Registration Dossier

Data platform availability banner - registered substances factsheets

Please be aware that this old REACH registration data factsheet is no longer maintained; it remains frozen as of 19th May 2023.

The new ECHA CHEM database has been released by ECHA, and it now contains all REACH registration data. There are more details on the transition of ECHA's published data to ECHA CHEM here.

Diss Factsheets

Administrative data

Description of key information

The European Food Safety Authority (EFSA) Scientific Committee on Food - Scientific Panel on Dietetic Products, Nutrition and Allergies (2006) set a tolerable upper intake level for nicotinic acid. The tolerable upper intake level for nicotinic acid of 10 mg/day is based on the available data indicating occasional flushing at 30 mg per day, using an uncertainty factor of 3 to allow for the fact that a slight effect was reported, and that the study was performed in a small number of subjects, but taking into account the steep dose-response relationship. This upper level is 300-fold below the dose frequently used clinically for the treatment of hypercholesterolaemia (3 g/day) and which is associated with a high incidence of serious adverse reactions.

Additional information

Human data on nicotinic acid is presented in a weight of evidence approach. Available human data was previously reviewed by the European Commission Scientific Committee on Food, the EFSA Panel on Additives and Products or Substances used in Animal Feed, the US FDA and the UK Food Standards Agency Expert Group on Vitamins and Minerals. An OECD SIDS review concluded that Niacin is essential for human and animal health. Based on all available human information the substance does not present a hazard to human health.

The European Food Safety Authority (EFSA) compiled relevant human data (2006):

HUMAN DATA

The principal identification of hazards associated with excessive intakes of niacin have arisen from studies in which high doses of nicotinic acid have been used for its therapeutic effects in lowering blood cholesterol and blood hyperlipidaemias. The most comprehensive study was that conducted by the Coronary Drug Project Research Group (1975). A number of hazards have been reported to be associated with high doses of nicotinic acid. These have been summarised by the US National Academy of Sciences Institute of Medicine in their evaluation of dietary reference intakes.

In addition, nicotinamide has been investigated as a method for reducing the risk of the development of diabetes (Knip et al, 2000). Studies have shown that nicotinamide can afford protection in an animal model of immune mediated insulin-dependent diabetes (Reddy et al, 1990), and it has been investigated in a number of clinical trials, some of which are still ongoing.

Vasodilatory effects (flushing)

Vasodilation is commonly seen in patients given high doses of nicotinic acid for the treatment of hyperlipidaemias. Very large single doses cause hypotension, although tolerance develops to this effect after several days of continued high dose intake. In general, flushing is a mild and transient effect although in many clinical trials it has resulted in patients withdrawing from treatment. The flushing activity appears to be related to the presence of a carboxyl group on the pyridine nucleus since compounds lacking this function, including nicotinamide, are not associated with facial flushing (Bean and Spies,1940). Flushing is associated with periods of rapid rises in blood concentrations, and sustained-release formulations were developed for the use of nicotinic acid in the treatment of hypercholesterolaemia, in order to minimise this side-effect. Flushing is produced via prostaglandin D2 release (Morrow et al, 1989 and 1992) and a “niacin flush-test” has been used as a method of investigating essential fatty

acid metabolism (Glen et al, 1996). Tolerance develops due to decreased formation of prostaglandin D2 on repeated dosage (Stern et al, 1991). Although flushing is not a clearly adverse effect and single oral doses of 100 mg do not alter heart rate or blood pressure, some patients in the study of Spies et al(1938) reported dizziness after oral nicotinic acid (doses not defined). Theoretically if flushing occurred in the elderly, it could exacerbate mild postural hypotension, and could increase the risk of falls, which are a common cause of morbidity in the elderly. This risk relates to taking supplements containing nicotinic acid (not nicotinamide), especially if taken on an empty stomach.

Gastrointestinal effects

Gastrointestinal effects such as dyspepsia, diarrhoea and constipation are common in patients with hypercholesterolaemia given high doses of nicotinic acid (3 g/day - especially as the sustained-release formulation; Knopp et al, 1985).

Hepatotoxicity

Severe and potentially life-threatening hepatotoxicity has been associated with treatment of patients with 3-9 g nicotinic acid per day for periods of months or years for the treatment of hypercholesterolaemia. Severe cases show liver dysfunction and fulminant hepatitis and may even proceed to encephalopathy requiring liver transplantation. Many of the patients showing hepatotoxicity were taking the slow release formulation of the compound, so that in contrast to the flushing discussed above, the development of hepatic toxicity is a function of long-term chronic exposure to relatively constant levels rather than the fluctuating levels and rapid rises which produce flushing.

Glucose intolerance

Nicotinic acid (3 g/day) has been reported to impair glucose tolerance in otherwise healthy individuals treated for hypercholesterolaemia.

Other effects

There have been rare reported cases of a range of effects including blurred vision, macular oedema and increased plasma homocysteine concentrations in patients given high doses of nicotinic acid. These effects were reported at doses similar to those producing hepatic dysfunction, and were reversible upon cessation of high dose treatment. There has been a single report of a possible association with congenital malformation in women taking nicotinamide during early pregnancy (Nelson and Forfar, 1971). On the basis of their retrospective survey of drug and vitamin prescriptions during pregnancy in 1369 mothers, the authors reported that a significantly (P<0.05) higher proportion of women with infants showing abnormalities took nicotinamide in the first 56 days (5/458), compared with mothers of normal babies (1/911). In contrast no such relationship was found during later phases of pregnancy or over the whole of the pregnancy. The paper

did not report the doses of nicotinamide taken. This finding is in contrast to the results of the large multicentre study on vitamins and the prevention of neural tube defect (MRC Vitamin Study Research Group, 1991). In that study 1817 women with high risk for producing a baby with neural tube defect were randomised into 4 groups; one group received folic acid, one group a multivitamin preparation

(that did not include folate but contained 15 mg/day of nicotinamide), one group was given both preparations and one group received neither preparation. Although the study focused on neural tube defects, any foetal malformation was recorded together with other pregnancy outcomes, and there was no difference in incidence between the multivitamin preparation and placebo.

DOSE-RESPONSE ASSESSMENT

Vasodilatory effects (flushing)

Low doses of nicotinic acid may produce mild but noticeable flushing when taken on an empty stomach (Hathcock, 1997) and this represents the adverse effect detected at the lowest doses. An early study (Smith et al, 1937) reported that a single oral dose of 60 mg nicotinic acid produced marked flushing, which was not associated with changes in heart rate or blood pressure. Spies et al (1938) reported flushing in 5% and about 50% of subjects given single oral doses of 50 mg and 100 mg nicotinic acid, respectively. The dose-response for flushing was examined further by Sebrell and Butler (1938) who gave 3 groups of 6 subjects daily dose of 10, 30 or 50 mg nicotinic acid for 92 days as single oral doses given in solution added to tomato juice and consumed with the mid-day meal; flushing was reported intermittently by 4, 2 and 0 of the subjects given 50, 30 and 10 mg, respectively.

The response is possibly related to periods of rapid increase in plasma concentrations of nicotinic acid, because the response is greater after intravenous dosage and is blunted if taken orally with food (Bean and Spies, 1940). Rash, pruritus and a sensation of warmth was reported following the consumption of pumpernickel bagels, accidentally made to contain 190 mg nicotinic acid per bagel (MMWR, 1983) and following the consumption of cooked meat containing 225 mg/100 g (Press and Yeager, 1962). This hazard does not seem to be related to nicotinamide. The facial flushing associated with low doses of nicotinic acid can be prevented by co-administration of an inhibitor of prostaglandin synthesis such as aspirin (although this is not always recommended - Schuna, 1997).

Gastrointestinal effects

Gastrointestinal effects such as dyspepsia, diarrhoea and constipation are common in patients given high doses of nicotinic acid for hypercholesterolaemia. Ruffin (cited in Sebrell and Butler, 1938) reported Nicotinic Acid and Nicotinamide (Niacin) nausea and vomiting in 2 out of 10 subjects given 1 g of nicotinic acid. Spies et al (1938) reported nausea and vomiting in subjects given oral doses of 300-1500 mg of nicotinic acid. Nausea is a common adverse effect in the studies of patients given 3 g of nicotinic acid daily for hypercholesterolaemia.

Hepatotoxicity

The first report of hepatotoxicity associated with the administration of nicotinic acid was in a study in dogs (Chen et al, 1938), which compared the toxicity of nicotine with nicotinic acid. In that study 2 dogs were given either 145 or 133 mg/kg bw/day nicotinic acid orally and both developed convulsions and excreted blood in their faeces about 2-3 weeks after treatment started. Post mortem observations included gastrointestinal adhesions, “fatty metamorphosis” of the liver and neuronal damage. The first case-report of hepatotoxicity of nicotinic acid in humans was in a 23 year old man who developed jaundice after taking 3 g per day for 72 weeks (Rivin, 1959). Subsequent case-reports included a man who had taken 3 g per day for 6 months (Pardue, 1961), and a woman who developed pruritus and jaundice after taking 3 g nicotinic acid (together with 3 g vitamin C and 100 mg pyridoxine for a psychological disturbance) per day for 2.5 years (Einstein et al, 1975). A survey of 66 patients treated with nicotinic acid, of whom 51 had taken 3 g/day for 12 months or more, found a high incidence of abnormal liver function tests (23 patients) while on treatment, with 2 patients developing jaundice (Berge et al, 1961).

Approximately one-third of the 1119 patients in the study of the Coronary Drug Project Research Group (1975), who received 3 g/day nicotinic acid for up to 5 years, were reported to have elevated serum glutamate-oxaloacetate transaminase (SGOT) and alkaline phosphatase levels. There have been a number of reports of individual cases of patients with severe hepatotoxicity resulting from the use of nicotinic acid for hypercholesterolaemia or hypertriglyceridaemia. Four cases of liver disease were associated with doses of 2.5 g of sustained-release nicotinic acid daily for 5 months, 1.5 g of sustained-release nicotinic acid per day for 3 months, 2.25 g of sustained-release nicotinic acid per day for an unrecorded period, and 2 g of sustained-release nicotinic acid daily for an unrecorded period (Coppola et al, 1994). In all cases, cessation of nicotinic acid administration resulted in resolution of the liver symptomatology.

A single case report gave some insight into the dose-response relationship for sustained-release nicotinic acid since symptoms of anorexia, fatigue and persistent nausea arose approximately one month at the end of a sequence of dose escalation from 1 g/day through 3 g/day for one month and finally 4 g/day for one month (Lawrence, 1993). A rapid reversal of the symptoms was found at 3 weeks after discontinuation of the nicotinic acid therapy.

Rader et al (1992) reviewed the available cases of hepatotoxicity and side-effects from conventional and sustained-release nicotinic acid and concluded that adverse effects were frequently seen shortly after an abrupt change from unmodified to sustained-release preparations. Their paper summarized both the dose and the duration of therapy in the different cases of hepatic toxicity and showed that in general toxicity was associated with doses of 3 g/day or more, although there were 2 cases who took less than 1 g/day for short periods (0.75 g conventional nicotinic acid per day for less than 3 months; 0.5 g sustained-release nicotinic acid for 2 months).

A comparison of an immediate release formulation and a sustained-release formulation of nicotinic acid in two groups of 23 patients with low density lipoproteinaemia studied the sequential effect of 0.5, 1, 1.5, 2 and 3 g per day for period of 6 weeks each. The therapeutic efficacy was similar for the two formulations but there were interesting differences in the side-effect profiles. About 39% of subjects on the immediate release formulation withdrew before completing the 3 g/day dose due to vasodilatory symptoms and fatigue, whereas 78% of subjects in the sustained-release group withdrew before completion of the study, primarily due to gastrointestinal tract symptoms, fatigue and changes in serum aminotransferases, indicative of hepatic dysfunction. Interestingly, the lowest dose of 0.5 g/day appeared to be better tolerated with the sustained-release preparation than the immediate release primarily because of vasodilatory symptoms (McKenney et al, 1994).

The study of McKenney et al has been criticised because the dosage regimen of twice daily administration was considered to minimise the tolerability of the protocol and give the greatest potential for side-effects. The authors of the critique (Kennan et al, 1994) report that there was only a 5% drop-out rate as a result of intolerance and toxicity after one year in a study of 1119 subjects receiving 3 g (1 g three times a day) of immediate release nicotinic acid. The study of McKenney et al was also criticised because of the high top dose administered since drop-out rates of only 3-4% had been reported in studies where the maximum dose was 2 g/day. The results of a multicentre study on the long-term safety and efficacy of a sustained-release preparation of nicotinic acid were reported by Guyton et al (1998). Nicotinic acid doses, ranging from 0.5-3 g were given once a day at bedtime to a total of 269 patients for a period up to 96 weeks. The average dose given at the end of the study was 2 g/day with a range from 1-3 g, which indicates the poor tolerability of doses greater than 2 g/day. The principal adverse effect was flushing which resulted in 4.8% of the participants discontinuing the study (although they were advised that they could take aspirin to reduce this symptom). Those individuals who showed flushing had an average of one episode of 1.2 hours duration every 4-5 weeks. A total of 9 patients showed elevated transaminase levels of at least 2 times the upper limit of normal. However 5 of these patients were on a combination therapy including nicotinic acid plus nystatin or a bile acid sequestrant. In 5 of the cases the transaminase elevation resolved while treatment with nicotinic acid continued and without a reduction in dose. Therefore this study demonstrates only mild hepatotoxicity in a group of subjects given controlled doses of sustained release nicotinic acid.

Dalton and Berry (1992) describe a single case of a woman who presented with hepatotoxicity after taking crystalline nicotinic acid for a period of 2 years and sustained-release formulation for a period of only 2 days prior to admission. Her symptoms on admission to hospital included hypothermia, hypotension and metabolic acidosis, and the authors suggested that this may have been a result of the change from conventional to sustained-release nicotinic acid associated with prolonged flushing and possibly significant transcutaneous heat loss. This observation is ironic, since the sustained release formulation was primarily developed to minimise the skin flushing reaction associated with conventional nicotinic acid (Rader et al, 1992). Some studies have suggested that sustained-release formulations of nicotinic acid produce a greater incidence of hepatotoxicity (Christensen et al, 1961; Knopp, 1989; Mullin et al, 1989; Henkin et al, 1990), although this is not a consistent observation in all studies (Gibbons et al, 1995).

Gray et al (1994) reported that the daily intakes of nicotinic acid in 42 elderly diabetic patients who developed hepatic dysfunction (2.33 ± 0.15 g/day) were significantly higher than the doses for the remaining 854 subjects (1.64 ± 0.03 g/day) who did not develop hepatic dysfunction.

Effects on prothrombin time have been reported in patients taking sufficient nicotinic acid to cause hepatic toxicity. Elevated prothrombin times have been reported in a small number of cases, which were associated with only mild elevation of transaminase levels so that blood-clotting disorders may become the limiting sign of hepatotoxicity in some cases. Three cases reported by Dearing et al (1992) were receiving 2.0, 2.0 and 3.0 g of nicotinic acid daily.

In contrast to the studies that have reported abnormal liver function in patients treated with nicotinic acid, a small study in the group of 30 patients with hyperlipidaemia who were given slow release nicotinic acid at 1 g/day for 2 months and then 2-3 g/day for 10 months reported a low incidence of symptoms other than skin flushing (which had an incidence of 26.7% - mostly at the start of the treatment period). There was no evidence of hepatic abnormalities as indicated by changes in serum aminotransferases, alkaline phosphatase or antipyrine test results (Chojnowska-Jezierska and Adamska-Dyniewska, 1998).

A large number of studies have defined the efficacy and tolerability of both conventional and sustained or controlled-release nicotinic acid in the treatment of hypercholesterolaemia and hyperlipidaemias.

The data from these studies provide adequate evidence of the hazard identification and some evidence of dose-response characterisation. A major problem with the use of such data for establishing an upper level is that the doses investigated were restricted to those that showed clinical efficacy in the conditions being treated (mostly 3 g/day), and there are few data available at lower levels (Rader et al, 1992). Hodis (1990) reported a case of acute hepatic failure, which was ascribed to treatment with 500 mg per day nicotinic acid, however there was no repeat challenge or other data to support causation (other than an absence of other recognised reasons).

Glucose intolerance

Although hyperglycaemia is a relatively rare side-effect associated with high doses of nicotinic acid, it can be of clinical significance. Administration of 3 g of nicotinic acid per day for 10-14 days to volunteers resulted in an increase in fasting blood glucose and immuno-reactive insulin in serum (Miettinen et al, 1969). An increase in blood glucose concentrations, glycosuria, elevated serum ketone bodies, and an increase requirement for hypoglycaemic medication were reported in 6 patients with diabetes mellitus, who were receiving between 1 g and 3 g of nicotinic acid daily for a period of 2 weeks or more (Molnar et al, 1964). Gray et al (1994) reported a high incidence of hyperglycaemia in elderly hyperlipidaemic patients who had been treated with high doses of nicotinic acid (average dose 1.7 g/day). Schwartz

(1993) described a patient who was hospitalised with severe hyperglycaemia following treatment with 3 g of nicotinic acid per day for 4 months; administration of insulin and oral hypoglycaemics reversed and stabilised the blood glucose levels.

Other effects and overall dose-response relationships

Thrombocytopaenia, which resolved on cessation of nicotinic acid treatment, was reported in a single patient who developed hepatitis 10 years after the initiation of nicotinic acid treatment (Reimund and Ramos, 1994).

The plasma concentrations of homocysteine were increased by 55% in patients with peripheral arterial disease who were treated with nicotinic acid (Garg et al, 1999). The 52 patients were a subgroup from a multicentre study in which patients were given increasing doses of 100, 500 and 1000 mg per day over periods of 3-4 weeks (in order to identify patients who tolerated nicotinic acid), following which the subjects were randomised to receive either placebo or nicotinic acid (up to 3 g per day). The plasma concentrations of homocysteine were measured at baseline, at randomisation and at 18 and 48 weeks

after randomisation. Plasma homocysteine increased from 13.1 ± 0.5 μM at baseline to 15.3 ± 0.8 μM at randomisation. After randomisation the levels increased further in those receiving nicotinic acid (to about 20 μM at 18 and 48 weeks; n=25 and 24, respectively), but decreased in those on placebo (to about 12 μM at 18 and 48 weeks; n=21 and 22, respectively). The clinical significance of this is unclear, but elevated plasma homocysteine is a recognised risk factor for coronary artery disease.

Severe reversible cystoid macular oedema was reported in 3 patients receiving high-doses of nicotinic acid (Gass, 1973). A survey of 116 patients who had received nicotinic acid (3 g or more per day) for treatment of hyperlipidaemia and a similar number of patients who were not treated with nicotinic acid revealed an increased incidence of decreased vision associated with sicca syndromes, eyelid oedema or macular oedema (Fraunfelder et al, 1995).

Because the majority of the data arise from studies designed to investigate the hypolipidaemic action of nicotinic acid, most of the data relate to doses of 1 g/day or more. In consequence, there are few data available on the tolerability and toxicity of doses less than 500 mg/day. In general the main adverse effect reported at intakes below the 500 mg/day has been flushing which is generally self-limiting in relation to continuation of treatment or intake of nicotinic acid.

High dose nicotinic acid (0.5-2.25 g daily) has been used for the treatment of severe hypercholesterolaemia in children. A retrospective review of 21 such cases reported similar adverse effects to those found in adults, with 6 children showing reversible dose-related elevations in serum transaminases, and 8 children who discontinued treatment because of flushing, abdominal pains and/or elevated serum transaminase levels. Hepatitis was reported in subjects with very high doses on a mg/kg bw/day basis (50, 67, 41, 34, 48 and 39 mg/kg bw/day) (Colletti et al, 1993).

In a study in the USA on elderly male veterans (age 62 +/- 9 years) the doses administered averaged 1.7 g/day with a mean duration of intake of 13 +/- 10 months (Gray et al, 1994). Almost one-half of the subjects discontinued treatment because of adverse effects with poor glycaemic control occurring in 41% of patients with diabetes mellitus. Probable and possible nicotinic acid-induced hepatotoxicity occurred in 2.2 and 4.7% of the patient group. These data indicate that the spectrum of toxicity is similar in elderly and young adults.

The side-effect profile of wax matrix sustained-release nicotinic acid was studied in groups of younger (< 50 years) and older (50-70 year old) hypercholesterolaemic subjects. The study was a randomized double-blind placebo controlled design of 8 weeks duration with doses of 1.0-2.0 g/day. Clinically significant side-effect included flushing, itching, tingling, upper gastrointestinal symptoms, constipation, diarrhoea, dizziness, palpitations and blurred vision; the overall incidence of adverse effects was similar in the two difference age groups (Keenan et al, 1992).

DERIVATION OF A TOLERABLE UPPER INTAKE LEVEL

The more severe forms of toxicity of nicotinic acid, as described above, occur principally at doses of greater than 500 mg/day. The limiting adverse effect at lower doses is flushing, and this has been reported at much lower intakes than the other adverse effects. The most severe and potentially life threatening adverse effects, such as hepatotoxicity, occur at doses one order of magnitude higher than have been reported for flushing. The dose of free nicotinic acid reported to produce flushing consistently in clinical studies is 50 mg/day (Sebrell and Butler, 1938; Spies et al, 1938). Spies et al (1938) reported a 5% incidence of flushing after a single oral dose of 50 mg nicotinic acid and a 50% incidence at 100 mg. The available data indicate that flushing would be unlikely to occur repeatedly in subjects given less than 50 mg/day, but occasional flushing was reported by Sebrell and Butler (1938) at a dose of 30 mg of nicotinic acid daily. A tolerable upper intake level for nicotinic acid of 10 mg/day is based on the available data indicating occasional flushing at 30 mg per day, using an uncertainty factor of 3 to allow for the fact that a slight effect was reported, and that the study was performed in a small number of subjects, but taking into account the steep dose-response relationship. This upper level is 300-fold below the dose frequently used clinically for the treatment of hypercholesterolaemia (3 g/day) and which is associated with a high incidence of serious adverse reactions. The only reports of flushing associated with the ingestion of nicotinic acid with food have occurred following the addition of free nicotinic acid to food prior to consumption. Although flushing might be considered a minor health effect, it has been used as the basis for setting the upper level for nicotinic acid, because of concerns about the possibility of a transient hypotensive episode, especially in the elderly.

Reference: EFSA Scientific Committee on Food - Scientific Panel on Dietetic Products, Nutrition and Allergies (2006) Tolerable Upper Intake Levels for Vitamins and Minerals.