Vitamin A deficiency is a significant public health problem throughout the world, affecting millions of school children. Among the proposed strategies to correct the deficiency, the distribution of capsules containing a high dose of preformed vitamin A to at-risk populations is practised and supported widely by international organizations due to its immediate impact and the possibility of it being implemented via the existing health care infrastructure (Sommer, Reference Sommer1989).
Vitamin A deficiency is a public health problem in Sri Lanka (Medical Research Institute, 1998). The Government of Sri Lanka in its national plan of action aimed to achieve virtual elimination of vitamin A deficiency and its consequences (Nutrition Division, 1999). The current national programme of vitamin A supplementation, commenced in 2001, includes using an oral megadose of 210 μmol (200 000 IU) for postpartum mothers within 4 weeks of delivery, and 105 μmol (100 000 IU) for infants at 9 months with measles immunization, preschool children at 18 months with oral polio vaccine and diphtheria, pertussis and tetanus immunization, and school children in Grades 1, 4 and 7 (approximately 5-, 9- and 12-year-olds, respectively). Although the programme was initiated in 2001, not all schools were included initially due to logistic reasons and the supplementation programme was phased over a period of time. The interval between megadose supplementation in children ranges from 9 months (between first and second dose) to 3 years (between fourth and final dose) (Family Health Bureau, 2000).
Vitamin A intervention programmes require rigorous and repeated evaluation to ensure they are achieving their goal (Wasantwisut, Reference Wasantwisut2002). The national vitamin A supplementation programme in Sri Lanka has not been evaluated for its efficacy at any level since its inception. We report here an evaluation of the programme in school children in a selected area of Sri Lanka.
Materials and methods
Study site and subjects
A cross-sectional study was carried out in the Yatiyantota Medical Officer of Health (MOH)/Divisional Director of Health Services (DDHS) area in the Kegalle District of the Sabaragamuwa Province from May to August 2002. Sri Lanka has nine provinces and each province is divided into districts and each district is divided into a number of MOH/DDHS areas administered by a MOH/DDHS. Each MOH/DDHS area has a population ranging from 70 000 to 100 000. The country is thus divided into 270 MOH/DDHS areas. In a survey carried out in 1995–6, the Sabaragamuwa province had the highest percentage of children with low serum retinol concentrations (less than 0·7 μmol/l) (Medical Research Institute, 1998). Approximately half of the school children in the area were supplemented with the vitamin A oral megadose (105 μmol) at the time of the survey. Each dose contained 105 μmol vitamin A and 9·1mg vitamin E. Capsules were manufactured by Accucaps Industries Ltd (Windsor, Canada). As supplementation was carried out in Grades 1 and 4, the survey was carried out among children in Grades 1, 2, 4 and 5. A sample of ninety-five children was required from each category to estimate a prevalence of night blindness of 1 % with a 95 % CI ranging from 0 to 3 %.
There were seventy-six schools in the area. The list of schools of the Yatiyantota DDHS area and their supplementation status were obtained from the DDHS. Supplemented children were recruited from seven randomly selected schools in which the supplementation programme was carried out. Non-supplemented children were recruited from two randomly selected schools in which the supplementation was not carried out.
Evaluation of ocular manifestations of vitamin A deficiency
Night blindness was assessed by history and by examining children in a dimly lit room at least 12 feet in length with several items of furniture (World Health Organization, 1996). Mothers and children were questioned about night blindness with appropriate terms in the local language to obtain a history of night blindness. All children were subject to a clinical examination to detect ocular manifestations of vitamin A deficiency by trained medical officers according to the guidelines specified by the World Health Organization (1995a).
Socio-demographic and anthropometric data
An interviewer-administered questionnaire was used to collect information on socio-demographic and health-related data. Anthropometric measurements were obtained according to WHO guidelines (World Health Organization, 1995b). The height of each child was measured to a precision of 0·1 cm using a measuring scale. Weights were measured to a precision of 0·1 kg using a digital electronic scale (Seca, Les Mureaux, France). The scale was standardised every twenty-fifth to thirtieth measurement with standard weights of 500 g and 5 and 10 kg. Previous heights and weights of the children were obtained from past school medical inspection records. Anthropometric indices were calculated using EPIINFO (Centers for Disease Control and Prevention, Atlanta, GA, USA) and children were classified as stunted, underweight or wasted if their z-score of height-for-age, weight-for-age or weight-for-height was less than two standard deviations below the National Center for Health Statistics median. z-scores for weight-for-height were calculated for males up to 138 months (11.5 years) of age and less than 145 cm and for females up to 120 months (10 years) of age and less than 137 cm.
Collection of morbidity data
Morbidity data were collected by questioning the mothers on the past history of diarrhoea and respiratory infections in children during the 4 weeks prior to the interview. Diarrhoea was defined as the passage of three or more loose motions a day and a respiratory infection was defined as the presence of cough with or without expectoration for at least 24 h.
Determination of serum vitamin A concentrations
A 2 ml sample of venous blood was drawn from each child and transferred at 4°C to a laboratory in Colombo within 4–5 h where serum was separated and stored at − 80°C until biochemical analysis. Serum vitamin A concentrations were assayed by reversed-phase HPLC according to the method of Bieri et al. (Reference Bieri, Tolliver and Catignani1979). A total of 200 μl serum was added to 100 μl standard solution of retinyl acetate and 100 μl ethanol and serum vitamin A was extracted with 600 μl hexane. A 400 μl portion of the hexane extract was evaporated to dryness under a stream of nitrogen gas, re-dissolved in 100 μl mobile phase, and injected on to a C18, reversed phase 150 mm × 4·6 mm HPLC column (5 μm particle size; Waters, MA, USA). The mobile phase (methanol–water, 95:5 v/v) was delivered at a flow rate of 1 ml/min. Vitamin A (eluting time 4 min) was detected at 325 nm in a Beakman ultraviolet detector. The vitamin A concentration was quantified to the peak area of the internal retinyl acetate standard. The CV was 7·2 % and inter-batch CV was 5·8 %. Serum vitamin A values are reported in μmol/l.
Statistical analyses
Data analyses were carried out using SPSS (SPSS Inc., Chicago, IL, USA). t Tests, χ2 tests and regression analysis were used to detect associations.
Ethical aspects
Ethical clearance to conduct the study was obtained from the Ethical Review Committee of the Faculty of Medical Sciences, University of Sri Jayewardenepura. Permission to conduct the study was obtained from the Deputy Provincial Director of Health Services, Kegalle District, the Zonal Director of Education, Dehiowita, and all principals of schools selected for the study. Children were recruited into the study after obtaining written informed consent from their parents or guardians. Children detected with signs of vitamin A deficiency were given an oral megadose of vitamin A. Children with other medical conditions were referred appropriately.
Results
A total of 746 students from nine schools in the Yatiyantota MOH/DDHS area were recruited into the study. A total of 452 children from seven schools were supplemented with vitamin A. The rest (n 294) were not supplemented. The socio-demographic profile of the children is given in Table 1.
Table 1 Socio-demographic profile of children
GCE OL/AL, GCE ‘O’ level/‘A’ level.
The majority of the non-supplemented children were female, older than 9 years and lived in permanent houses. Mothers of supplemented children were more educated than mothers of non-supplemented children.
The prevalence of eye signs and symptoms among school children was extremely low and there was no difference between supplemented and non-supplemented children (Table 2). When the analyses were performed considering the sexes separately, there was no significant difference in nutritional status between supplemented and non-supplemented boys. Girls supplemented with vitamin A tended to be more malnourished than their non-supplemented counterparts (data not shown). Among non-supplemented children, only one case of night blindness was detected by history. Among supplemented children, five children had night blindness by dark adaptometry.
Table 2 Health profile of children
* Fisher's Exact Test.
† Children were classified as stunted, underweight or wasted if their z-score of height-for-age, weight-for-age or weight-for-height was less than two standard deviations below the National Center for Health Statistics median.
Although the nutritional status of supplemented children was poorer than that of the non-supplemented children, only the difference for height-for-age was statistically significant (Table 2). There was no significant difference in the reported incidence of infections between the two groups of children.
Of the supplemented children who were stunted at the last school medical examination 53 % were normal in the current survey (Table 3). Among the non-supplemented children classified as stunted in the last school medical examination, 23 % were classified as normal in the current survey.
Table 3 Prevalence of stunting* in the last school medical examination and current survey in relation to vitamin A supplementation status
* Children were classified as stunted if their z-score of height-for-age was less than two standard deviations below the National Center for Health Statistics median.
Serum vitamin A concentrations of supplemented children (1·4 (sd 0·49) μmol/l) were significantly higher than those of controls (1·2 (sd 0·52) μmol/l) (P = 0·003). Serum vitamin A concentrations declined gradually from the time of supplementation (Table 4; Fig. 1). Serum vitamin A concentrations of the children who were supplemented more than 6 months prior to the survey (1·3 (sd 0·44) μmol/l in those supplemented 7–12 months back and 1·1 (sd 0·44) μmol/l in those supplemented 13–18 months back) were similar to those of control children (1·2 (sd 0·52) μmol/l). Children who were supplemented within 1 month had the highest serum vitamin A concentrations (1·6 (sd 0·45) μmol/l) followed by children supplemented 1–6 months (1·4 (sd 0·50) μmol/l) previously.
Table 4 Serum vitamin A concentrations and prevalence of vitamin A deficiency by duration since supplementation*
a,b,c Mean values within a column with unlike superscript letters were significantly different (P < 0·05; Student–Newman–Keuls test).
* For details of procedures, see p. 154.
† For vitamin A deficiency, a serum vitamin A concentration of 0·7 μmol/l was taken as the cut-off value.
‡ F test from ANOVA.
§ χ2 value.
Fig. 1 Serum vitamin A levels by time since supplementation. For details of procedures, see p. 154. Values are means with their 95 % CI depicted by vertical bars.
In a multiple regression analysis, taking serum vitamin A concentrations as the dependent variable, time since supplementation ( < 1 month, 1–6 months, 7–12 months and >12 months) was the only significant predictor of serum vitamin A concentrations after controlling for other variables (Table 5). Children who were supplemented within 6 months of the survey had a higher serum vitamin A concentration than controls. The difference was as much as 0·4 μmol/l within 1 month of the survey and 0·2 μmol/l between 1 and 6 months.
Table 5 Regression analysis using serum vitamin A levels (μmol/l) as the dependent variable
* Regression coefficient.
† Reference group is the control group.
‡ F test.
§ t test.
∥ Reference group is female children.
¶ Reference group is education above Grade 9.
** Reference group is education above Grade 9.
†† Reference group is permanent house type.
‡‡ Reference group is children ≥ 5 years.
Discussion
The present study clearly shows that children supplemented with vitamin A within 1 month had a significantly higher concentration of vitamin A than those supplemented earlier. The ability to sustain a high serum vitamin A concentration following an oral megadose depends on many factors. The higher demand for nutrients during the period of growth, inadequate consumption of vitamin A and increased catabolism of vitamin A under a heavy burden of parasitic infections are critical to the vitamin A status of a child (Semba, Reference Semba1994). Some children were living on estates where a high prevalence of geo-helminth infections had been reported (Sorensen et al. Reference Sorensen, Ismail, Amarasinghe, Hettiarachchi and Dassenaike1996; Gunawardena et al. Reference Gunawardena, Karunaweera and Ismail2005). Supplemented children came from a poorer socioeconomic background and this may be a reason for their inability to sustain a high serum vitamin A level after 6 months of supplementation.
The single most important predictor of vitamin A levels was the time since supplementation, with vitamin A levels declining over time. Immediately after supplementation, vitamin A levels may increase by as much as 0·4 μmol/l, i.e. a 30–40 % increase in serum levels. By the end of 6 months this drops by about 0·2 μmol/l.
The dose of 105 μmol (100 000 IU) used in the supplementation programme in Sri Lanka may be a reason for the rapid decline in vitamin A levels. Different dosage and frequency forms have been used elsewhere. Most studies using a megadose of 210 μmol (200 000 IU) for supplementation every 3–4 or 6 months report beneficial effects despite co-existing malnutrition and heavy infectious disease burdens (Sommer et al. Reference Sommer, Tarwotjo, Djunaedi, West, Loeden and Tilden1986; West et al. Reference West, Pokhrel and Katz1991; Daulaire et al. Reference Daulaire, Starbuck, Houston, Church, Stukel and Pandey1992). In Africa, the Ghana Vast Study Team (1993) trial reported the beneficial effects of vitamin A supplementation on morbidity and mortality on children over 1 year with children receiving 210 μmol every 4 months. Herera et al. (Reference Herera, Nestel, El Amin, Fawzi, Mohamed and Weld1992) reported that a 6-monthly dosing schedule of vitamin A supplementation had no impact on child survival in Sudan. Similarly, Pedro et al. (Reference Pedro, Madriaga, Barba, Habito, Gana, Deitchler and Mason2004) reported that the effect of high-dose vitamin A capsules on serum retinol concentrations do not persist for 6 months in children from 1 to 5 years of age in the Philippines.
The success of a vitamin A supplementation programme using an oral megadose relies on the body's ability to store vitamin A. The bulk of the body's vitamin A is stored as retinyl esters in the liver. To meet constant tissue needs, despite day-to-day variability of dietary vitamin A, a steady concentration of circulating retinol is maintained by drawing on hepatic reserves through secretion of retinol bound to its specific carrier protein, retinol-binding protein (Gamble et al. Reference Gamble, Rajasekhar, Palafox, Briand, Berglund and Blaner2001). Protein deficiency status reduces the absorption of vitamin A, and the synthesis and release of retinol-binding protein from the liver. In such situations, an oral megadose supplementation programme will be ineffective.
Donnen et al. (Reference Donnen, Dramaix, Brasseur, Bitwe, Vertongen and Hennart1998) reported that high-dose supplementation of vitamin A did not reduce morbidity in malnourished and vitamin A-deficient children, and recommended a daily low-dose vitamin A supplementation for severely malnourished children. There is evidence that weekly low-dose supplementation is more effective than periodic megadose supplementation (Rahmathullah et al. Reference Rahmathullah, Underwood, Thulasiraj, Milton, Ramaswamy, Rahmathullah and Babu1990). More frequent supplementation schedules have inherent practical problems associated with their implementation.
Baseline vitamin A status may influence the impact of a supplementation programme. It has been reported that serum concentrations will only increase if they are already low and low concentrations are due to vitamin A deficiency (Olson, Reference Olson1984). We could not elucidate baseline values as the survey was a cross-sectional survey. Considering average vitamin A concentrations of children which ranged from 1·4 (sd 0·49) to 1·2 (sd 0·52) μmol/l for supplemented and control groups, respectively, there is very little evidence to suggest a high prevalence of vitamin A deficiency or extreme deficiency among these children. The average serum vitamin A concentrations in children of the present study were similar to those reported in the UK (Thurnham et al. Reference Thurnham, Mburu, Mwaniki and Wagt2005) and USA (Ballew et al. Reference Ballew, Bowman, Sowell and Gillespie2001), though measured using different assays. It is possible that the impact of a single mega dose supplementation of 105 μmol vitamin A observed in the present study may be due to the relatively low prevalence of vitamin A deficiency in this population.
Serum vitamin A levels and the prevalence of ocular manifestations of deficiency are important outcome indicators in evaluating vitamin A intervention programmes (Wasantwisut, Reference Wasantwisut2002). Though vitamin A concentrations were significantly different between non-supplemented and supplemented children 6 months before the survey, there was no significant difference in the prevalence of ocular manifestations of vitamin A deficiency or morbidity between the two groups of children. It may be concluded that a single dose of 105 μmol vitamin A improved serum vitamin A concentrations significantly in the short term but the effect on morbidity was not clear.
Night blindness and conjunctival xerosis are often associated with poor nutritional status, particularly stunting (Brink et al. Reference Brink, Perera, Broske, Cash, Smith, Sauberlich and Bashor1979; Santos et al. Reference Santos, Dricot, Asciutti and Dricot-d'ans1983). There are reports that vitamin A supplementation improves linear growth (Arroyave et al. Reference Arroyave, Aguilar, Flores and Gunzman1979; Muhilal et al. Reference Muhilal, Idjradinata, Muherdiyantiningsih and Karyadi1988; Hadi et al. Reference Hadi, Stoltzfus, Dibley, Moulton, West, Kjolhede and Sadjimin2000). Even though the single administration of a 105 μmol megadose may be inadequate to sustain high levels after 6 months, there is evidence suggestive that it may have a beneficial effect on nutritional status of children as documented by the reduction in the prevalence of stunting in supplemented children. However, the present results need to be interpreted with caution given the baseline differences, and age and sex differences between the two groups.
A typical child in a developing country would need to increase the portion size of fruits and vegetables by about 10-fold to control vitamin A deficiency by eating fruits and vegetables alone, a goal which cannot be achieved (Miller et al. Reference Miller, Humphrey, Johnson, Marinda, Brookmeyer and Kartz2002). Therefore, without supplementation, a child in a developing country is not able to attain and maintain ‘minimally adequate’ vitamin A stores in the liver. The WHO recommends a dose of 210 μmol vitamin A every 4–6 months for children above 1 year of age (World Health Organization, 2000). Hence, the dose of 105 μmol every 3 years used in Sri Lanka for school children is grossly inadequate.
The problem of overdosing may be the single most important reason for using the current dosing schedule. Only about 1 % of children showed signs of intolerance to 210 μmol retinyl palmitate in acid solution, which disappeared after a few hours (Arroyave, Reference Arroyave1988). Sommer (Reference Sommer1996) did not report any deaths linked to isolated vitamin A toxicity. Studies conducted among preschool children receiving 210 μmol every 4–6 months reported no overdosing signs and symptoms (Sommer et al. Reference Sommer, Tarwotjo, Djunaedi, West, Loeden and Tilden1986; West et al. Reference West, Pokhrel and Katz1991; Daulaire et al. Reference Daulaire, Starbuck, Houston, Church, Stukel and Pandey1992).
In Sri Lanka, vitamin A deficiency has been a long-felt public health problem and there was no sustained programme prior to the current supplementation programme. Although supplementation with high-dose vitamin A is the most widely used strategy to combat vitamin A deficiency, other strategies have been adopted or are being considered in other countries. These include food fortification, homestead food production and social marketing (Bloem et al. Reference Bloem, Kiess and Moench-Pfannner2002). Revising the dosing schedule coupled with other interventions may produce better results.
Acknowledgements
We thank the staff of the MOH/DDHS Office, Yatiyantota, Kegalle for their assistance in data collection, Dr Gayathri Gauthamadasa and Dr S. M. Dunuwila for assisting in conducting clinical examinations, and Dr Renu Wickremasinghe and the staff of the Departments of Parasitology and Biochemistry of the Faculty of Medical Sciences, University of Sri Jayewardenepura, Sri Lanka for their assistance in analysing serum samples.
