Viral infection of infants and children with benign transient hyperphosphatasemia

Abstract

In this study, we screened serum samples for transient hyperphosphatasemia (TH) using cellulose acetate membrane electrophoresis over a period of 3 years. In the patients found to suffer from TH, we examined the relationship between the clinical condition and viral infection. The frequency of TH was 0.26%, and all of the cases detected were in infants or young children. The female to male ratio of TH was 1.29/1. While there was no clear seasonal fluctuation or periodicity in the appearance of TH, two peaks were recognized in spring and autumn. Research on the clinical manifestations clarified that most of the TH cases had infectious diseases of the upper airways accompanied by symptoms of fever and diarrhea. We examined antibody titers for viruses causing upper airway infectious diseases and identified antibodies for enteroviruses such as Echo 22, Entero 71, and Coxsackie B4. Our results suggested that TH might be caused by an infection of the enterovirus group.

Transient hyperphosphatasemia, Enterovirus, Upper airway infectious disease

1 Introduction

In transient hyperphosphatasemia (TH), infants and young children commonly exhibit very high alkaline phosphatase (ALP; EC 3.1.3.1) activity, with levels frequently exceeding 20-fold the upper limit for normal adults. However, the ALP level usually returns to the normal range within a few months [1–3], and no severe diseases are known to correlate with excess increases of ALP activity. Based on clinical and laboratory information, Posen et al. [4] described this syndrome as TH of infancy in 1977, and two further reports on this syndrome were published in 1980 [5] and 1988 [6]. Growing evidence in the current literature suggests that TH originates from a viral infection [7,8]. However, the virus responsible for TH has not been identified, and the cause of the excess increase of serum ALP remains unknown. Serum ALP represents a group of isozymes originating mainly from bone, liver, intestine, kidney, and placenta [9,10]. In TH cases, a specific isoform fraction distinct from the common ALP isozymes mentioned above appears on electrophoresis [5,11]. Using cellulose acetate membrane electrophoretic analysis of this specific ALP isoform fraction, we tested serum samples for TH over a period of 3 years. In the 50 TH cases identified during the 3-year period from 1998 to 2001, we analyzed the relationship between TH and viral infection.

2 Materials and methods

2.1 Serum samples

This study was performed on 19,230 serum samples sent in for ALP isoenzyme analysis by clinical doctors from 1998 to 2001.

2.2 Measurement of ALP

The overall activity of ALP was measured with an Olympus AU5200 Auto Analyzer (Olympus, Tokyo, Japan) using a continuous monitoring system according to the recommendations of the SSCC, and nitrophenyl phosphate-2Na (PNPP) as the substrate. The normal range of ALP activity assessed by this method in adults was 90–270 IU l⁻¹.

2.3 Analysis of ALP isoenzyme

To measure the total ALP activity, ALP isoenzymes were separated by cellulose acetate membrane (Helena Laboratory, Urawa, Japan) electrophoresis in a tris-barbital–sodium barbital buffer solution (ionic strength 0.042) for analysis. ALP enzymatic staining was performed with 3-indoxylphosphatase-di-p-toluidine salt as a substrate, using the reaction between 3-hydroxyindole and nitrotetrazolium blue (NTB), and diformazan obtained via the reduction by the chromogenic diformazan method.

2.4 Frequency, seasonal fluctuation, age distribution, sex difference, and clinical conditions

Over the 3-year period from 1998 to 2001, the serum samples of patients with TH were investigated to determine the frequency, seasonal fluctuation, age distribution, and sex differences of TH patients, and a survey was carried out at several hospitals treating TH patients to research their clinical conditions. The definition applied for TH was based on the following ALP activity data and electrophoretic pattern of ALP isoenzymes, based on the report by Griffiths et al. [11]. (1) ALP activity is at least 3-fold the upper limit of the normal range. (2) A TH-specific isoform band is found in the fast α₂ of the electrophoretic pattern of ALP isoenzymes.

To analyze the seasonal fluctuation, we tabulated the number of cases detected month by month during the 3-year study.

2.5 Investigation of the titer of viral antibodies

The neutralization tests for viral antibodies (NT) in the serum of patients with TH were carried out using an antiserum kit (Neutralization Test for Antiserum, Denka Seiken, Tokyo, Japan), according to the manufacturer's instructions.

3 Results

3.1 Isoenzyme analysis by electrophoresis

Serum samples were subjected to ALP isoenzyme analysis by electrophoresis. Only the serum samples from TH cases in all of the 19,230 samples were found to have the peculiar isoform band of ALP isoenzyme at the position of fast α₂, between α₁ (membrane-associated, biliary ALP) and α₂ (liver ALP) (Fig. 1). The ALP activity in the serum of TH cases was elevated from 3.33-fold to 42.4-fold the upper limit of normal for adults (270 IU l⁻¹).

Figure 1

Electrophoretic patterns of ALP isoenzymes in the serum of patients with TH. 1, a control sample exhibiting macromolecular α₁ and liver type α₂ isoenzymes, collected from a patient with jaundice; 2–5, representative patterns of 50 TH found in 19,230 ALP isoenzyme analytical cases. TH cases have a TH-specific band at the position of fast α₂.

3.2 Frequency, seasonal fluctuation, age distribution, and sex difference of TH cases

A total of 19,230 serum samples were screened for TH over the 3-year period from 1998 to 2001. Fifty TH cases were identified based on high ALP activity and the peculiar band in the electrophoretic mobility of fast α₂ (Fig. 1). The frequency of TH emergence was 0.26%.

The seasonal fluctuation of TH was analyzed from the number of TH cases detected on a month by month basis over the 3-year period (Fig. 2). There was a slight increase in the number of TH patients detected in the spring and autumn, but the tendency was not clear.

Figure 2

Seasonal fluctuation of TH cases in infancy and childhood. TH cases are grouped month by month for 3 years.

All 50 of the TH patients were children less than 8 years old. The frequency of TH was highest in infants up to the age of 1 and progressively decreased with age. The female/male ratio among the TH cases was 1.29/1, and the female prevalence was especially high in infants up to the age of 1 (ratio of 1.78/1) (Fig. 3).

Figure 3

Age distribution and sex differentiation of TH cases in infancy and childhood.

3.3 Analysis of clinical conditions

The ALP activity data and clinical conditions of the TH cases were obtained from the hospitals where the patients were treated from 1998 to 2001 (Fig. 4). Two of the treated patients were a pair of 2-year-old twins. The four most common symptoms, bronchitis and pneumonia, followed by diarrhea and fever, were observed in the TH cases, symptoms frequently associated with so-called upper airway infectious diseases. The increases in ALP activity were most marked in the TH cases with bronchitis, followed by the cases with diarrhea and fever. Cases with epilepsy also showed a strong tendency towards increased ALP activity.

Figure 4

Clinical conditions and total ALP activity of 50 TH cases in infancy and childhood.

3.4 Viral antibodies in the TH cases

In the investigation of the titer of viral antibodies using serum from TH patients, Echo 22, Entero 71, and Coxsackie B4 of the enterovirus group were identified at proportions of 32%, 14%, and 8% respectively, and the Echo 22 antibody was most frequent in TH cases. In addition, other antibodies to enteroviruses such as Echo 3, 12, and 9 and Coxsackie B5, A16, B3, A9, and A4 were also detected (Fig. 5).

Figure 5

Number of TH patients positive for each antiviral antibody by serum analysis. E, Echo viruses; Ent, enteroviruses; CA and CB, Coxsackie A and B viruses.

4 Discussion

Almost half a century has passed since the first reported case of TH during infancy. Several TH cases have been reported until now, and some physicochemical analyses have been used to study ALP in the specimens taken from TH cases [5,11,12]. ALP activity in TH cases showed very high levels and was shown to be heat labile, and inhibited by l-homoarginine but resistant to l-phenylalanine. Further, electrophoretic analysis using TH specimens treated with sialidase demonstrated that the peculiar ALP in TH cases was liver type ALP with excess sialic acid. However, the cause of the ALP elevation in these TH specimens was not clarified.

In a 3-year screening of serum samples for TH by analyzing ALP isoenzyme and assessing excess elevation of ALP activity, we found 50 TH cases, all of whom were infants and young children. The ALP activity in these detected cases ranged from 3.33-fold to 42.4-fold the upper limit detected in normal adults, and the clarity of the TH isoform band observed at the position between α₁ (membrane-associated ALP) and α₂ (liver type ALP) changed according to the fluctuation of ALP activity. Cellulose acetate membrane electrophoresis was very useful for the identification of TH in infants and eliminated the need for extensive and unneeded evaluation in the hospital.

In order to clarify the relationship between the TH emergence and viral infection, we examined the frequency, seasonal fluctuation, age distribution, sex difference, clinical conditions, and viral antibodies among the TH cases.

The seasonal fluctuation data on TH cases showed small peaks in spring and autumn, following a pattern very similar to that reported by Crofton's group [6]. Temperature changes are more pronounced during spring and autumn, hence people may be more susceptible to upper airway infections at these times of the year. Many TH cases were seen to occur concomitantly with infectious diseases of the upper airways such as bronchitis or pneumonia, and twins with TH became sick at the same time. We inferred from this that elevated ALP levels in the TH cases were the result of viral infection. Many types of antibodies to enteroviruses were identified in the serum of TH patients in this study (i.e., Echo 22, 3, 12, 9, Entero 71, Coxsackie B4, B5, A16, B3, A9), suggesting that those enteroviruses might have been responsible for the TH. It will be necessary to clarify the mechanisms responsible for the co-presence of viral infections in TH patients with ALP elevation.