisher Z transform test showed there were no statistically significant differences between the correlation coefficients obtained for these parameters in the control group and those for G-I.

Statistically significant positive correlations of CHOL, TG, VLDLc, LDLc and atherogenic ratios, LDLc/HDLc and (VLDLc+LDLc)/HDLc, with HbA1c (r=0.38, P=0.008; r=0.42, P=0.003; r=0.42, P=0.003; r=0.44, P=0.002; r=0.42, P=0.003; r=0.42, P=0.003, respectively) were found.

After the IDDM children group classification following the criteria of the American and British Associations for IDDM metabolic control (Butler et al, 1995), the results obtained in our groups means that 53% of the patients included in our study had poor metabolic control of their illness, 26% of the patients presented moderate metabolic control of their diabetes and 21% of the IDDM children showed good metabolic control. Table 2 shows the statistical differences found between the two groups for the biochemical parameters studied. Among the IDDM children, when the metabolic control of the illness is worsening, there is a clear increase in the atherogenic risk indicators and also a diminution in the vitamin A/CHOL and vitamin A/TG ratios.

Table 2 Comparison of the mean values of the biochemical variables analysed (Student's/Welch t-test) in children with type 1 diabetes mellitus (G-I) in the three subgroups obtained after its classification following their own metabolic control, measured by HbA1c (good control, HbA1c≤7%; moderate control, HbA1c 7–8%; poor control, HbA1c>8%) and in the healthy children group


Children with type 1 diabetes mellitus do not present a higher atherogenic risk than do healthy children

Table 1 shows that the concentrations of total cholesterol and triglycerides in the diabetic children (G-I) who followed the ‘home-based self-control therapeutic optimisation programme’ did not differ statistically from the values found for the group of healthy children (G-II); these data have been confirmed by other authors (Peña et al; 1994). No statistically significant differences were found in the serum concentrations of HDLc, LDLc and VLDLc with respect to the control group. Similar results were obtained on comparing the atherogenic ratios [LDLc/HDLc and (VLDL+LDL)c/+HDLc] (Table 1). These data suggest that children with IDDM who follow the ‘home-based self-control programme’ are in a situation of equal atherogenic risk to healthy children. This fact might be explained by the home-based self-control, together with insulin therapy and optimum lifestyle, as discussed in Bayés et al (1991). These results agree with those obtained by Kobbah et al (1988), who examined a group of IDDM children aged 3–15 y.

Influence of metabolic control of the illness on atherogenic risk

In the present study, 53.2% of the children presented poor control of the illness. These data show that, despite the hard work of the Paediatric Endocrinology Clinical Units to ensure that all diabetic children control their illness, in order to prevent short-, medium- and long-term complications, the results are not encouraging. Achieving the desired metabolic control among a majority of the children affected is a difficult task.

Therapeutic optimisation of diabetic patients is associated with a significant decrease in levels of HbA1c (Bagdade et al, 1990). Ostlund et al (1989) have reported that among children with IDDM the plasma levels of cholesterol and triglycerides are closely related to HbA1c. Thus, analysis of the influence of control of the illness on the potential development of hypercholesterolemia reveals that 48% of the children with poor control of their illness present a moderate to severe degree of hypercholesterolemia. The reduction of 1% in HbA1c is associated with a fall of 2.2% in cholesterolaemia and of 8% in triglyceridemia. These results show that a two-unit reduction in HbA1c, as described by Ostlund et al (1989), produces a fall in the incidence of hypercholesterolaemia, as only 31.8% of the diabetic children with HbA1c≤8% present moderate to severe hypercholesterolemia; moreover, none of the children with good control of their illness presented severe hypercholesterolemia. Similar conclusions have been drawn by Wolffenbuttel et al (1990). This study also reveals a simple, direct correlation of HbA1c with triglyceridemia (n=47, r=0.42, P=0.003). These results coincide with those of other authors (Winocour et al, 1989; Peña et al, 1994). Atabani et al (1989) found positive correlations between serum levels of total cholesterol and HbA1c (r=0.36, P<0.05). Our study presents a statistically significant correlation of VLDLc with HbA1c (r=0.42, P=0.003), together with a positive significant correlation of the different atherogenic indices with HbA1c (LDLc/HDLc, r=0.42, P=0.003; (VLDLc+LDLc)/HDLc, r=0.42, P=0.003).

The data obtained in this study show that diabetic children with poor control of their illness are at increased atherogenic risk due to the significant increase in mean serum concentrations of CHOL, LDLc, VLDLc, LDLc/HDLc, (VLDL+LDL)c/HDLc and TG with respect to children with good control of their illness (Table 2). Comparison of diabetic children with a moderate control of their illness (7%<HbA1c≤8%) with respect to the former group (HbA1c≤7%) reveals that above HbA1c≥7% there is a statistically significant increase in the concentrations of CHOL, VLDLc and TG, with no variation in the marker indices of atherogenic risk.

These results demonstrate that poor control of the illness produces a marked change in lipid metabolism, with a significant increase in atherogenic risk in the diabetic child, thus producing medium- to long-term clinical effects that negatively affect the quality of life. Thus, evaluation of the atherogenic risk of the diabetic population must be preceded by an initial classification into categories of good, moderate or poor control of the illness.

Is there any relation between retinol and/or RBP with the lipid parameters that are markers of atherogenic risk in IDDM?

Experiments with animals have shown that chemically induced diabetes is associated with alterations of the patterns of antioxidant enzymatic activity, especially when diabetic control is sub-optimum. The main non-enzymatic antioxidants are vitamins C, E and A; to ensure their efficacy against free radicals, their levels within plasma and tissues must be adequate and the nutritional intake of these vitamins must be optimum (Gey, 1995).

Krill et al (1997), in a study of adult diabetic insulin-dependent patients, found that plasma concentrations of retinol and serum levels of RBP were significantly lower than in healthy adults. The results of the present study do not agree with earlier published data, and reveal significantly higher serum concentrations of RBP among children with IDDM than among the control group (2.39±0.13 vs 2.07±0.08 mg/dl, P<0.05, Table 1); nevertheless, the values reported by Krill et al (1997) were obtained from adults with IDDM. Most of the recent studies done in animals show the same results as Krill et al, but no studies in diabetic children are reported in the literature except the one by Basu et al (Basu et al, 1989), who reported lower serum RBP levels in diabetic children with respect to healthy children.

In agreement with Basu et al (1989, 1997, Martinoli et al (1993) and Krill et al (1997), we found significantly lower plasma concentrations of vitamin A among the children with IDDM than in the control group (1.03±0.03 vs 1.17±0.06 μg/ml, P<0.05, Table 1). This reduction in the plasma concentration of retinol among the children with IDDM might be a consequence, as suggested by Martinoli et al (1993), of an increase in the consumption of vitamin A among the IDDM group resulting from the process of eliminating free radicals of oxygen. The IDDM children in our study presented a mean illness development period of 5.99±2.66 y, and do not currently present any hepatic, renal or thyroid-type complications that might affect the transport of vitamin A and RBP metabolism. Martinoli et al (1993) showed in a large-scale study of insulin-dependent diabetics, classified by age, that vitamin A levels remain lower among younger patients than among adults aged 45 y and over, who presented clear signs of microangiopathic complications.

None of the patients in the study group showed signs of insufficiency (<0.20 μg/ml) or deficiency (<0.10 μg/ml) of vitamin A (Lindblad et al, 1998). Levels remained within the range of 0.61–1.67 μg/ml, while mean plasma concentrations were higher (1.03 μg/ml) than those found in Spanish adults with insulin-dependent diabetes (Olmedilla et al, 1997).

Plasma concentrations of retinol have been related to those of cholesterol and triglycerides. Studies of patients with hypercholesterolemia have revealed that plasma concentrations of vitamin A and RBP are higher than among a healthy control group (Smith et al, 1992). The present study also reveals that children with IDDM present simple, direct, linear correlations between vitamin A and RBP with triglyceridemia and with the atherogenic indices, finding no statistically significant differences in plasma concentrations of vitamin A and serum levels of RBP between the diabetic children presenting severe hypercholesterolemia and those with cholesterol levels of less than 200 mg/dl (NCEP, 1992). Vitamin A has also been correlated in a direct, linear way with plasma concentrations of total lipids among the group of children with IDDM (n=47, r=0.36, P<0.02). Herbeth et al (1991) obtained similar results in healthy children aged 10–15 y. These results suggest that RBP performs a protective function in situations of hypertriglyceridemia or atherogenic risk, perhaps in order to improve the transport of retinol (diminished) as an antioxidant to where it is required, or perhaps for some other important, but as yet unknown, function. Further investigation is required to determine the role of RBP in IDDM and to define it as a marker for atherogenic risk (Baena et al, 1999).

Levels of vitamin A with respect to cholesterol (vitamin A/CHOL) were significantly lower among the children with IDDM than among the control group (0.60±0.02 vs 0.71±0.04 μg/mg, P<0.02; Table 1). This was not so, however, for the vitamin A/TG index, which did not reveal significant differences between the two study groups. These results agree with those of Martinoli et al (1993) for patients with IDDM.

Krill et al (1997) found no relation between vitamin A and glycosilate haemoglobin, and concluded that control of glucose in the blood is not related to low levels of vitamin A among type 1 diabetics. Our results confirm this; no statistical differences were found in plasma vitamin A concentrations among the children with good to poor control of their illness and there was no correlation between vitamin A and glycosilate haemoglobin (Table 2). The vitamin A/CHOL and vitamin A/TG indices, which are markers of the proportion of vitamin A to CHOL or TG, are clear markers of the relative insufficiency or deficiency of this antioxidant as regards the efficient functioning of its protective action. A linear, inverse correlation was found between the vitamin A/TG index and HbA1c (n=47, r=−0.41, P<0.004), which is supported by the inverse correlation found among this group of IDDM children of the vitamin A/TG index and basal glucaemia (n=47, r=−0.41, P<0.004); Table 2 shows that the vitamin A/TG index falls significantly as metabolic control of the illness worsens, and even among children with moderate control it is lower than among children with good control; thus, those children who do not present good control of their illness have a relative deficit or insufficiency of vitamin A as regards the prevention of lipid peroxidation phenomena and the tissue damage caused by the formation of free radicals. Obviously, further research is needed to establish the value of this index, which would be indicative for possible treatment. Our results, therefore, suggest that diabetic children with moderate or poor control of their illness could benefit from a nutritional, therapeutic supplement of vitamin A. The quantity of such a supplement remains to be determined and further studies are necessary to corroborate our results.

In conclusion: (1) children with IDDM who follow a home-based self-control programme present a similar level of atherogenic risk to that of healthy children; (2) children with a poor metabolic control of their illness show higher markers of atherogenic risk than those with good metabolic control; (3) there is a relation between markers of atherogenic risk and plasma concentrations of retinol and RBP; (4) there appears to be a protective mechanism by which RBP levels increase in patients with IDDM to counteract high levels of atherogenic risk markers, which would explain the simple, direct, linear correlations found; (5) our results suggests that vitamin A supplementation in IDDM children may be useful to prevent the development of a relative insufficiency or deficiency of this vitamin.


  1. Abahusain MA, Wright J, Dickerson JW, De-Vol EB . 1999 Retinol, alpha-tocopherol and carotenoids in diabetes Eur. J. Clin. Nutr. 53: 630–635

    CAS  Article  Google Scholar 

  2. Atabani GS, Saeed BO, el Mahdi EM, Adam ME, Hassan DA . 1989 Glycated haemoglobin and other biochemical parameters in Sudanese diabetics Ann. Clin. Biochem. 26: 332–334

    Article  Google Scholar 

  3. Baena RMa, Campoy C, Blanca E, Fernández JM, Bayés R . 1999 Proteína transportadora de retinol y riesgo aterogénico en la diabetes mellitus tipo 1 Quím. Clín. 18: 76 (abstract).

    Google Scholar 

  4. Bagdade JD, Buchanan WE, Kuusi T, Taskinen MR . 1990 Persistent abnormalities in lipoprotein composition in noninsulin-dependent diabetes after intensive insulin therapy Arteriosclerosis 10: 232–239

    CAS  Article  Google Scholar 

  5. Basu TK, Basualdo C . 1997 Vitamin A homeostasis and diabetes mellitus Nutrition 13 Suppl 9: 804–806

    CAS  Article  Google Scholar 

  6. Basu TK, Tze WJ, Leichter J . 1989 Serum vitamin A and retinol-binding protein in patients with insulin-dependent diabetes mellitus Am. J. Clin. Nutr. 50: 329–331

    CAS  Article  Google Scholar 

  7. Basualdo CG, Wein EE, Basu TK . 1997 Vitamin A (retinol) status of first nation adults with non-insulin-dependent diabetes mellitus J. Am. Coll. Nutr. 16: 39–45

    CAS  Article  Google Scholar 

  8. Bayés R, Campoy C, Pérez-Bonilla E, Molina JA . 1991 Optimización del autocontrol domiciliario en niños diabéticos: Disminución del riesgo aterogénico An. Esp. Pediatr. 34: 194–202

    Google Scholar 

  9. Bayés R, Campoy C, Fernández JM, Molina-Font JA . 1998 Relationship between alfa-tocopherol and glucosilated haemoglobin in normal and in type 1 diabetes affected children Pediatr. Res. 44: 423 (abstract).

    Article  Google Scholar 

  10. Biesalski HK, Frank J, Beck SC, Heinrich F, Illek B, Reifen R, Gollnick H, Seeliger MW, Wissinger B, Zrenner E . 1999 Biochemical but not clinical vitamin A deficiency results from mutations in the gene for retinol binding protein Am. J. Clin. Nutr. 69 Suppl 5: S931–S936

    Article  Google Scholar 

  11. Butler C, Peters J, Stott N . 1995 Glycated haemoglobin and metabolic control of diabetes mellitus: external versus locally established clinical targets for primary care Br. Med. J. 310: 784–788

    CAS  Article  Google Scholar 

  12. Campoy C, Cano MD, Bayés R . 1996 Glucosilación Proteica y vitamina E en la diabetes insulino-dependiente An. Esp. Pediatr. 87 Suppl 87: S44–S45 (abstract).

    Google Scholar 

  13. Friedewald WT, Levy RI, Frederickson DS . 1972 Estimation of plasma low-density lipoprotein cholesterol concentration without use of preparative ultracentrifugation Clin. Chem. 18: 499–509

    CAS  PubMed  Google Scholar 

  14. Gey KF . 1995 Cardiovascular disease and vitamins. Concurrent correction of ‘suboptimal’ plasma antioxidant levels may, as important part of ‘optimal’ nutrition, help to prevent early stages of cardiovascular disease and cancer, respectively Bibl. Nutr. Dieta 52: 75–91

    Google Scholar 

  15. Gonzalez-Corbellá MJ, Lloberas N, Castellote MaI, López-Sabater MC, Rivero M . 1994 Determination of α-tocopherol in plasma and erythrocytes by high-performance liquid chromatography J. Chromatogr. B 600: 395–400

    Article  Google Scholar 

  16. Herbeth B, Spyckerelle Y, Deschamps JP . 1991 Determinants of plasma retinol, beta-carotene, and alpha-tocopherol during adolescence Am. J. Clin. Nutr. 54 Suppl 5: S884–S889

    Article  Google Scholar 

  17. Hozumi M, Murata T, Morinobu T, Manago M, Kuno T, Tokuda M, Konishi K, Mingci Z, Tamai H . 1998 Plasma β-carotene, retinol, and α-tocopherol levels in relation to glycemic control of children with insulin-dependent diabetes-mellitus J. Nutr. Sci. Vitaminol. 44: 1–9

    CAS  Article  Google Scholar 

  18. Jain SK, Mcvie R, Duett J, Herbst JJ . 1989 Erythrocyte membrane lipid peroxidation and glycosylated hemoglobin in diabetes Diabetes 38: 1539–1543

    CAS  Article  Google Scholar 

  19. Kerscher L, Town MH et al. 1985 The combined determination of two parameters on the same sample and in the same cuvette using the Hitachi 705 Clin. Chem. 21: 94–98

    Google Scholar 

  20. Kobbah AM, Vessby B, Tuvemo T . 1988 Serum lipids and apolipoproteins in children with type 1 (insulin-dependent) diabetes during the first two years of the disease Diabetologia 31: 195–200

    CAS  Article  Google Scholar 

  21. Krill D, O'leary L, Koehler AN, Kramer MK, Warfy V, Wagner MA, Dorman JS . 1997 Association of retinol binding protein in multiple-case families with insulin-dependent diabetes Hum. Biol. 69 Suppl 1: S89–S96

    Google Scholar 

  22. Laakso M . 1995 Epidemiology of diabetic dyslipemia Diabetes Rev. 3: 408–422

    Google Scholar 

  23. Lindblad BS, Patel M, Hamadeh M, Helmy N, Ahmad I, Dawodu A, Zaman S . 1998 Age and sex are important factors in determining normal retinol levels J. Trop. Pediatr. 44: 96–99

    CAS  Article  Google Scholar 

  24. Martín-Andrés A, Luna JA . 1998 Bioestadística para Ciencias de la Salud, 4th edn. Madrid: Norma

    Google Scholar 

  25. Martinoli L, Di Felice M, Seghieri G, Ciuti M, De Giorgio LA, Fazzini A . 1993 Plasma retinol and α-tocopherol concentrations in insulin-dependent diabetes mellitus: their relationship to microvascular complications. Internat J. Vitamin Nutr. Res. 63: 87–93

    CAS  Google Scholar 

  26. NCEP . 1992 Report of the expert panel on blood cholesterol levels in children and adolescents Pediatrics 89: 525–583

    Google Scholar 

  27. Olmedilla B, Granado F, Gil-Martinez E, Blanco I, Rojas-Hidalgo E . 1997 Reference values for retinol, tocopherol, and main carotenoids in serum of control and insulin-dependent diabetic Spanish subjects Clin. Chem. 43: 1066–1071

    CAS  PubMed  Google Scholar 

  28. Ostlund RE Jr., Semenkovich CF, Schechtam KB . 1989 Quantitative relationship between plasma lipids and glycohemoglobin in type I patients. Longitudinal study of 212 patients Diabetes Care 12: 332–336

    Article  Google Scholar 

  29. Peña M, Bayés R, Cano MD . 1994 Marcadores de riesgo aterogénico en la diabetes tipo I Rev. Diagn. Biol. 43: 193–197

    Google Scholar 

  30. Romero D, Villaba MP, Camacho C, Gonzalez F . 1992 Estrés oxidativo y su relación con la patología pediátrica An. Esp. Pediatr. 36: 85–97

    Google Scholar 

  31. Sale AK, Moussa MAA . 1985 An automated liquid chromatographic system for convenient determination of glycated hemoglobin Alc Clin. Chem. 31: 1872–1876

    Google Scholar 

  32. Sato Y, Hotta N, Sakamoto N, Matsuoka S, Ohishi N, Yagi K . 1979 Lipid peroxide level in plasma of diabetic patients Biochem. Med. 21: 104–107

    CAS  Article  Google Scholar 

  33. Smith DK, Greene JM, Leonard SB, Kuske TT, Feldman DS, Feldman EB . 1992 Vitamin A in hypercholesterolemia Am. J. Med. Sci. 304: 20–24

    CAS  Article  Google Scholar 

  34. Sochett E, Daneman D . 1999 Early diabetes-related complications in children and adolescents with type 1 diabetes. Implications for screening and intervention Endocrinol. Metab. Clin. N. Am. 28: 865–882

    CAS  Article  Google Scholar 

  35. Sugiuchi H, Uji Y, Okabe H, Irie T, Uekama K, Kayahara N, Miyauchi K . 1995 Direct measurement of high density lipoprotein cholesterol in serum with polyethylene glycol-modified enzymes and sulfated α-cyclo-dextrin Clin. Chem. 41: 717–723

    CAS  PubMed  Google Scholar 

  36. The Expert Committee on the Diagnosis and Classification of Diabetes Mellitus . 1997 Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus Diabetes Care 20: 1183–1197

    Article  Google Scholar 

  37. Thurnham DI, Northrop-Clewes CA . 1999 Optimal nutrition: vitamin A and carotenoids Proc. Nutr. Soc. 58: 449–457

    CAS  Article  Google Scholar 

  38. Tomlinson PA, Dalton RN, Turner C, Chantler C . 1990 Measurement of β-2-microglobulin, retino-binding protein, α-1-microglobulin and urine protein 1 in healthy children using enzyme-linked immunosorbent assay Clin. Chim. Acta 192: 99–106

    CAS  Article  Google Scholar 

  39. Tsai EC, Hirsch IB, Brunzell JD, Chait A . 1994 Reduced plasma peroxy radical trapping capacity and increased susceptibility of LDL to oxidation in poorly controlled IDDM Diabetes 43: 1010–1014

    CAS  Article  Google Scholar 

  40. Winocour PH, Durrington PN, Ishola M, Gordon C, Jeacock J, Anderson DC . 1989 Does residual insulin secretion (assessed by C-peptide concentration) affect lipid and lipoprotein levels in insulin-dependent diabetes mellitus? Clin. Sci. 77: 369–374

    CAS  Article  Google Scholar 

  41. Wolffenbuttel BH, Weber RF, Weeks L, Van Koetsveld PM, Verschoor L . 1990 Twice daily insulin therapy in patients with type 2 diabetes and secondary failure to sulphonylureas Diabetes Res. 13: 79–84

    CAS  PubMed  Google Scholar 

Download references


The authors thank Glenn Harding for checking and revising the English and style of the article.

Author information



Corresponding author

Correspondence to C Campoy.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Baena, R., Campoy, C., Bayés, R. et al. Vitamin A, retinol binding protein and lipids in type 1 diabetes mellitus. Eur J Clin Nutr 56, 44–50 (2002).

Download citation


  • diabetes mellitus type 1
  • vitamin A
  • retinol binding protein
  • atherogenic risk

Further reading