点击显示 收起
1 From the Center for Human Nutrition, Department of International Health, The Johns Hopkins School of Hygiene and Public Health, Baltimore, and the Nepal Society for the Prevention of Blindness, Kathmandu, Nepal.
2 Supported by Career Development Awards from the National Eye Institute and Research to Prevent Blindness (K-23 EY 00388-02) (NGC). Also supported by Cooperative Agreement HRN-A-00-97-00015-00 between the Center for Human Nutrition, Johns Hopkins School of Public Health, and the Office of Health and Nutrition, US Agency for International Development, with additional support from the Sight and Life Institute, Johns Hopkins School of Public Health, and the Bill and Melinda Gates Foundation, in collaboration with Nepal Netra Jyoti Sangh, Kathmandu, and the Sushi Kedia Seva Mandir, Hariaun, Nepal. 3 Address reprint requests to NG Congdon, Wilmer 120, 600 North Wolfe Street, Baltimore, MD 21287–5001. E-mail: ncongdon{at}jhmi.edu.
ABSTRACT
Background: Impaired dark adaptation occurs commonly in vitamin A deficiency.
Objective: We sought to examine the responsiveness of dark-adaptation threshold to vitamin A and ß-carotene supplementation in Nepali women.
Design: The dark-adapted pupillary response was tested in 298 pregnant women aged 15–45 y in a placebo-controlled trial of vitamin A and ß-carotene; 131 of these women were also tested at 3 mo postpartum. Results were compared with those for 100 nonpregnant US women of similar age. The amount of light required for pupillary constriction was recorded after bleaching and dark adaptation.
Results: Pregnant women receiving vitamin A had better dark-adaptation thresholds (-1.24 log cd/m2) than did those receiving placebo (-1.11 log cd/m2; P = 0.03) or ß-carotene (-1.13 log cd/m2; P = 0.05) (t tests with Bonferroni correction). Dark-adaptation threshold was associated with serum retinol concentration in pregnant women receiving placebo (P = 0.001) and in those receiving ß-carotene (P = 0.003) but not in those receiving vitamin A. Among women receiving placebo, mean dark-adaptation thresholds were better during the first trimester (-1.23 log cd/m2) than during the second and third trimesters (-1.03 log cd/m2; P = 0.02, t test). The mean threshold of nonpregnant US women (-1.35 log cd/m2) was better than that of all 3 Nepali groups (P < 0.001, t test, for all 3 groups).
Conclusions: During pregnancy, pupillary dark adaptation was strongly associated with serum retinol concentration and improved significantly in response to vitamin A supplementation. This noninvasive testing technique is a valid indicator of population vitamin A status in women of reproductive age.
Key Words: Vitamin A • ß-carotene • night blindness • pregnancy • lactation • pupil • dark adaptation • dark vision • night vision • Nepal • pregnant women • postpartum women
INTRODUCTION
In previous reports, we described the use of a rapid, portable device to assess the dark-adaptation threshold as an index of population vitamin A status among preschool children (1, 2), the group most at risk of mortality and morbidity associated with vitamin A deficiency (3). This noninvasive technique, which relies on observation of the pupil, is more acceptable to children than is conventional psychophysical dark-adaptation testing. Measurements of the dark-adaptation threshold with this technique correlate with serum retinol concentrations and are responsive to vitamin A supplementation (1, 2).
Night blindness, a marker of moderate vitamin A deficiency, is also prevalent among pregnant and lactating women in Nepal, affecting 16% and 6% of these groups, respectively (4). In pregnancy, night blindness is associated with low serum retinol concentrations (<0.7 µmol/L) and chronic low intake of vitamin A–rich foods (5). Vitamin A supplementation was shown to reduce the incidence of night blindness during pregnancy by nearly 70% (6). Furthermore, low-dose vitamin A and ß-carotene supplementation in women of reproductive age reduced pregnancy-related mortality by 44% in the rural plains region of Nepal (7). Because of these findings, there is growing interest in rapid, noninvasive techniques for assessing milder stages of vitamin A deficiency before the occurrence of night blindness in pregnant and lactating women in populations at risk of vitamin A deficiency.
We sought to determine whether dark-adaptation testing with our protocol, which was initially developed for children, would be acceptable to women of reproductive age living in the rural plains of Nepal. This is a population group known to be at risk of clinical vitamin A deficiency (4). Our original protocol called for both pupillary and conventional visual dark-adaptation testing on all subjects. However, previous results suggested that visual dark adaptation provided little, if any, additional information beyond pupillary testing (2). Only the results of pupillary testing are reported here. We also sought to examine the responsiveness of dark-adaptation thresholds to vitamin A and ß-carotene supplementation in pregnant and lactating women and to evaluate the relations between dark-adaptation thresholds and serum retinol concentrations, gestational age, and other indicators of general nutritional status.
SUBJECTS AND METHODS
Study population
The study was conducted in Sarlahi District in the southern, rural plains (terai) of Nepal, bordering India. Protein-energy and micronutrient deficiencies are prevalent among adults and children in this region (8). The study population was composed of women aged 15–45 y who were participating in a randomized community intervention trial of the effects of low-dose vitamin A or ß-carotene supplementation on maternal, fetal, and early infant mortality and morbidity (7).
Married women living in 30 village development communities, each consisting of 9 wards (local administrative units), were randomly assigned at the ward level (270 wards), blocked on village development community. Thus, an equal number of wards received each of the 3 treatments, which were weekly supplements containing 7000 µg retinol equivalent (RE) preformed vitamin A (retinyl palmitate), 7000 µg RE all-trans-ß-carotene, or an identical-appearing placebo in peanut oil. All supplements contained 5 mg -tocopherol. During weekly household visits, the women were given their assigned supplements. When a woman stated that she was pregnant, she was enrolled in a more intensive protocol involving multiple interviews during pregnancy and the postpartum period. Administration of the supplements continued throughout the postpartum period and any subsequent pregnancies.
Three village development communities (27 wards) were selected because they were the most accessible by road to participate in a substudy involving nutritional and clinical assessment of women during pregnancy and at 3 mo postpartum. All women in these wards who reported that they were pregnant during weekly home interviews were invited to come to a project clinic to have a venous blood sample drawn for measurement of nutritional biochemical indexes, to have a stool examination for intestinal helminths, and to permit collection of dietary and anthropometric data. One-half of the pregnant women who visited the clinic between July 1995 and July 1997 were randomly selected to participate in dark-adaptation testing. Of the 345 women selected to participate, 324 (93.9%) had their pregnancies confirmed by urine testing for B-chorionic gonadotropin. Of the 324 individuals, 4 visited the clinic for 2 different pregnancies each, and only data from the first pregnancy were analyzed. Of the remaining 320 eligible subjects, 22 (6.9%) either refused or were unable to complete dark-adaptation testing. This resulted in a total of 298 pregnant women with results for dark-adaptation testing (Figure 1).
FIGURE 1. . Flow chart showing enrollment and dark-adaptation testing of Nepali women in the study.
Of the women who participated in dark-adaptation testing during pregnancy, 172 gave birth to a live infant and returned to the clinic at 3 mo postpartum for another health examination and follow-up dark-adaptation testing. Of these women, one had a positive B-chorionic gonadotropin test and was found to be pregnant again. Of the remaining 171 subjects, 40 (23.4%) visited the clinic at a time when the machine was undergoing repair and was unavailable for testing. The remaining 131 postpartum women completed dark-adaptation testing successfully (Figure 1).
A total of 100 apparently healthy, nonpregnant women aged 15–45 y underwent dark-adaptation testing at a central site in Baltimore. The testing protocol was the same as was used with the Nepali women. Subjects of European (n = 43), Asian (n = 27), and African (n = 30) descent were recruited from among the students and faculty at the Johns Hopkins School of Hygiene and Public Health. Of these women, 100% completed dark-adaptation testing successfully. The protocol of this study was approved by the Committee on Human Research of the Johns Hopkins School of Hygiene and Public Health and the Joint Committee for Clinical Investigation of the School of Medicine.
Protocol for dark-adaptation threshold testing
The protocol was described elsewhere in detail (1). Briefly, scotopic sensitivity was assessed by using a rechargeable, battery-powered, hand-held illuminator with a yellow-green light-emitting diode (LED) light source that had 11 intensity settings at 0.4 log unit intervals. The instrument was designed to fit entirely over the subject's eye while the contralateral eye was illuminated by an obliquely mounted red LED light source.
Before testing, subjects underwent a binocular partial bleach (ie, exposure to bright light to counteract previous partial dark adaptation) and then were allowed to dark adapt for 10 min in a room darkened to standard specifications. These specifications were such that a dark-adapted observer should be unable to read hand-sized black letters a finger's breadth wide printed on white paper.
The pupillary threshold was measured with the machine over the subject's left eye while the right eye was observed under oblique red illumination with a loupe of 2.5x magnification. The subject's attention was directed to a nonaccommodative target at a distance of 2 m (ie, a target located at a sufficient distance to prevent accommodation or close-up focusing of the eye, which causes the pupil to constrict). Then the stimulus intensity was increased until a pupillary response in the contralateral (right) eye was clearly visible to the observer on 2 successive trials.
All testing in Nepal was carried out by 3 local fieldworkers who had a 10th-grade education. These examiners performed the testing after a 2-wk period of training and standardization directed by one of the authors (NGC). Validation tests of both inter- and intraobserver reliability were carried out and examiners were not certified until their scores came within one grade of the trainer for 10 successive subjects. A 1-wk period of retraining was provided by the original trainer (NGC) roughly midway through the study. In the United States, all testing was carried out by a single examiner (DH) after a similar period of training and standardization provided by the same trainer.
Assessment of nutritional and health status
During a home interview, each subject's age, date of last menstrual period (LMP), and pregnancy history were collected for enrollment of pregnant women into the supplementation trial. The date of the LMP used for calculation of gestational age at the clinic visit was based on a combination of prospectively reported menstrual histories and recall of the LMP.
Assessment of nutritional and health status by various clinical, anthropometric, and biochemical indicators was carried out at the clinic for women enrolled in the clinic substudy. Midupper arm circumference (MUAC) was measured to the nearest mm at the midpoint of the left arm with an insertion tape (9). The median of 3 measurements was recorded.
Venous blood was collected from the subjects and was centrifuged (1530 x g for 10 min at room temperature). The serum was collected in 1-mL cryotubes and stored immediately in liquid nitrogen until arrival at the Johns Hopkins Nutrition Laboratory in Baltimore, where samples were stored at -70°C until analyzed. Serum retinol concentrations were measured by reversed-phase isocratic HPLC (10).
Statistical methods
All analyses were performed with SAS version 6.12 (SAS Institute Inc, Cary, NC). Ranges and frequency distributions of all continuous and categorical variables were examined. The t test was used to compare differences in mean age, gestational age, MUAC, serum retinol concentration, and dark-adaptation threshold between the placebo, ß-carotene–supplemented, and vitamin A–supplemented groups. Because it was not expected that treatment with vitamin A would worsen dark-adaptation threshold or lower serum retinol concentrations, a one-tailed t test was used for these comparisons, whereas a two-tailed test was used for all other comparisons. A Bonferroni correction for multiple comparisons was used for all comparisons among the 3 groups. Linear regression analysis was performed to examine the relation between serum retinol concentration and dark-adaptation threshold. Repeated-measures analysis of variance was performed to examine the between-subject effects (such as treatment), within-subject effects (such as time), and interactions between the 2 types of effects. A P value of 0.05 was considered significant for all tests.
Dark-adaptation threshold was measured in log cd/m2, the SI unit of luminance. Subjects with better dark vision could see a dimmer light, meaning a smaller luminance value. On a logarithmic scale, these were expressed as more negative numbers. For example, a dark-adaptation threshold of -2.00 log cd/m2 would be a full order of magnitude (3 units on our machine) more sensitive than a threshold of -1.00 log cd/m2.
RESULTS
Among the pregnant subjects who underwent dark-adaptation testing, there were no significant differences by group (vitamin A, ß-carotene, or placebo) in terms of age, MUAC, or gestational age at the time of testing (Table 1). Women receiving vitamin A had higher serum retinol concentrations than did women receiving ß-carotene or placebo (Table 2). Pregnant women receiving vitamin A had lower dark-adaptation thresholds (ie, better dark vision) than did those receiving ß-carotene or placebo (Table 2).
View this table:
TABLE 1.. Comparability between treatment groups for Nepali women during pregnancy and postpartum1
View this table:
TABLE 2.. Serum retinol concentration and dark-adaptation threshold for Nepali pregnant and postpartum women and nonpregnant women in the United States1
The mean dark-adaptation threshold of apparently healthy, nonpregnant women tested in the United States was lower (better) than that of all 3 groups of women tested in Nepal (Table 2). There was no significant difference between mean dark-adaptation thresholds for the 3 main racial groups tested in the United States, although the power to have detected a difference between racial groups in the United States which was as large as that seen between US women and untreated Nepali women was <80% (data not shown).
Among pregnant women receiving the placebo, there was a strong association between dark-adaptation threshold and serum retinol concentration; women with higher serum retinol concentrations had lower (better) dark-adaptation thresholds (regression coefficient = -0.014, P = 0.0005). This association was also observed among women receiving ß-carotene (regression coefficient = -0.009, P = 0.005) but not among women receiving vitamin A (regression coefficient = -0.003, P = 0.39) (Figure 2). The slopes of the regression lines for women receiving vitamin A and placebo differed significantly (P = 0.04, t test for regression interaction term), whereas those of the groups receiving vitamin A and ß-carotene did not (P = 0.25, t test for regression interaction term).
FIGURE 2. . Dark-adaptation threshold and serum retinol concentration of pregnant Nepali women receiving ß-carotene, vitamin A, and placebo. Dark-adaptation threshold and serum retinol concentration were inversely correlated in the placebo (regression coefficient = -0.014, P = 0.0005) and ß-carotene (regression coefficient = -0.009, P = 0.005) groups.
In the placebo group, the mean dark-adaptation threshold of women tested in the second and third trimesters (-1.03 log cd/m2; n = 51) was significantly worse than that of women tested in the first trimester (-1.23 log cd/m2; n = 36) (P = 0.02). Dark-adaptation threshold did not differ significantly by gestational age among women receiving ß-carotene or vitamin A. Note that each subject was tested only once during pregnancy.
There was no significant association between dark-adaptation threshold and wasting malnutrition, reflected by MUAC, in any of the 3 groups of women during pregnancy or the postpartum period.
The serum retinol concentrations of women who returned to the clinic at 3 mo postpartum differed significantly by group assignment. Mean serum retinol of women receiving vitamin A was higher than that of women receiving placebo (1.53 ± 0.41 compared with 1.15 ± 0.57 µmol/L, P = 0.003). There was no significant difference among mean dark-adaptation thresholds of the 3 groups at 3 mo postpartum (Table 2).
Two-factor repeated measures analysis of variance was performed for matched pregnant and postpartum women who received either vitamin A or placebo during pregnancy, with time and group assignment as the 2 factors. Each woman was matched to herself, in that her dark-adaptation threshold in pregnancy was matched to her postpartum value for the analysis. Time, group, and the interaction were all not significant with regard to dark-adaptation threshold (P > 0.3 for all factors). The power to detect a difference in thresholds between groups within the subset of women who returned for postpartum follow-up was as high as the power for comparisons between pregnant women receiving vitamin A and pregnant women receiving placebo (power = 0.83). With regard to serum retinol concentration, the group effect was of borderline significance for pregnant women (P = 0.05) and was significant for postpartum women (P = 0.004). The time effect was significant (P = 0.0004).
DISCUSSION
Dark-adaptation testing with our pupillary protocol was acceptable to most of the women; >90% of women selected to undergo dark-adaptation testing completed the full protocol. Dark adaptation showed the expected response to vitamin A supplementation in pregnancy: women assigned to the vitamin A group had dark-adaptation thresholds that were significantly lower (indicating better dark adaptation) than those of women in the placebo and ß-carotene groups. Serum retinol concentrations of women supplemented with vitamin A were significantly better than those of women who received placebo and ß-carotene, both in pregnancy and postpartum.
In the placebo group, which had the poorest vitamin A status and received no intervention, dark-adaptation thresholds were significantly associated with serum retinol concentrations. This relation disappeared altogether in women supplemented with vitamin A (Figure 2). This suggests that weekly supplementation with 7000 µg RE vitamin A was sufficient in this group of pregnant women to eliminate the association between dark-adaptation threshold and serum retinol concentration on a population basis. This does not suggest that there were no remaining women with poor dark vision in the supplemented group, but does provide a physiologic basis for a normal cutoff for pupillary dark-adaptation threshold in this population (-1.24 log cd/m2 for the group receiving vitamin A). It should be stressed that such a cutoff refers to the status of populations rather than of individuals; the standard could presumably be of value in assessing the vitamin A status of comparable populations elsewhere.
It is interesting to note that dark-adaptation thresholds for even the supplemented Nepali women were significantly worse than thresholds for apparently healthy, nonpregnant women tested in the United States. It is possible that this represents an effect of pregnancy that cannot be improved with vitamin A supplementation, although this is not likely given that there was no significant improvement in dark-adaptation threshold among these Nepali women after delivery. Another possibility is that poorer dark adaptation in Nepali women was due to some other nutritional factor particular to Nepal, such as low zinc concentrations. This would be consistent with studies that found that vitamin A supplementation reduced, but did not eliminate, self-reported maternal night blindness in Nepal (6). Nepali women are generally considerably thinner than Americans, suggesting that protein-energy malnutrition could be responsible; however, the lack of association between the results of dark-adaptation testing and MUAC in Nepal argues against this explanation. A genetic difference between the Nepali women and those tested in the United States seems less likely, because no significant differences were found between the different racial groups (Asian, African, and white) tested in the United States. However, as noted above, the power to find such differences was limited.
Among women who received the placebo, dark-adaptation thresholds were significantly worse in the second and third trimesters than in the first trimester. This finding agrees with clinical observations that compromised dark adaptation is more common later in pregnancy (3) and with biochemical evidence of more severely depleted vitamin A status during this same period (11). It is interesting to note that among women supplemented with ß-carotene or vitamin A, dark-adaptation thresholds did not differ by gestational age at the time of testing.
The association between dark-adaptation threshold and measures of vitamin A status in this population was specific to pregnancy. Although serum retinol concentrations differed significantly between the 3 groups at 3 mo postpartum, dark-adaptation threshold did not differ significantly by group assignment in the postpartum period. It may be that dark-adaptation threshold was more strongly associated with serum retinol concentration during pregnancy because of high nutritional demands and lower systemic reserves at this time. However, there was no significant difference between the pregnancy and postpartum threshold means in the 2-factor analysis for women tested during both pregnancy and the postpartum period. This may have been because some of the women tested in pregnancy were assessed in early pregnancy, when dark-adaptation threshold had not yet been severely affected. Also, our power to detect such differences was only 80% because only a subset of pregnant women returned for postpartum follow-up.
In view of the fact that postpartum follow-up rates were <50% for all 3 groups (Figure 1), we must consider the possibility that differential follow-up between groups could explain the lack of a postpartum association between serum retinol concentration and dark-adaptation threshold. In fact, for the vitamin A group, mean dark-adaptation thresholds during pregnancy for the women who later failed to return postpartum were significantly better than those for the women who did return postpartum (-1.19 ± 0.41 compared with -1.12 ± 0.34 log cd/m2, respectively, P = 0.02). This would tend to mask an association between dark-adaptation threshold and group assignment in the postpartum period.
The lack of association during pregnancy or the postpartum period between dark-adaptation threshold and MUAC, an indicator of wasting, contrasts with a rather strong association between wasting and maternal night blindness reported in this same population (5); this finding presumably reflected a more nutritionally compromised state among women with night blindness. Our findings suggest that dark-adaptation testing results can be interpreted to reflect vitamin A status across a range of nutritional status levels, in that the thresholds appear to be relatively unaffected by general nutritional state.
It should be stressed that pupillary dark-adaptation testing is recommended as a method for assessing vitamin A status in populations. We do not suggest using cutoffs to attempt to distinguish between normal and abnormal individuals. For the purpose of planning public health programs, there is little value in being able to distinguish between deficient and nondeficient individuals; once a population has been identified as being at risk and is targeted for prophylactic intervention, all individuals will receive the treatment.
Measuring the dark-adaptation threshold with our pupillary protocol offers several advantages over commonly used indicators of vitamin A status. The testing is noninvasive and is well tolerated, unlike phlebotomy to measure biochemical indicators such as serum retinol concentration and relative-dose-response measurements (12). Our approach avoids the cumbersome and expensive transport of delicate specimens that is necessary with both serum testing and conjunctival impression cytology. There are no long delays necessitated by the off-site processing of specimens and feedback on the status of a population or the response to an intervention is obtained immediately. Examination of the pupillary response by an observer allows measurement of the dark-adaptation threshold in individuals, such as young children and less educated subjects, who might be unable to comply with traditional psychophysical dark-adaptation testing. The technique lends itself to large-scale national programs for the assessment and prevention of vitamin A deficiency, and to measuring the effects of intervention programs, in which low cost, portability, wide acceptability, and rapidity are of paramount importance.
ACKNOWLEDGMENTS
We acknowledge the assistance of Alfred Sommer, Steven LeClerq, Elizabeth Kimbrough-Pradhan, Sharada Ram Shrestha, Andre Hackman, Gwendolyn Clements, Ramesh K Adhikuri, and Sanu M Dali.
REFERENCES