Journal of Perinatology (2011) 31, 430–433 r 2011 Nature America, Inc. All rights reserved. 0743-8346/11 www.nature.com/jp ORIGINAL ARTICLE Measurement of and glycated albumin in umbilical cord: evaluation of the glycemic control indicators in neonates

M Koga1, J Murai1, H Saito1, Y Yamada2, T Mori2, S Suno2, K Takeuchi3, S Suzuki4, K Fujieda{,4 and S Kasayama5 1Department of Internal Medicine, Kinki Central Hospital, Hyogo, Japan; 2Department of Obstetrics and Gynecology, Kinki Central Hospital, Hyogo, Japan; 3Department of Clinical Laboratory, Kinki Central Hospital, Hyogo, Japan; 4Department of Pediatrics, Asahikawa Medical College, Asahikawa, Japan and 5Department of Medicine, Nissay Hospital, Osaka, Japan

these glycated proteins are suggested to be involved in the Objective: As neonatal blood contains a high proportion of fetal development and progression of chronic diabetic complications.1 hemoglobin (HbF), it is difficult to use high-performance liquid Among these glycated proteins, glycated hemoglobin (GHb) is chromatography (HPLC) method, latex-immunoturbidimetry (LA) method commonly used as the gold standard to monitor glycemic control 2,3 and enzymatic methods, which determine hemoglobin A1C (HbA1C) in order in patients with mellitus. Although high-performance to provide the glycemic control indicators of neonates. In this study, we liquid chromatography (HPLC) is most frequently used for the evaluated glycated hemoglobin (GHb) and glycated albumin (GA) as determination of GHb, there are other methods available such as appropriate indicators of the glycemic control in the neonatal period. latex-immunoturbidimetry (LA) method, enzymatic method and Study Design: Umbilical cord blood samples collected during delivery affinity method. Since lifespan of erythrocytes is approximately were subjected to measurements of GHb (HPLC methods using two 120 days, GHb reflects the plasma for the past few months. different instruments, LA method, enzymatic method and affinity method) Furthermore, GHb measurements are affected by variant and serum GA. hemoglobin and some diseases that shorten the lifespan of erythrocytes, such as hemolytic anemia and renal anemia. Thus, Result: HbA levels determined by the HPLC method, the LA method 1C GHb does not properly represent the status of glycemic control in and the enzymatic method were as low as <3.0% in all the cases. such conditions.4,5 Although GHb determined by the affinity method was 3.6 ± 0.2%, this Neonatal diabetes mellitus (NDM) is a generic term for diabetes method may not measure accurately the values of glycated HbF plus that develops before early infancy, and one in every 300 000 to glycated HbA. Serum GA was 9.4 ± 1.1%. 500 000 neonates suffers from this disease.6,7 Although the Conclusion: We speculate that serum GA, but not GHb, could be used as pathological condition of NDM is -dependent diabetes glycemic control indicators in neonates. mellitus, the etiology of NDM is different from that of type 1 Journal of Perinatology (2011) 31, 430–433; doi:10.1038/jp.2010.144; diabetes mellitus. NDM is defined as that it develops within the first published online 16 December 2010 6 months of life, while mellitus develops in the period after 7 months of life. Recently, the involved in the Keywords: glycated albumin; glycated hemoglobin; neonate; glycemic control development of various pancreatic b-cells and their functions have been identified as the causes of NDM.8,9 As GHb levels cannot be used as an indicator of glycemic control status in patients with Introduction NDM, most of these patients receive insulin therapy to monitor plasma glucose levels alone. It is known that glycation among various proteins is increased in Neonatal blood contains a high proportion of fetal hemoglobin diabetic patients compared with non-diabetic subjects. Some of (HbF). Immediately after birth, HbF accounts for 80 to 90%, while Correspondence: Dr M Koga, Department of Internal Medicine, Kinki Central Hospital, hemoglobin A (HbA) accounts for only 10 to 20%. During postnatal Kuruma-zuka 3-1, Itami, Hyogo 664-8533, Japan. periods, HbF is gradually replaced by HbA, and HbA accounts for E-mail: [email protected] most of Hb by 6 months after birth. Therefore, because HPLC {Prof Kenji Fujieda deceased on 19 March 2010. Received 13 July 2010; revised 8 September 2010; accepted 16 September 2010; published online method, LA method and enzymatic methods determine hemoglobin 16 December 2010 A1C (HbA1C) specifically, these methods cannot provide neonatal HbA1C and GA in umbilical cord M Koga et al 431 glycemic control indicators. In addition, serum 1,5-anhydroglucitol Table 1 Levels of GHb in umbilical cord blood, serum GA, hemoglobin, serum levels are known to appear to be low in the neonatal period.10 albumin and plasma glucose measurements, which measure absolute amount of n Mean±s.d. glycated proteins, are also low, because serum protein levels are low in neonates.11,12 GHb (%) On the other hand, as serum glycated albumin (GA) levels are HPLC method (HA-8160) 5 <3.0 expressed as ratios relative to serum albumin levels, serum GA HPLC method (G8) 5 1.5±0.2 levels are not affected by serum protein levels.12 Accordingly, we LA method 5 <3.0 Enzymatic method 4 0.8±0.3 hypothesized that serum GA would be an appropriate indicator Affinity method 5 3.6±0.2 of glycemic control status in the neonatal period. In this study, we determined GHb levels using various methods and serum Serum GA (%) 4 9.4±1.1 GA levels of umbilical cord in order to evaluate the validity of Serum albumin (g per 100 ml) 4 3.7±0.2 our hypothesis. Plasma glucose (mg per 100 ml) 3 94±27 Hemoglobin (g per 100 ml) 5 16.4±1.3

Abbreviations: GA, glycated albumin; GHb, glycated hemoglobin; HPLC, high-performance Methods liquid chromatography; LA, latex-immunoturbidimetry. Participants Five different methods were used to determine GHb. The subjects were five pregnant women who did not suffer from any complications. The umbilical cord blood samples collected during delivery were used in this study. The institutional committee abHb F Hb A0 approved the protocol of this study, and all participants gave their 640 mOD written informed consent. 15%

Analysis Plasma glucose, serum albumin and hemoglobin were determined by standard laboratory assays. GHb levels were Hb F determined by the following five methods: HPLC method using 13 ADAMS-A1C HA-8160 (Arkray, Kyoto, Japan) and that using HLC- 14 Hb A0 723 G8 (Tosoh, Tokyo, Japan), LA method (RAPIDIA Auto HbA1C- 15 L; Fujirebio, Tokyo, Japan), enzymatic method (Nordia-N HbA1C; Sekisui Medical, Tokyo, Japan)16 and affinity method (GHb V, A1c affinity mode; Tosoh).17,18 All these methods were corrected by using Japan Diabetes Society lot 2 GHb19 and the normal control 0% range of GHb for adults is from 4.3 to 5.3%. Serum GA was 0 30 6090 (sec) 0.0 0.5 1.0 determined by enzymatic method using albumin-specific protease, No A1c peak ketoamine oxidase and albumin assay reagent (Lucica GA-L; Asahi Figure 1 Chromatogram of umbilical cord blood. (a) ADAMS-A HA-8160. Kasei Pharma, Tokyo, Japan).20,21 GA was hydrolyzed to amino 1C (b) HLC-723 G8. Hb F, hemoglobin F; Hb A0, hemoglobin A0;A1C, acids by an albumin-specific proteinase, and then oxidized by hemoglobin A1C. ketoamine oxidase to produce hydrogen peroxide, which was measured quantitatively. Serum GA levels were calculated as the percentage of GA relative to total albumin, which was measured in 22 the same serum sample using a new bromocresol purple method. the prominent peak of HbF was followed by the small peak of Normal control of serum GA levels for adults is from 11.0 to 16.0%. HbA1C, and the level was determined as 1.5±0.2% (Figure 1b). By Results are expressed as mean±s.d. the LA method, GHb levels were below the measurement sensitivity. The result of measurement using the enzymatic method was as low as 0.8±0.3%. By the affinity method using borate, a clear peak of Results GHb was observed and the level was 3.6±0.2%, which was slightly The results of measurements of glycated proteins in umbilical cord lower than the lower limit of normal controls for adults. Serum GA blood are shown in Table 1. By the HPLC method using HA-8160, level was 9.4±1.1%, which was slightly lower than the lower limit the peak of HbA1C that was eluted a little behind HbF was of normal controls for adults. Although serum albumin level was undetected (Figure 1a). By the HPLC analysis using HLC-723 G8, slightly low (3.7±0.2 g per 100 ml), both plasma glucose

Journal of Perinatology HbA1C and GA in umbilical cord M Koga et al 432

(94±27 mg per 100 ml) and Hb levels (16.4±1.3 g per 100 ml) levels.25–27 In the previous reports, the affinity method was used to were within the normal ranges. determine GA levels in umbilical cord serum.27 This study is the first report of the results of GA measurement using the enzymatic method. Discussion Serum GA levels, but not GHb levels determined by the affinity By two HPLC methods, LA method and enzymatic method, GHb method, may be used as glycemic control markers of NDM. The levels were considerably low compared with normal control ranges enzymatic method can determine serum GA levels by auto- for adults. As these techniques determine glycated b-chains analyzer, and thus can give the results within short time. In the specifically, glycated g-chains derived from HbF cannot be future, we need to accumulate data on serum GA levels of the determined as GHb. In measurements using the former HPLC patients with NDM to verify whether serum GA levels can be used as instrument, the peak of HbF overlapped with that of HbA1C because an indicator of glycemic control in these patients. of poor separation by HPLC, and GHb levels were measured high in patients who persistently had HbF.22 As a result of subsequent improvement of separation by HPLC, the models released after Conflict of interest HA-8160 (Arkray) and GHb V (Tosoh) have been able to separate The authors declare no conflict of interest. HbF from HbA1C. As HbA1C level is calculated by dividing the measured HbA1C by the whole Hb, the level appears to be low in blood samples containing a large amount of components other References than HbA.23 Neonatal blood contains a large amount of HbF, whereas HbA accounts for only 10 to 20%. For this reason, 1 Cohen MP. Nonenzymatic glycation: a central mechanism in diabetic micro- HbA levels are determined extremely low and such data cannot vasculopathy? J Diabet Complications 1998; 2: 214–217. 1C 2 Koenig RJ, Peterson CM, Jones RL, Saudek C, Lehrman M, Cerami A. Correlation of be used in the clinical setting. HbA1C levels determined by glucose regulation and hemoglobin AIc in diabetes mellitus. N Engl J Med 1976; 295: HLC-723 G8 are considered to contain a part of the peak of 417–420. HbF (Figure 1b). 3 Bunn HF, Gabbay KH, Gallop PM. The glycosylation of hemoglobin: relevance to GHb levels determined by the affinity method, which reflect the diabetes mellitus. Science 1978; 20: 21–27. entire amount of GHb, can be used as an indicator of glycemic 4 Jeffcoate SL. Diabetes control and complications: the role of glycated haemoglobin, 18,24 25 years on. Diabet Med 2004; 21: 657–665. control state in the neonatal period. Although the affinity 5 Bry L, Chen PC, Sacks DB. Effects of hemoglobin variants and chemically modified method was used as the main method for measurement of GHb derivatives on assays for glycohemoglobin. Clin Chem 2001; 47: 153–163. levels, the results of its measurement in umbilical cord blood were 6 Tammaro P. Neonatal diabetes. Endocr Dev 2007; 11: 70–82. reported. According to these reports, the levels of GHb in umbilical 7 Aguilar-Bryan L, Bryan J. Neonatal diabetes mellitus. Endocr Rev 2008; 29: 265–291. cord blood samples collected from diabetic mothers were higher 8 Sperling MA. The genetic basis of neonatal diabetes mellitus. Pediatr Endocrinol Rev than those from mothers without diabetes mellitus.25,27 Later, 2006; 4(Suppl 1): 71–75. 9 Suzuki S, Makita Y, Mukai T, Matsuo K, Ueda O, Fujieda K. Molecular basis of however, the problem of binding specificity between boronic acid neonatal diabetes in Japanese patients. J Clin Endocrinol Metab 2007; 92: and glycated proteins was pointed out, and so the affinity method 3979–3985. has now become rarely used. Furthermore, the speed to form GHbs 10 Yoshioka S. New metabolic parameter for diabetes-1,deoxyglucose (1,5-anhydroglu- may be different between HbF and HbA, because g-chain in HbF citol). Shonika 1983; 24: 405–410 (in Japanese). does not have an N-terminal valine, but has glycation sites at 11 Sene´cal PE, Douville P, Simard S, Coulombe R. The relationship between serum fructosamine and albumin. Clin Chim Acta 1988; 173: 239–241. several different amino-residues. Therefore, it is difficult to assume 12 Abe F, Yano M, Minami Y, Ueda T, Chikakiyo H, Miyamoto N et al. Alterations in that the GHb levels determined by affinity method may be used as fructosamine and glycated albumin levels during childhood. Ann Clin Biochem 1989; glycemic control marker for NDM. 26: 328–331. Similar to the results by the affinity method of GHb, the levels of 13 Schnedl WJ, Lahousen T, Wallner SJ, Krause R, Lipp RW. Silent hemoglobin variants serum GA in umbilical cord blood were slightly lower than the and determination of HbA1c with the high-resolution program of the HPLC HA-8160 normal control ranges for adults. Although serum albumin levels hemoglobin analyzer. Clin Biochem 2005; 38: 88–91. 14 Chapelle JP, Teixeira J, Maisin D, Assink H, Barla G, Stroobants AK et al. Multicentre in umbilical cord serum were previously shown low,28 unlike evaluation of the Tosoh HbA1c G8 analyser. Clin Chem Lab Med 2010; 48: 365–371. serum fructosamine levels, serum GA levels are not affected by 15 Guthrie R, Hellman R, Kilo C, Hiar CE, Crowley LE, Childs B et al. A multisite serum protein (albumin) levels.12 Thus, the previously reported physician’s office laboratory evaluation of an immunological method for the results also showed that GA in umbilical cord serum and GHb measurement of HbA1c. Diabetes Care 1992; 15: 1494–1498. levels determined by the affinity method were lower than those of 16 Sakurabayashi I, Watano T, Yonehara S, Ishimaru K, Hirai K, Komori T et al. New enzymatic assay for glycohemoglobin. Clin Chem 2003; 49: 269–274. maternal blood samples. As plasma glucose levels in umbilical cord 17 Nagisa Y, Kato K, Watanabe K, Murakoshi H, Odaka H, Yoshikawa K et al. Changes in blood were lower than those in maternal blood samples, the lower glycated haemoglobin levels in diabetic rats measured with an automatic affinity HPLC. GA and GHb levels probably reflect the lower plasma glucose Clin Exp Pharmacol Physiol 2003; 30: 752–758.

Journal of Perinatology HbA1C and GA in umbilical cord M Koga et al 433

18 Moiz B, Hashmi MR, Sadaf S. Performance evaluation of ion exchange and affinity 23 Felner EI, McGrath M. Inaccurate hemoglobin A(1C) levels in patients with

chromatography for HbA1c estimation in diabetic patients with HbD: a study of 129 type 1 diabetes and hereditary persistence of hemoglobin F. J Pediatr 2008; 153: samples. Clin Biochem 2008; 41: 1204–1210. 137–139. 19 Tominaga M, Makino H, Yoshino G, Kuwa K, Takei I, Aono Y et al. Japanese standard 24 Weykamp CW, Penders TJ, Muskiet FA, van der Slik W. Influence of haemoglobin reference material for JDS Lot 2 haemoglobin A1c. I: comparison of Japan diabetes variants and derivatives on glycohemoglobin determinations, as investigated by 102 society-assigned values to those obtained by the Japanese and USA domestic laboratories using 16 methods. Clin Chem 1993; 39: 1717–1723. standardization programmes and by the International Federation of Clinical Chemistry 25 Hall PM, Cawdell GM, Cook JG, Gould BJ. Measurement of glycosylated haemoglobins reference laboratories. Ann Clin Biochem 2005; 42: 41–46. and glycosylated plasma proteins in maternal and cord blood using an affinity 20 Kouzuma T, Usami T, Yamakoshi M, Takahashi M, Imamura S. An enzymatic method chromatography method. Diabetologia 1983; 25: 477–481. for the measurement of glycated albumin in biological samples. Clin Chim Acta 2002; 26 Worth R, Ashworth L, Home PD, Gerrard J, Lind T, Anderson J et al. Glycosylated 324: 61–71. haemoglobin in cord blood following normal and diabetic . Diabetologia 21 Kouzuma T, Uematsu Y, Usami T, Imamura S. Study of glycated amino acid 1983; 25: 482–485. elimination reaction for an improved enzymatic glycated albumin measurement 27 John WG, Webb AM, Jones AE. Glycosylated haemoglobin and glycosylated albumin in method. Clin Chim Acta 2004; 346: 135–143. non-diabetic and diabetic mothers, and their babies. Diabet Med 1985; 2: 103–104. 22 Krause JR, Stolc V, Campbell E. The effect of hemoglobin F upon glycosylated 28 Abe F, Shima E, Hamai S, Nagamine Y, Hirota M, Oshima I et al. Variation in the hemoglobin determinations. Am J Clin Pathol 1982; 78: 767–769. levels of serum constituents in childhood. Tokushima J Exp Med 1987; 34: 75–81.

Journal of Perinatology