Introduction

Para-aminobenzoic acid (PABA) has long been an accepted objective marker to verify completeness of 24 h urine sampling (Bingham & Cummings, 1983). PABA-verification is inter alia used to validate the accuracy of dietary intake studies (Bingham et al, 1995), through the whole adult age range (Black et al, 1997), often by calculating the association between dietary protein intake and urinary protein (nitrogen) excretion (Bingham & Cummings, 1985). The underlying assumption is that PABA is excreted almost quantitatively in 24 h.

Normally, three 80 mg PABA tablets are taken orally in conjunction with main meals on the day of urine collection, and PABA recovery in the urine above 85% of total ingested dose indicates urine has been collected for 24 h. There is some evidence that PABA excretion is lower in elderly subjects compared to the younger age groups (Leclercq et al, 1991). An earlier study by us introduced a highly specific HPLC method for PABA determination in urine and, further, suggested that advancing the last PABA dosage 3 h, from 18:00 to 15:00 h in subjects aged 80 y might compensate for a lower PABA recovery in this age group (Jakobsen et al, 1997).

Studies measuring PABA in urine specimens collected separately during 24 h after single or repeated doses in healthy subjects have rarely been performed (Bingham & Cummings, 1983), and not at all with the HPLC method. Although PABA verification has proved to be a useful marker, 24 h urine could be incorrectly excluded (false-negative) or included (false-positive) depending on intake schedule or variability in pharmacokinetics of PABA.

The primary aim of this study is to analyse the age dependency of the urinary PABA excretion, and to examine whether a delayed PABA excretion can be overcome by advancing the intake schedule; and further to validate the previously assessed delimiting level of PABA recovery for completeness of 24 h urine. Lastly, it is the aim to examine the excretion of PABA in fractionated urinary samples collected during 24 h after single and repeated doses of oral PABA.

Materials and methods

Subjects and urine collections

A total of 99 subjects (61 females, 38 males) spanning the ages 30–91 y (average 55.9) with a body mass index (BMI) between 14.5 and 36.0 kg/m² (average 24.2) took part in the study. All subjects gave informed consent. The study protocol was approved by the regional ethical committee (no. KA 97049g).

All were well-functioning, ambulatory and in good health without acute or severe chronic illness. They were employees, or relatives or friends of employees at the Food Administration of Denmark. No dietary changes were imposed and all followed their normal daily routine. For the collection of urine all subjects were given 21 brown bottles containing 10 ml/HCl (1 mol/l) for preservation, a smaller 'visiting bottle', a funnel, a safety pin for attachment to a suitable place as a reminder, three 80 mg PABAcheck tablets (Laboratories for Applied Biology, London, UK) and a set of printed instructions. Subjects were asked to record the time of start and finish of the urine collections, together with the time of taking the tablets and any lost specimen. Urine was received at the laboratory immediately after collection, volumes were estimated by weight (specific gravity=1 g/ml), and aliquots were stored at -20°C for a maximum duration of 1 month.

From each of the participating subjects 24 h urine was sampled twice at least 1 week apart. On collection days the subjects ingested 240 mg of PABA, divided into three doses of 80 mg. On both collection days subjects were carefully instructed to take the first two doses of PABA at 8:00 and 12:00 h, and the last dose at 15:00 h (PABA15) or 18:00 h (PABA18). Random numbers determined the order of collection for each subject. Information of the time of the start and the finish of the urine collection was reported as well as the exact hour for the intake of the third PABA tablet. Twenty-four hour urine was analysed for PABA without knowing whether the sample was PABA15 or PABA18.

Eight subjects (six females, two males) aged 30–70 y (average 53.3) with a BMI between 19.6 and 30.1 kg/m² (average 24.7) took one 80 mg tablet of PABA with breakfast after urination. Urinations were then collected in separate containers for 24 h. The times of urinations were meticulously reported. The subjects had their habitual diet and an ad libitum fluid intake. Individual urine specimens were analysed separately for PABA.

Ten subjects (eight females, two males) aged 32–55 y (average 43.2) with a BMI between 18.7 and 25.5 kg/m² (average 25.5) volunteered to collect separate urinations for 24 h. Each subject ingested 80 mg PABA three times (total 240 mg) precisely at 08:00, 12:00 and 18:00. Subjects were asked to void approximately every 3 h until bedtime. The times of urination were meticulously reported. The first urination was done precisely at 08:00 on the day of PABA collection and the last urination precisely at 08:00 the following morning. No dietary changes were made and normal amounts of fluid were drunk. To secure PABA intake at the scheduled hours subjects were equipped with a digital alarm clock. Individual urine specimens, including the urine voided at 08:00 before intake of PABA, were analysed separately for PABA.

Analytical method

For the determination of PABA the HPLC method was used. The method, which is described in detail by Jakobsen et al (1997), includes alkaline hydrolysis for conversion of PABA metabolites to PABA, followed by separation on a reversed-phase column (VYDAC 201 TP 54), detection at 290 nm and quantification by an external standard method. The HPLC system (Waters, Milford, MA, USA) was equipped with controller 600E with pump, autoinjector 717, UV detector model 2487 and software Millennium 2010, version 2.1.

Statistical analysis

For the statistical test of the accepted data set the InStat (GraphPad Software, Inc., CA, USA) was used. Linear regressions with test runs were performed and P-values <0.05 were considered statistically significant. Values are given as mean (x)s.d. or 95% confidence intervals (CI).

Results

A total of 26 subjects were excluded because of (1) lack of information on time of ingestion of third PABA dosage (four subjects); (2) PABA recovery less than 160 mg indicating grossly incomplete urine sampling or PABA ingestion (six subjects); (3) a difference of less than 2 h between times of taking the third PABA-dosage (nine subjects); and (4) a collection period less than 12 h and longer than 15 h after the third PABA dosage for PABA18 and less than 15 h or more than 18 h for PABA15 (seven subjects). The 73 included subjects did not differ from the original data material with respect to age and BMI, however most of the excluded subjects came from the middle age groups (Table 1). Data on urine volumes and durations of urine collections are given in Table 2. About half of the included subjects took a wide range of medications, including herbal medicine and vitamin–mineral supplements.

Table 1 - Information about the subjects.

Full table (9K)

Table 2 - Volume and duration of urine collections from included subjects (n=73).

Full table (11K)

Figure 1 shows a scattergram of the recovery of 24 h PABA. Linear regression for PABA15 as well as PABA18 demonstrates significantly less recovery with age of the subjects (PABA15: r²=0.1784; slope deviates significantly from zero with P=0.0002; test runs: NS; PABA18: r²=0.1273; slope deviates significantly from zero with P=0.0019; test runs: NS). The finding of almost identical slopes of the two regression lines (PABA15: -0.1403 (95% CI -0.2117, -0.0690); PABA18: -0.1330 (95% CI -0.2154, -0.0505) indicates against an increased recovery with PABA15 compared to PABA18 with increasing age of the subjects. This is further substantiated when differences in recovery between PABA15 and PABA18 (PABA=PABA15-PABA18) vs age are plotted (Figure 2). Linear regression shows an r² of close to 0 (r²=0.0003), and the best-fit line to be horizontal (slope: -0.0066, P=0.89; 95% CI -0.1046, 0.0915; test runs: NS and with a y-intercept not significantly different from 0 (1.575; 95% CI -4.176, 7.326).

Figure 1.

Linear regressions of recoveries of PABA15 (last dosage taken at 15:00) and PABA18 (last dosage taken at 18:00) with age in 73 healthy subjects.

Full figure and legend (17K)

Figure 2.

Linear regression of difference between PABA15 (last dosage taken at 15:00) and PABA18 (last dosage taken at 18:00) with age in 73 healthy subjects.

Full figure and legend (11K)

To determine the lower limit for a complete 24 h urine using the HPLC method, those subjects who did not meet the demands of a certain time interval and collection periods for the two dosage schedules were included (n=16). Further, recoveries of PABA15 and PABA18 were combined for each subject (within-subject s.d.: 4.2%). Mean 24 h PABA recovery for all subjects (n=89) was 87.5% (range 68.5–106.4%) and the lower limit for complete 24 h urine collection comprising 95% of the subjects was 75.2% (183 mg).

Recovery of PABA in individual urine specimens

The excretion of PABA in individual urine specimens in eight subjects during 24 h after ingestion of one single dosage of 80 mg of PABA at 08:00 is shown in Figure 3. After 4 h about 60–80%, after 6 h 70–85% and after 8 h 75–90% had been recovered.

Figure 3.

Recovered PABA in urine (%) after intake of a single dosage of 80 mg at 08:00.

Full figure and legend (25K)

Twenty-four-hour recoveries in 10 subjects after ingestion of 80 mg PABA at 08:00, 12:00 and 18:00 is shown in Figure 4. These 10 subjects collected urine at 08:00 before intake of PABA. Analyses of these urines showed interferences for PABA at a level of 1.2–1.2 g/ml, corresponding to a total of 0.2–3.7 mg PABA in 24 h urine. After 16 h (24:00) with this dosage schedule the lower limit for a complete 24 h urine collection (78.1%) was exceeded in all subjects. A total of 21 (25%) individual urine specimens were collected after this limit was reached.

Figure 4.

Recovered PABA in urine (%) after intake of 80 mg PABA at 08:00, 12:00 and 18:00.

Full figure and legend (29K)

Discussion and conclusion

This study shows that mean PABA excretion decreases with the age of the subject. This finding confirms the result of our previous study (Jakobsen et al, 1997). However, in contrast to the previous cross-sectional study, we could not in this randomized cross-over study 'normalize' PABA excretion by advancing the last PABA dosage by 3 h, from 18:00 to 15:00 h. The limit previously determined for a complete 24 h urine using the HPLC method was 77.9% (187 mg; Jakobsen et al, 1997), which is almost the same limit as found in the present study (75.2% (183 mg)).

The higher specificity of the HPLC method compared to the usually employed colorimetrical method has been demonstrated by us (Jakobsen et al, 1997) and others (Chan et al, 1988; Kastel et al, 1994; Yung-Jato et al, 1988). Mean PABA recovery in this study was 87.4%, somewhat lower than mean recoveries found when using colorimetry. The lower PABA recoveries with HPLC are probably explained by the co-determination of aromatic amines from other sources, including commonly used drugs, such as acetaminophen and sulphonamides, with colorimetry.

The rate of decline of PABA recovery in this study of healthy subjects was about 1% per decennium from age 30. The lowest acceptable limit for a complete 24 h collection for the group of elderly (>70 y) can be calculated to 72.9% (175 mg). Upholding the lower 95% confidence limit of 77.9% PABA for a complete 24 h urine collection would reject two urines from this group of elderly as incomplete that in fact were complete collections.

The explanation for the gradual decline in PABA recoveries with age is not known. A study by Leclercq et al (1991) showed that 24 h PABA excretion was 15% lower in elderly than in a young control group. One-third of this decrease was recovered in the following 24 h urine and, consequently, could be explained by delayed excretion of subjects aged >70 y, however the major cause was suggested to be unreported spilling of small volumes of urine. This is, of course, also a possibility in our study, but it is conspicuous that both 24 h PABA excretions (PABA15 and PABA18) decline gradually and with similar rate with age of the subjects.

After rapid absorption from the gut, PABA is extensively metabolized and conjugated in the liver, before being cleared by the kidneys and excreted in the urine. Endogenous creatinine clearance—and accordingly glomerular filtration rate—decreases with age at a rate approximately 1% per year after 40 y (Lubran, 1995). PABA metabolites are also excreted by active tubular secretion (Tune et al, 1969), which also decreases with age. Thus, reduced 24 h PABA recovery with increasing age could be caused by a delayed renal clearance due to these reductions in the functions of the aging kidneys, as has been demonstrated for other water-soluble drugs mostly cleared by the kidneys (Lobel & Bergeron, 1987). Nevertheless, a delayed excretion—if this is the explanation for the lower recovery—cannot be overcome by advancing the last dosage of PABA by 3 h, from 18:00 to 15:00 h.

It has been suggested that the PABA method becomes unreliable in patients with plasma creatinines in excess of normal values (Bingham et al, 1992). However, plasma creatinines usually fail to increase with age because creatinine production, which is dependent on muscle mass, falls at nearly the same rate as the renal clearance of creatinine (Lindeman, 1993). Because of this and because invasive tests complicate any field study involving healthy subjects we opted not to include measurements of plasma creatinines in this study.

Decreased PABA recovery with age could also be explained by a decreased bioavailability. Overall gastro-intestinal function undergoes little change and is well preserved with age. However, PABA is consumed by bacteria to produce needed folic acid for proliferation, and bacterial overgrowth due to chronic atrophic gastritis and the inability to secrete gastric acid, observed in one-third of the elderly, might decrease availability of PABA (Saltzman & Russell, 1998).

Lastly, increased formation of metabolites that are not measured by the used HPLC method, could explain the results. Although the quality control of the analytical method included tests for complete conversion of the metabolites including glucuronides (Kitamura et al, 1960), there might still be unrecognized PABA metabolites that are not converted to PABA by the alkaline hydrolysis.

We found that 75–90% of the marker had been recovered in the urine 8 h after a single dosage of PABA had been taken. These are generally lower recoveries than those observed in four subjects studied by Bingham and Cummings (1983). An explanation for the difference could be that the subjects in that study to force diuresis had to drink 1 l of water with the PABA ingestion and furthermore were urged to drink water during the study, while our eight subjects were studied under their normal lifestyle conditions.

PABA test normally implies that a total of 240 mg is ingested, split in three doses of 80 mg taken at the main meals, usually around 8 am, 12 am and 6 pm. Our study showed that with this dosage schedule the lower limit for a complete 24 h urine collection (78.1%) was exceeded in all subjects at 12 pm and that a total of 21 (25%) individual urine specimens were collected after this recovery was achieved. Bingham and Cummings (1983) argued that there was no need for a dose of PABA at bedtime to cover the 24 h period, since almost all urine specimens contained significant amounts of PABA. However, their study was performed with subjects in a calorimeter under strictly controlled conditions and, unfortunately the number of urinations during nighttime is not indicated. Urinations after bedtime occur frequently-as this study shows-and it is probably these specimens that most often are 'forgotten', giving rise to erroneously low solute excretions in supposedly 24 h urines.

In conclusion this study has shown that there is a gradual decline of PABA recovery with age and that this decline cannot be overcome by advancing the dosage schedule. It is known that a substantial proportion of urine collections are truly incomplete (Heitmann, 1993; Bingham et al, 1995), however among the 'incomplete' collections some, especially in the elderly, will be rejected unjustly (false-negatives). Also, with the currently recommended dose schedule (PABA taken with the main meals) the risk of false-positive 24 h urine collections prevail, ie collection periods less than 24 h are considered to be complete. With refinement of the PABA test procedure, ie employing a specific analytical method and age-dependent cut-off values, the test may achieve a higher specificity and sensitivity.

References

1. Bingham, S & Cummings, JH (1983). The use of 4-aminobenzoic acid as a marker to validate the completeness of 24 h urine collections in man. Clin. Sci., 64, 629–635. | PubMed | ISI | ChemPort |
2. Bingham, S & Cummings, JH (1985). Urine nitrogen as an independent validatory measure of dietary intake: a study of nitrogen balance in individuals consuming their normal diet. Am. J. Clin. Nutr., 42, 1276–1289. | PubMed | ChemPort |
3. Bingham, SA, Murphy, J, Waller, E, Runswick, SA, Neale, G, Evans, D & Cummings, JH (1992). para-Amino benzoic acid in the assessment of completeness of 24-hour urine collections from hospital outpatients and the effect of impaired renal function. Eur. J. Clin. Nutr., 46, 131–135.
4. Bingham, SA, Cassidy, A, Cole, TJ, Welch, A, Runswick, SA, Black, AE, Thurnham, D, Bates, C, Khaw, KT, Key, TJA & Day, NE (1995). Validation of weighed records and other methods of dietary assessment using the 24 h urine nitrogen technique and other biological markers. Br. J. Nutr., 73, 531–550. | Article | PubMed | ChemPort |
5. Black, AE, Bingham, SA, Johansson, G & Coward, WA (1997). Validation of dietary intakes of protein and energy against 24 hour urinary N and DLW energy expenditure in middle-aged women, retired men and post-obese subjects: comparisons with validation against presumed energy requirements. Eur. J. Clin. Nutr., 51, 405–413. | Article | PubMed | ISI | ChemPort |
6. Chan, K, Miners, JO & Birkett, DJ (1988). Direct and simultaneous high-performance liquid chromatographic assay for the determination of P-aminobenzoic acid and its conjugates in human urine. J. Chromatogr., 426, 103–109.
7. Heitmann, BL (1993). The influence of fatness, weight change, slimming history and other lifestyle variables on diet reporting in Danish men and women aged 35–65 years. Int. J. Obes. Relat. Metab. Disor., 17, 329–336.
8. Jakobsen, J, Ovesen, L, Fagt, S & Pedersen, AN (1997). Para-aminobenzoic acid used as a marker for completeness of 24 hour urine: assessment of control limits for a specific HPLC method. Eur. J. Clin. Nutr., 51, 514–519. | Article | PubMed | ChemPort |
9. Kastel, R, Rosival, I & Blahovec, J (1994). Simultaneous determination of P-aminobenzoic acid and its metabolites in urine by high performance liquid chromatography. Biomed. Chromatogn., 8, 294–296.
10. Kitamura, M, Nakao, M & Yanagisawa, I (1960). Metabolism of para-aminobenzoic acid. J. Biochem., 47, 60–68.
11. Leclercq, C, Maiani, G, Polito, A & Ferro-Luzzi, A (1991). Use of PABA test to check completeness of 24-h urine collections in elderly subjects. Nutrition, 7, 350–354.
12. Lindeman, RD (1993). Assessment of renal function in the old. Special considerations. Clin. Lab. Med., 13, 269–277.
13. Lobel, M & Bergeron, MG (1987). Pharmacokinetics in the elderly. Studies on ciprofloxacin. Am. J. Med., 82, (4A)108–114.
14. Lubran, MM (1995). Renal function in the elderly. Ann. Clin. Lab. Sci., 25, 122–133. | PubMed |
15. Saltzman, JR & Russell, RM (1998). The aging gut. Nutritional issues. Gastroenterol. Clin. N. Am., 27, 309–324.
16. Tune, BM, Burg, MB & Patlak, CS (1969). Characteristics of PAHA transport in proximal renal tubules. Am. J. Physiol., 217, 1057–1063. | PubMed | ISI | ChemPort |
17. Yung-Jato, LLY, Durie, PR & Soldin, SJ (1988). Liquid-chromatographic measurement of P-aminobenzoic acid and its metabolites in serum. Clin. Chem., 34, 2235–2238.

Acknowledgements

We wish to thank Heddi Worsøe, Kirsten Pinndal and Morten Lindberg for their skilful assistance with the analytical work.