Absorption and Transport

Absorption of nicotinic acid and nicotinamide from the stomach and the intestine is rapid (Bechgaard and Jespersen, 1977) and at low concentrations is mediated by sodium ion-dependent facilitated diffusion. At higher concentrations, passive diffusion predominates, with doses of 3 to 4 g of niacin almost completely absorbed (Bechgaard and Jespersen, 1977). Glycohydrolases in the liver and intestines catalyze the release of nicotinamide from NAD (Henderson and Gross, 1979). Nicotinamide is then transported to tissues to be used in synthesis of NAD when needed. Both forms of the vitamin enter cells by simple diffusion; however, both nicotinic acid and nicotinamide also enter erythrocytes by facilitated transport (Lan and Henderson, 1968).

Metabolism and Excretion

The niacin coenzymes NAD and NADP are synthesized in all tissues of the body from nicotinic acid or nicotinamide. Tissue concentrations of NAD appear to be regulated by the concentration of extracellular nicotinamide, which in turn is under hepatic control and is hormonally influenced. Hydrolysis of hepatic NAD allows the release of nicotinamide for transport to tissues that lack the ability to synthesize the NAD and NADP coenzymes from tryptophan. In the liver some excess plasma nicotinamide is converted to storage NAD (i.e., NAD not bound to enzymes). Tryptophan and nicotinic acid also contribute to storage NAD following the biosynthetic pathway, going through NAMN, which is then reamidated to NAD. In the degradation of NAD, the nicotinamide formed can be reconverted to NAD via nicotinamide ribonucleotide. Nicotinamide can be deamidated in the intestinal tract by intestinal microflora (Bernofsky, 1980).

The body's niacin requirement is met not only by nicotinic acid and nicotinaminde present in the diet, but also by conversion from the dietary protein containing tryptophan. The relative contribution of tryptophan is estimated as follows: 60 mg of tryptophan = 1 mg of niacin = 1 mg of niacin equivalents (Horwitt et al., 1981).

Excess niacin is methylated in the liver to N¹-methyl-nicotinamide, which is excreted in the urine along with the 2- and 4-pyridone oxidation products of N¹-methyl-nicotinamide. The two major excretion products are N¹-methyl-nicotinamide and its pyridone derivative (Mrochek et al., 1976). The proportions differ somewhat depending on the amount and form of niacin ingested and the niacin status of the individual.

Method Used to Set the Adequate Intake

As for other nutrients, the Adequate Intake (AI) level for niacin is set for infants based on the observed mean intake of infants fed principally with human milk.

Ages 0 through 6 Months. One study (Ford et al., 1983) of unsupplemented mothers estimated the niacin concentration of their milk at 1.8 mg/L. The adequate intake for niacin for infants ages 0 through 6 months is based on the reported mean volume of milk consumed by this age group (0.78 L/day; see Chapter 2) and the estimate of the niacin concentration in human milk of 1.8 mg/L (0.78 L × 1.8 mg/L = 1.4 mg). The tryptophan content of human milk is approximately 210 mg/L (Committee on Nutrition, 1985). Because of the high rate of protein turnover and net positive nitrogen retention in infancy, it is likely that the standard method for estimating niacin equivalents (NEs) would overestimate the contribution from tryptophan. Thus, the AI for niacin for infants is given in milligrams of preformed niacin only—2 mg/day after rounding up.

Ages 7 through 12 Months. No difference in human milk composition was noted between the first and second 6 months of lactation (Ford et al., 1983). If the reference body weight ratio method described in Chapter 2 to extrapolate from the AI for niacin for infants ages 0 through 6 months is used, the AI for preformed niacin for the older infants is 2 mg/day after rounding. The second method (see Chapter 2), extrapolating from the Estimated Average Requirement (EAR) for adults and adjusting for the expected variance to estimate a recommended intake, gives an AI of 4.1 mg of NEs. A 15 percent coefficient of variation (CV) is used (for discussion, see “Niacin EAR and RDA Summary, Ages 19 Years and Older”). This AI value is expected to be higher than that obtained with the first method because it reflects NEs rather than preformed niacin.

Alternatively, the AI for niacin for infants ages 7 through 12 months can be calculated by using the estimated niacin content of 0.6 L of human milk and the average volume consumed by this age group (niacin content equals 1.1 mg), and adding the amount of niacin provided by solid foods (8 mg), as estimated by Montalto et al. (1985) (see Chapter 2). The result in NEs equals more than 9 mg/day. This value was judged to be unreasonably high because it is more than twice the extrapolated values and because mean intakes of older age groups appear to be much higher than their requirements. Thus the AI for niacin given in NEs is 4 mg/day for infants ages 7 through 12 months—the value extrapolated from estimates of adult requirements.

Niacin AI Summary, Ages 0 through 12 Months

Method Used to Estimate the Average Requirement

No data were found on which to base the EAR for niacin for children or adolescents. In the absence of additional information, EARs and RDAs for children and adolescents have been estimated by using the method described in Chapter 2, which extrapolates from adult values.

Niacin EAR and RDA Summary, Ages 1 through 18 years

The RDA for niacin is set by using a coefficient of variation (CV) of 15 percent (see Chapter 1 and the discussion of adult requirements that follows) because information is not available on the standard deviation of the requirement for these age groups; the RDA is defined as equal to the EAR plus twice the CV to cover the needs of 97 to 98 percent of the individuals in the group (therefore, for niacin the RDA is 130 percent of the EAR).

Method Used to Estimate the Average Requirement

The best biochemical measure for estimating the average requirement was judged to be niacin metabolite excretion data, namely the metabolites N¹-methyl-nicotinamide and its 2-pyridone derivative. These measures have been reported the most extensively. The niacin intakes that result in these measures being above the levels considered barely adequate represent some degree of body pool reserve. Niacin metabolites are not excreted until adequate tryptophan is available to meet needs for the synthesis of protein, nicotinamide adenine dinucleotide, and nicotinamide adenine dinucleotide phosphate (Fu et al., 1989; Vivian et al., 1958). Metabolite excretion measures are more sensitive to niacin depletion than are other biochemical measures, such as blood levels of pyridine nucleotides or tryptophan. Whereas pellagra is prevented with NEs of about 11 mg/day (at 2,500 kcal/day), restoration of niacin metabolite excretion beyond minimal occurs at intakes of NEs of about 12 to 16 mg/day. Excretion of N¹-methyl-nicotinamide rather than the 2-pyridone derivative is preferred as the target measure for estimating the niacin requirement because this metabolite better differentiates marginal from adequate niacin intakes (Jacob et al., 1989), interpretive guidelines exist for this metabolite (ICNND, 1963), and more data are available for this than for other metabolites.

An average niacin requirement can be estimated as the niacin intake corresponding to an excretion of N¹-methyl-nicotinamide that is above the minimal excretion at which pellagra symptoms occur. Urinary excretion of N¹-methyl-nicotinamide was found to be 1.2 mg/day in men on low but adequate niacin diets that resulted in no signs or symptoms of pellagra (Horwitt et al., 1956; Jacob et al., 1989). Urinary N¹-methyl-nicotinamide excretion was found to be 0.6 mg/day in three females on corn-based, niacin-deficient diets that resulted in clinical signs and symptoms of pellagra (Goldsmith et al., 1952, 1955). A urinary excretion value for N¹-methyl-nicotinamide of 1.0 mg/day has been chosen as an interpolated level of niacin excretion; it reflects a niacin intake that is above the intake that results in clinical niacin deficiency and thus is minimal or barely adequate.

Niacin intakes (expressed in NEs) that would correspond to N¹-methyl-nicotinamide excretions of 1.0 mg/day are calculated from the results of four experimental studies in which NEs of 8 to 12 mg/day were fed (Table 6-1). A study by Patterson and colleagues (1980) was not used because no individual data were provided. In calculating the niacin intakes that would result in the N¹-methyl-nicotinamide excretion level of 1 mg/day, a linear relationship was assumed. This approach is conservative; if more niacin was fed, the ratio of niacin to the excretion of N¹-methyl-nicotinamide would become somewhat lower. Because all the studies were judged satisfactory in quality, a weighted average of the results was calculated. The overall average intake equivalent to the excretion of 1 mg/day of N¹-methyl-nicotinamide is 11.6 ± 3.94 (standard deviation), with a CV of 34 percent.

TABLE 6-1

Experimental Human Studies of Niacin Intake and Urine N¹-methylnicotinamide Excretion.

From Table 6-1, it might be interpreted that women require more niacin per 1,000 kcal than do men. However, the studies are too few and not sufficiently comparable to justify such a conclusion, especially considering a lack of evidence that utilization of niacin is less efficient for women than for men. The studies varied widely in methods, including the types of foods provided, which could have marked effects on the bioavailability of niacin. It is thus assumed that women have a slightly lower requirement than do men because of their size and average energy utilization. Therefore, the EAR is estimated to be 12 mg of NEs for men and 11 mg of NEs for women. Data are not available for determining whether the niacin requirement changes with age in adults.

The results of other studies were examined (Goldsmith, 1958; Goldsmith et al., 1956; Horwitt, 1958; Leklem et al., 1975), but coexisting deficiencies, a lack of valid dietary measurements, or both made it necessary to exclude them when deriving the EAR.

Niacin EAR and RDA Summary, Ages 19 Years and Older

On the basis of the data in Table 6-1 and with a minor (approximately 10 percent) decrease for energy in women, the EAR is set at 12 mg/day of NEs for men and 11 mg/day of NEs for women.

The data in Table 6-1 also suggest a CV of the niacin requirement that is greater than 10 percent. The wide variation in the efficiency of converting tryptophan to niacin may contribute to the large apparent variation. Thus, a CV of 15 percent is used; the RDA is defined as equal to the EAR plus twice the CV to cover the needs of 97 to 98 percent of the individuals in the group (therefore, for niacin the RDA is 130 percent of the EAR).

Method Used to Estimate the Average Requirement

There is no direct evidence that would suggest a change in the niacin requirement during pregnancy. To derive the EAR for pregnant women, it is estimated that the need for niacin increases by 3 mg/day of NEs to cover increased energy utilization and growth in maternal and fetal compartments, especially during the second and third trimesters.

Niacin EAR and RDA Summary, Pregnancy

By adding 3 mg of NEs to the EAR of 11 mg of NEs for nonpregnant, nonlactating women, the EAR for pregnancy becomes 14 mg of NEs. No adjustment is made for the woman's age.

The data in Table 6-1 suggest a CV for the niacin requirement that is greater than 10 percent. The wide variation in the efficiency of converting tryptophan to niacin may contribute to the larger apparent variation. Thus, a CV of 15 percent is used because information is not available on the standard deviation of the requirement for pregnant women; the RDA is defined as equal to the EAR plus twice the CV to cover the needs of 97 to 98 percent of the individuals in the group (therefore, for niacin the RDA is 130 percent of the EAR).

Method Used to Estimate the Average Requirement

An estimated 1.4 mg of preformed niacin is secreted daily during lactation. Added to this is a small amount (1 mg) to cover energy expenditure involved in milk production. Thus, the additional amount of niacin needed is 2.4 mg/day of NEs for women who are exclusively breastfeeding an infant.

Niacin EAR and RDA Summary, Lactation

Adding 2.4 mg of NEs to the EAR of 11 mg of NEs for non-pregnant, nonlactating women gives an EAR for niacin for lactation of 13.4 mg, rounded down to 13.

The data in Table 6-1 suggest a CV for the niacin requirement that is greater than 10 percent. The wide variation in the efficiency of converting tryptophan to niacin may contribute to the larger apparent variation. Thus, a CV of 15 percent is used because information is not available on the standard deviation of the requirement during lactation; the RDA is defined as equal to the EAR plus twice the CV to cover the needs of 97 to 98 percent of the individuals in the group (therefore, for niacin the RDA is 130 percent of the EAR).

Adverse Effects

There is no evidence of adverse effects from the consumption of naturally occurring niacin in foods. Therefore, this review is limited to evidence concerning intake of niacin as a supplement, food fortificant, or pharmacological agent.

One report showed adverse effects after consumption of bagels to which 60 times the normal amount of niacin had been added inadvertently (CDC, 1983). Most of the data on the adverse effects of excess niacin intake are from studies and case reports involving patients with hyperlipidemia or other disorders who were treated with pharmacological preparations containing immediate-release nicotinic acid or slow- or sustained-release nicotinic acid. The Tolerable Upper Intake Level (UL) developed here applies to all forms of niacin added to foods or taken as supplements (e.g., immediate-release, slow or sustained-release nicotinic acid, and niacinamide [nicotinamide]). Adverse effects such as nausea, vomiting, and signs and symptoms of liver toxicity have been observed at nicotinamide intakes of 3,000 mg/day (Rader et al., 1992) compared with intakes of nicotinic acid of 1,500 mg/day (McKenney et al., 1994). The generic term niacin may be considered interchangeable with nicotinic acid. As described below, the critical adverse effect selected was flushing to the extent that it results in a change in the dosing pattern or withdrawal from treatment.

Vasodilatory Effects (Flushing). The term flushing covers a burning, tingling, and itching sensation as well as a reddened flush primarily on the face, arms, and chest. Flushing occurs in many patients treated with nicotinic acid therapeutically. It is often accompanied by pruritus, headaches, and increased intracranial blood flow (Miller and Hayes, 1982). Occasionally, it is accompanied by pain (Bean and Spies, 1940). Case reports and clinical trials have reported flushing effects at oral doses of 30 to 1,000 mg/day within 30 minutes to 6 weeks of the initial dose (CDC, 1983; Estep et al., 1977; Henkin et al., 1990; McKenney et al., 1994; Sebrell and Butler, 1938; Spies et al., 1938). Although flushing is a transient effect, it often results in patients deciding to withdraw from treatment.

In a study of the flushing effects of nicotinic acid and other pyridine compounds in humans, Bean and Spies (1940) suggest that pyridine compounds without a carboxyl radical in the 3 position of the pyridine ring do not produce flushing effects. Nicotinamide, which does not have a carboxyl radical in the 3 position, does not appear to be associated with flushing.

Flushing appears to be more closely related to a continuous rise in plasma nicotinic acid concentrations than to the absolute dose. Tolerance to nicotinic acid-induced flushing can develop whereby the effects are minimized when the dose is slowly increased over time (Stern et al., 1991). Flushing effects can also be reduced somewhat by taking niacin with food (Knodel and Talbert, 1987). Flushing has been shown to be prostaglandin mediated, and tolerance results from reduction in prostaglandin levels with continued administration (Morrow et al., 1989, 1992).

Gastrointestinal Effects. In addition to flushing, nonspecific gastrointestinal effects are common in patients treated with nicotinic acid (Knodel and Talbert, 1987), especially with slow-release preparations (McKenney et al., 1994; Rader et al., 1992). The gastrointestinal effects in patients taking the slow-release form have been shown to be associated with liver enzyme elevations (Gibbons et al., 1995).

Hepatotoxicity. The hepatotoxicity of niacin has been demonstrated in numerous case reports (Clementz and Holmes, 1987; Dalton and Berry, 1992; Einstein et al., 1975; Etchason et al., 1991; Ferenchick and Rovner, 1989; Frost, 1991; Goldstein, 1988; Henkin et al., 1990; Hodis, 1990; Knapp and Middleton, 1991; Knopp, 1991; McKenney et al., 1994; Mullin et al., 1989; Palumbo, 1991; Patterson et al., 1983). In the most severe cases, patients develop liver dysfunction and fulminant hepatitis and may progress to stage 3 and 4 encephalopathy requiring liver transplantation (Clementz and Holmes, 1987; Hodis, 1990; Mullin et al., 1989). The most frequently observed manifestations of niacin-induced hepatitis are jaundice and increased levels of serum transaminases (Etchason et al., 1991). These effects are typically associated with high doses (3 to 9 g/day of niacin) used to treat patients with hypercholesterolemia for periods of months to years (Clementz and Holmes, 1987; Einstein et al., 1975; Pardue, 1961; Patterson et al., 1983; Rivin, 1959; Winter and Boyer, 1973). Almost all of the patients in these case reports were taking the slow-release form. A recent double-blind comparison study suggested that the slow-release form is more hepatotoxic than the immediate-release form (McKenney et al., 1994). However, another study reported hepatotoxicity with both forms of niacin (Gray et al., 1994).

Glucose Intolerance. Large doses (3 g/day) of nicotinic acid used to treat patients with hypercholesterolemia have produced impaired glucose tolerance in otherwise apparently healthy individuals (Miettinen et al., 1969). The adverse effects on glucose tolerance have been observed during short- and long-term administration of the drug (Schwartz, 1993).

Ocular Effects. Niacin treatment may produce significant ocular effects, including blurred vision, toxic amblyopia, macular edema, and cystic maculopathy. However, there are very few cases in the literature (Fraunfelder et al., 1995). Doses of 1.5 to 5 g/day of niacin have been associated with ocular effects (Fraunfelder et al., 1995; Gass, 1973; Millay et al., 1988). Niacin-induced ocular effects appear to be reversible and dose dependent.

Summary

Flushing is the adverse effect first observed after excess niacin intake and is generally observed at lower doses than are other effects. Flushing that results in patients deciding to change the pattern of niacin intake (i.e., reduce the amount taken at a time or withdraw from treatment) was selected as the most appropriate endpoint on which to base a UL. Although nicotinamide appears not to be associated with flushing effects, a UL for nicotinic acid that is based on flushing is considered protective against potential adverse effects of nicotinamide. The data on hepatotoxicity are considered less relevant to the general population because they involve large doses taken for long periods of time for the treatment of a medical condition.

Adults

Data Selection. The data sets used to identify the lowest-observed-adverse-effect level (LOAEL) for niacin included anecdotal reports and clinical trials involving oral intake of niacin by healthy individuals. Studies involving parenteral administration were not considered in the dose-response assessment. Studies involving immediate-release forms of niacin were considered more relevant to niacin intake by the general population than were studies involving sustained-release forms.

Identification of a LOAEL. The data are not adequate to identify a no-observed-adverse-effect level (NOAEL) for flushing. To identify a LOAEL, flushing reactions were considered if they resulted in a patient either changing the form or amount of niacin used or withdrawing from treatment. A LOAEL of 50 mg/day was identified based on a study by Sebrell and Butler (1938) in which four (66 percent) of six persons experienced a flushing sensation after oral intake of 50 mg/day of nicotinic acid given with meals for 92 days. In one of the four subjects who experienced flushing effects, the daily dose of 50 mg was given as 25 mg in the morning and evening. Although this study also reported a flushing reaction in one of six subjects taking 30 mg of nicotinic acid daily on day 32 of intake, this reaction was not bothersome enough to change the dosing pattern. Sebrell and Butler (1938) was selected as the critical study for identifying a LOAEL and deriving a UL because it provides the lowest effect level. A study by Spies et al. (1938) provides supportive evidence for a LOAEL of 50 mg/day. In this study, five of 100 individuals (5 percent) experienced flushing after a single oral dose of 50 mg of nicotinic acid, 50 individuals (50 percent) experienced flushing after 100 mg, and all individuals experienced flushing after 500 mg.

There is one case report showing that 14 of 69 persons (20 percent) experienced onset of rash, pruritus, and a sensation of warmth about 30 minutes after consuming one or more pumpernickel bagels to which niacin had been inadvertently added from an improperly labeled container (CDC, 1983). The bagels were found to contain an average of 190 mg of niacin.

Uncertainty Assessment. Because of the transient nature of the flushing effect, a small uncertainty factor (UF) of 1.5 was selected. A smaller UF was not appropriate because it is applied to a LOAEL rather than a NOAEL.

Derivation of a UL. A LOAEL of 50 mg/day was divided by a UF of 1.5 to obtain the UL for adults of 35 mg/day, a rounded estimate.

Other Life Stage Groups

For infants the UL was judged not determinable because of a lack of data on adverse effects in this age group and concern about the infant's ability to handle excess amounts. To prevent high levels of intake, the only source of intake for infants should be from food. No data were found to suggest that other life stage groups have increased susceptibility to flushing effects from excess niacin intake. Therefore, the UL of 35 mg/day is also set for pregnant and lactating adult women. The UL of 35 mg/day for adults was adjusted for children and adolescents on the basis of relative body weight as described in Chapter 3 and by using reference weights from Chapter 1, Table 1-2. Values have been rounded down.

Special Considerations

A review of the literature identified individuals with the following conditions as being distinctly susceptible to the adverse effects of excess niacin intake: hepatic dysfunction or a history of liver disease, diabetes mellitus, active peptic ulcer disease, gout, cardiac arrhythmias, inflammatory bowel disease, migraine headaches, and alcoholism. Therefore, people with these conditions may not be protected by the UL for niacin for the general population.