Aniline

Toxicological Summary

Workers - Hazard via inhalation route

Systemic effects

Long term exposure

Hazard assessment conclusion:

DNEL (Derived No Effect Level)

Value:

7.7 mg/m³

Most sensitive endpoint:

repeated dose toxicity

DNEL related information

DNEL derivation method:

other: German Occupational Exposure Limit

Justification:

methemoglobin concentration in blood, the most sensitive effect in humans, was only slightly enhanced after exposure to the DNEL concentration of 7.7 mg aniline/m³ for a work shift duration (6-8 hours), consistently dropped after the end of exposure and after about 24 hours the levels were in the range of those observed for non-exposed individuals. Thus, repeated exposure in the range of the DNEL and below is not expected to cause adverse health effects.

Justification:

the relevant data were obtained in humans

Justification:

the human volunteer study already included susceptible persons (slow acetylators)

Justification:

the database is of exellent quality (recent human data)

Acute/short term exposure

Hazard assessment conclusion:

DNEL (Derived No Effect Level)

Value:

15.4 mg/m³

Most sensitive endpoint:

repeated dose toxicity

DNEL related information

Local effects

Long term exposure

Hazard assessment conclusion:

hazard unknown (no further information necessary)

Acute/short term exposure

Hazard assessment conclusion:

hazard unknown (no further information necessary)

DNEL related information

Workers - Hazard via dermal route

Systemic effects

Long term exposure

Hazard assessment conclusion:

DNEL (Derived No Effect Level)

Value:

2 mg/kg bw/day

Most sensitive endpoint:

repeated dose toxicity

Acute/short term exposure

Hazard assessment conclusion:

DNEL (Derived No Effect Level)

Value:

4 mg/kg bw/day

Most sensitive endpoint:

repeated dose toxicity

DNEL related information

Local effects

Long term exposure

Hazard assessment conclusion:

medium hazard (no threshold derived)

Most sensitive endpoint:

sensitisation (skin)

Acute/short term exposure

Hazard assessment conclusion:

no DNEL required: short term exposure controlled by conditions for long-term

Workers - Hazard for the eyes

Local effects

Hazard assessment conclusion:

medium hazard (no threshold derived)

Additional information - workers

Aniline - Toxicological Information and DNEL Derivation

Protection against aniline induced erythrotoxicity and its consequences are the basis for DNEL derivation for workers.

1. Erythrotoxicity of aniline and its consequences in animal experiments after repeated exposure

In rats aniline causes toxicity to erythrocytes and the hematopoietic system with corresponding effects on the spleen, bone marrow, liver and kidney. Cyanosis with increased MetHb levels, erythrocyte lesions with formation of Heinz bodies and hemolytic anemia are characteristic observations. Reticulocytosis, increased bone marrow and extramedullary erythropoiesis are compensatory reactions of the red blood cell toxicity. The damaged erythrocytes are scavenged mostly in the spleen. Accumulation of hemosiderin in the spleen and sometimes also in liver and kidney, spleen congestion, dark coloration and increased spleen weight are observed after repeated exposure to aniline. Further signs of toxicity after repeated exposure at relevant doses are splenitis, spleen hyperplasia and fibrosis.

Comparison of the effects in the chronic and subacute feeding studies does not support a strong increase in the magnitude of the erythrotoxicity with prolonged repeated exposure and is in line with a repetitive erythrotoxicity and related effects after repeated exposure. A NOAEL (oral) for erythrotoxicity in rats could not be derived from the available dietary studies; the LOAEL (oral) is 4 mg aniline/kg bw/d.

After repeated exposure of rats to airborne aniline erythrotoxicity and consequential toxicity to the spleen are the leading toxic effects. Minimal histopathological alterations to the spleen have been observed after exposure to the lowest concentration of 64.7 mg/m³ (17 ppm) for 14 days (Du Pont de Nemours and Co., 1981). In another 14-days study on rats a slight increase in splenic extramedullary hematopoiesis was seen at 32.4 mg/m³ (8.4 ppm), whereas at 9.2 mg/m³ (2.4 ppm) no effects on erythrocytes and spleen were observed (Pauluhn, 2004). Due to the lack of erythrotoxicity observed after exposure to airborne aniline for 20 to 26 week, Oberst et al. (1956) observed a general NOAEC of 5 ppm in rats, dogs, mice and guinea pigs. After single exposure to aniline vapor (head-only) an acute reference concentration of 20.6 mg/l/8h (5.4 ppm/8 h) has been derived based on MetHb formation in dogs (Pauluhn, 2005).

In a study of Pauluhn (2002) beagle dogs were exposed head-only to aniline vapor in concentrations to attain a targeted total exposure dose of approximately 15 mg aniline/kg bw. One group of dogs received this calculated dose by gavage. Maximal MetHb levels at the end of the inhalation exposure attained approximately 5%, whereas administration by gavage produced a maximum MetHb response of 26%. This study demonstrates that for aniline, an agent known to be bioactivated by a hepatic first-pass metabolism, the conversion of MetHb formation after oral exposure to inhalation exposure concentrations is subject to overestimating dramatically the magnitude of MetHb formation. As to whether the 5-fold lower potency by inhalation is solely related to the hepatic first-pass bioactivation, to the rate of delivery, or to a less than 100% retention of the inhaled vapor within the respiratory tract remains to be elucidated.  

2. Genotoxicity and carcinogenicity assessment in animal experiments

The wide database on aniline genotoxicity studies and specific investigations point to a low potency and an indirect mechanism of clastogenicity, mediated by erythrotoxicity followed by splenic toxicity and compensatory extramedullary hematopoiesis in the bone marrow of rats. The consequential increased cell turnover and iron cycling in the bone marrow is discussed to encourage chromosomal damage (MAK, 2007). Overall, the genotoxicity of aniline appears very low, if any (SCOEL, 2010).

Aniline caused mesenchymal tumors in the spleen of rats but not in mice after dietary chronic exposure. Tumors were mainly observed in male rats and the tumor incidence was non-linear (CIIT, 1982). Tumor development could be connected to erythrocyte toxicity - indicated by the formation of MetHb and Heinz bodies - and the resulting splenic histopathological changes (fibrosis, fatty metamorphosis, capsulitis, hyperplasia and hematopoiesis). Subacute exposure of male rats to aniline in the diet caused hemolytic anemia, resulting in inflammatory reaction in the spleen and perturbations in iron metabolism (Mellert et al., 2004). Continuous dietary exposure to aniline elicits repetitive erythrotoxicity and related effects on hematopoiesis and spleen in male rats, which could play a decisive role for the development of tumors. Repetitive toxic effects play a decisive role for the development of tumors, and therefore no increased tumor risk should be expected in the absence of an increased erythrocyte turnover (MAK, 2007). This view could experimentally be supported by Mellert et al. (2004) corroborating the contention that experimentally carcinogenic doses of aniline cause early effects on hematological parameters, inflammatory reaction in the spleen and perturbations in iron metabolism as a result of hemolytic anemia. Accordingly, the experimental carcinogenicity of aniline can reasonably be linked to a defined threshold-related process. Aniline is categorized by SCOEL (2010, 2014) and MAK (2007) as carcinogen with a practical threshold.

3. Methemoglobin (MetHb) formation and threshold for adversity in humans

The primary and most sensitive toxicological effect after uptake of aniline is the formation of MetHb. Depending on the concentration of MetHb in human blood clinically relevant effects as blue discoloration of lips, fatigue and exhaustion, consequences of cyanosis, have been reported after aniline poisoning. While MetHb formation is reversible due to action of the enzyme MetHb reductase and is not toxic to erythrocytes per se it is nevertheless a component of potential oxidative injury to red blood cells. Differences in metabolism and sensitivity between species render interspecies transfer of toxicity data difficult. Human data are obviously the data of choice for risk assessment of aniline.

SCOEL (2010) states an increase in MetHb above the normal background level in blood up to 15% will in general be without signs and symptoms while Muller et al. (2006) reported the human body can compensate for a reduction of up to around 20% hemoglobin without any significant clinical effects. Chronic methemoglobinemia at low levels (below 10%) are generally asymptomatic although a blue/gray appearance of the extremities may already occur at MetHb >6% in some animal species including humans (Muller et al., 2006).

In humans normal physiological MetHb levels have been reported to range from 1% (SCOEL, 2010; Muller et al., 2006) to 1.5 % (ACIGH, 2001). The mean MetHb level in the general population is given with 0.78% (95th percentile 1.28%) by Fairbanks (cited in ACIGH, 2001). The latter value was supported in a currently performed study on human volunteers that shows a mean base level Met-Hb saturation of 0.58±0.15% (range: 0.20-1.00) with a 95th percentile of 0.80% (eight non-smokers, 3 day observation with in sum 168 blood samples; Käfferlein et al., 2013, 2015). A variety of intracellular mechanisms are responsible for maintenance of the physiological MetHb level in blood via reduction of MetHb back to Hb. In this regard MetHb reductase is the most important and effective enzymatic system but there are other secondary pathways utilizing NADPH, glutathione (and its related enzyme systems), ascorbic acid, and the general redox potential of the cell (ACGIH, 2001). Estimated half-life in human red blood cells for high levels of MetHb (80-100%) ranges between 6 and 24 hours (Bolyai et al., 1972, cited in ACGIH, 2001).

The MAK Commission notes in its documentations of biological threshold limits (BAT values) that Met-Hb levels >1.5% (the upper limit of the physiological background) must be considered as a biomarker of exposure. However, adverse health effects are not expected in humans at Met-Hb levels up to 5% (BAT 2007; SCOEL 2010; Bolt et al., 1985) in analogy to the maximal occupational exposure limit for CO (MAK 1981, BAT 1994) which was determined under partly consideration of an outstanding susceptible sub-population versus a reduced oxygen transport capacity (persons with latent restricted coronary or arterial function). The limit of 5% Met-Hb is suggested for all chemical compounds capable of inducing cyanosis via the formation of Met-Hb and where cyanosis is considered the primary toxic effect in humans (BAT 2007). Overall, based on practical experiences gained over several decades and investigations on several MetHb inducers and the hemoglobin binding agent carbon monoxide (CO) a MetHb level of up to 5% is generally accepted as tolerable and safe.

4. Potentially vulnerable human subgroups for chemically induced MetHb formation

4.1 Acetylator status

The proportion of aniline which enters the oxidative pathway (formation of N-phenyl hydroxylamine and N-nitrosobenzene) and thus is associated with the formation of MetHb is directly competing with the acetylation of aniline and its urinary excretion. Acetylation of aniline is determined by the activities of N-acetyltransferases (NAT1, NAT 2) which show differences in activities, thus, individuals can be roughly separated in "slow" and "fast" acetylators. Large epidemiological studies show that about 60% and 40% of Caucasians can be assigned to be “slow” and “fast” acetylators (Moore et al. 2011, García-Closas et al. 2005). In a study of Lewalter and Korallus (1985) the influence of the acetylator phenotype on the formation of MetHb was investigated. For aromatic amines a high acetylation rate resulted in relative low levels of hemoglobin conjugates and vice versa. Lewalter and Korallus showed that "slow" acetylators revealed higher mean physiological MetHb levels (1.4%, range 1.0-1.5%) compared to "fast" acetylators (0.9%, 0.7-1,2%). In addition, “slow” acetylators also showed 5-10-fold higher levels of hemoglobin adducts of N-nitrosobenzene compared to “fast” acetylators thus suggesting a small to moderate influence of the acetylator phenotype on Met-Hb levels in blood. However, the acetylation phenotype of the workers remains uncertain in this study because it was deduced only based upon the ratio of N-acetylaniline vs. unconjugated (free) aniline in post-shift urine samples rather than determining the phenotype of the workers in a controlled and standardized manner (Blaszkewicz 2004). In a recently performed study on human volunteers with exposure to aniline vapour at 2 ppm, no differences in Met-Hb formation were observed between slow (15 individuals) and fast (4 individuals) acetylators (Käfferlein et al., 2013, 2015). However, as expected a slightly higher Hb-aniline adduct level was seen in the slow acetylators (mean 6.08 µg/L versus 4.69 µg/L) since only ‘free aniline’ but not acetanilide enters the metabolic pathway which finally leads to the formation of Hb-adducts and this is dependent on the acetylator phenotype (Käfferlein et al., 2013, 2015).

Because of the high number of slow acetylators of > 50% in the population and the potential risk of higher susceptibility when exposed to aniline, the slow acetylator phenotype individuals have to be considered as a relevant vulnerable human subgroup which should be considered in the DNEL derivation for aniline.

4.2 Glucose-6-phosphate dehydrogenase (G6PD) deficiency

A concern has been raised that workers with G6PD deficiency might be more susceptible to anemia when exposed to aniline. This is based on the fact that main role of G6PD is production of the reduced form of NADPH, which has an important role in preventing oxidative damage to proteins and to other molecules, especially in erythrocytes (Luzzatto and Poggi, 2009). G6PD catalyses the breakdown of glucose in the pentose phoshate metabolism by oxidation of G6P to 6-phosphoglucono-δ-lactone, whereby NADP⊕ is reduced to NADPH (MAK, 2010). Individuals with G6PD deficiency are susceptible to developing anemia and hemolysis when challenged by oxidative stress. This can be triggered by infections, by working materials as aromatic amino and nitro compounds, by medication with e.g. sulphonamides or paracetamol, or by specific food such as fava beans (Vicia fava – ‘favism’), which disturb the cellular redox equilibrium (MAK, 2010).

Under controlled work place conditions (compliance with technical measures for occupational health and safety) G6PD deficient workers are not per se more susceptible to aniline as workers without this genetic disorder. However, they bear a higher risk of hemolytic anemia induced by oxidative stress, e.g. via medical treatment of a potentially induced cyanosis after accidentally high aniline exposure. Thus, in the extreme unlikely case of emergency, e.g. aniline-induced cyanosis in terms of Met-Hb levels > 15%, G6PD deficiency carriers cannot be treated with Met-Hb lowering drugs (e.g. methylene blue) because a side effect of these drugs is oxidative stress thus specifically increasing the risk of hemolytic anemia in drug-treated individuals (Liao et al., 2002). For individuals with severe and extremely severe deficiencies (for classification see below and MAK 2010) handling of amino and nitro aromatic compounds is to be avoided from the medical point of view, since accidentally high exposure may lead to extremely severe hemolytic crises with life-threatening consequences (MAK, 2010).

G6PD deficiency is an X-linked recessive hereditary disease (Beutler, 2011) characterized by abnormal levels of G6PD. Because of the type of heredity (e.g. males are more affected than females) and the polymorphism of G6PD (more than 300 allelic variants of G6PD were known) five classes of G6PD activity levels in erythrocytes have been determined by WHO (1989), ranging from severely deficient variants (Class I) to even increased G6PD activity (Class V). Class I variants can be easily detected because of a chronic non-spherocytic hemolytic anemia. Class II variants have less than 10% of residual enzyme activity without anemia. Class III variants are moderately deficient (10-60% residual enzyme activity) and Class IV variants have normal enzyme activities. In practice, because the majority of G6PD deficient persons are mostly asymptomatic, their G6PD deficiency is referred to as mild, simple, or common (corresponding to Class II or III) (Luzzatto and Poggi, 2009).

Although distributed all over the world, G6PD deficiency has prevalence in most tropical and subtropical parts of the world (Luzzatto and Poggi, 2009). African, Middle Eastern and South Asian people are affected the most. A “side effect” of this disease is that it confers protection against malaria, so G6PD deficiency carriers have an evolutionary advantage by increasing their fitness in malarial endemic environments (WHO, 1989). Therefore, the geographic distribution of G6PD deficiency correlates well with the distribution of malaria.

The frequency of G6PD deficiency is usually expressed as the proportion of a sample of males that is found to be hemizygous (WHO, 1989). About 7.5% of the world population carries one or two genes for G6PD deficiency, the proportion ranges from a maximum of 35% in parts of Africa and is rare in Japan and in parts of Europe with 0.1% (WHO, 1989). For some European Mediterranean countries the frequency of hemizygous men is given with 0.5 to 2.9%.

Overall, WHO (1989) has estimated a total frequency of 0.4% G6PD deficient people for the general European population. This number is in line with occupational human health examinations of one industrial aniline producing company for 3 different plants in the years 2010 and 2011. In the total collective of 1021 examined employees only 3 individuals with G6PD deficiency were identified (1/173 in 2011 and 2/848 in 2010), counting for a frequency of 0.3% (Leng, 2012).

The existence and the severity of G6PD deficiency can be determined by clinical and biological laboratory parameters in voluntarily provided blood samples using commercially available, validated test kits. G6PD activity in erythrocytes is measured based on a photometric method (for details see MAK 2010). Following general screening principles (G-Sätze) for preventive occupational medicine of the German Berufsgenossenschaft [Employers’ Liability Insurance Association] pathological G6PD findings are classified as mild, moderate, severe and extremely severe by MAK (2010) as cited here:

‘From the viewpoint of occupational medicine

-mild G-6-PDH deficiency with decreased activity up to 30% are of no clinical relevance,

-moderate G-6-PDH deficiency with erythrocytic activity reductions up to 100% are still tolerable, even when aromatic amino and nitro compounds are handled, if the methaemoglobin value of the affected person in the non-exposed state, which is measured simultaneously, is less than 2% and the reticulocyte counts are in the normal range, and as long as the 4-aminodiphenyl haemoglobin adduct findings in the case of non-smokers remain less than 0.010 µg/L blood and in the case of smokers less than 0.035 µg/L blood, and if there is strict compliance with all the health and safety guidelines of the work area.

In all other cases any handling of amino and nitro aromatic compounds is to be avoided from the medical point of view, as the extent of the hereditary G-6-PDH deficiency, in particular in the case of exposure to aromatic amino and nitro compounds as the result of an accident, may lead to extremely severe haemolytic crises with life-threatening consequences.’

Following this definition only mild and moderate G6PD deficient individuals showing blood parameters within the norm are allowed to work in aniline producing plants under controlled work place conditions.

According to §8 of the German Genetic Diagnosis Act (Gendiagnostikgesetzt, GenDG, 2009 with adaption of 2013) ‘any genetic examination or analysis may only be conducted, and any genetic sample may only be acquired for such a purpose, after the responsible medical person has received the express, written consent of the subject person, both in regard to the respective genetic examination and genetic sample’. Thus, the G6PD analysis test can be offered by the employer but the test can only be performed if the employee agrees.

In the past analysis of G6PD deficiency was already offered to a considerable number of workers in three aniline producing plants (Leng, 2012). Now, aniline producing industry in Europe is going to establish a systematic and comprehensive offer for G6PD analyses including all employees working in aniline producing and handling plants following the method described by MAK (2010). This system will enable the responsible medical persons, under consideration of all legal requirements of agreement and confidentiality, to inform G6PD deficient workers and reach consent in finding individual solutions to ensure safe working conditions.

5. Investigations on aniline induced MetHb formation in humans

5.1. Oral exposure of human volunteers:

The dose-response relationship between oral aniline uptake and MetHb formation was studied in an early study in 20 volunteers who received a bolus dose of 5, 15 or 25 mg/day on three consecutive days (Jenkins et al., 1972). Some volunteers were given higher doses (35, 45 or 65 mg) on subsequent days. The mean maximum increase in MetHb-formation occurred in less than 4 hours. A dose in the region of 15 mg aniline was determined by Jenkins et al. as no-effect dose (NOEL) with regard to MetHb formation. The lowest dose inducing a significant increase in MetHb formation (mean maximal increase of 2.46%) was determined with 25 mg aniline. 35 mg aniline induced a mean maximal increase of 3.68% MetHb. Although the authors determined 25 and 35 mg aniline as lowest effective doses for a slight increase in MetHb nothing is said with regard to adversity of such an increase.

One of the deficiencies of the Jenkins study is that basic blood MetHb levels of the volunteers prior to treatment are not mentioned in the publication. Therefore, the general basic value of about 1% was considered appropriate for assessment by SCOEL (2010) and MAK (2007) and was also taken for the DNEL derivation discussed here. Considering the physiological background of about 1% MetHb the resulting maximal MetHb values after oral uptake of 35 mg aniline were 4.7% (3.68% plus 1%) and therefore below the generally accepted threshold of 5% MetHb. Therefore, a single oral dose of 35 mg aniline was determined by SCOEL (2010) and MAK (2007) as an acceptable starting point for derivation of an OEL because the 5% MetHb level as reference value was not exceeded at that dose. This argumentation is followed also for DNEL derivation.

A further problem with the study by Jenkins et al (1972) is that the volunteers received (a) oral doses of aniline and (b) bolus doses. Pauluhn (2002 and 2005) has demonstrated for beagle dogs (as indicated above a sensitive species for MetHb formation) that MetHb concentrations obtained with exposures via oral route being about 5-fold more potent for inducing MetHb than exposures via inhalation. Hepatic and intestinal first-pass activation of aniline is assumed to be the cause of this discrepancy (Pauluhn 2002). In addition, the rate of dose delivery of the aniline during inhalation seems to be of great importance for the severity of methemoglobinemia (Pauluhn 2005). In beagle dogs an inhalation exposure to aniline of about 21 mg/m3, (averaged for an 8 hour working day), did not induce MetHb formation above the physiological background of about 0.8%.

5.2. Dermal and inhalation exposure of human volunteers:

In an early and poorly reported study on humans (Dutkiewicz and Piotrowski, 1961) uptake of aniline vapor was observed via the lungs and via the skin. A pulmonary retention of nearly 90% was reported for aniline. Based on the urinary p-aminophenol excretion the absorption rate of aniline vapor through the human skin was shown to be approximately 1000 times lower than that of liquid aniline. It ranged from 0.2-0.4 μg/cm2/h. The rate of dermal absorption increased with the temperature and humidity of the air. Overall, the deficient description of the results does not allow definite calculations of dermal absorption.

In a newly performed study to determine the formation of MetHb, human volunteers were exposed to airborne concentrations of aniline under worst-case work place conditions (pilot study: IPA, 2012; main study: Käfferlein, 2013, 2015). All volunteers (4 in the pilot and 19 in the main study), proved to be healthy non-smokers, were exposed to the German official occupational exposure limit of 2 ppm aniline for a complete work shift duration of 8 hours in the pilot study and 6 hours in the main study. This study was approved by the Ethic Commission of the Faculty of Medicine (Ruhr University, Bochum in 2012). The pilot study was carried out with 4 individuals (2 males, 2 females; all slow acetylators) and the main study with 19 individuals (10 males, 9 females; 15 slow acetylators and 3 fast acetylators). The volunteers wore standardized clothes typical for aniline manufacturing industry. To mimic worst-case workplace conditions no skin or respiratory protection measures in terms of gloves or breathing masks were used to allow dermal contact with the aniline airflow. Additionally, the volunteers were exercising 4/3 x 20 min on a cycle ergometer. Blood MetHb concentrations were assessed shortly before the exposure started (0 h), during exposure (2 h, 4 h, 6 h and in the pilot study 8 h) and during a post-exposure examination period (up to 24 and 48 h).

The mean basic MetHb levels before exposure were 0.21% in the pilot study and 0.72% in the main study. An additionally examined group of eight non-exposed volunteers showed physiological MetHb levels of 0.58%. Exposure to aniline at 2 ppm resulted in a slight increase in MetHb levels in all volunteers, reaching a maximal individual level of 2.07%. The mean maximal value of 1.21% MetHb was reached after 6 hours of exposure. Thus, all maximal levels reached were clearly below the threshold limit of adversity of 5% MetHb (BAT, 2007). In line with the observed MetHb-levels no clinical adverse effects in terms of irritative effects on skin, eyes, respiratory tract or cyanosis could be observed in any of the volunteers. A plateau of MetHb was achieved after approximately 6 hours of exposure most likely based on equilibrium between MetHb formation and its degradation via MetHb reductase. MetHb concentrations consistently dropped after end of exposure and after about 24 h the levels were in the range of those observed for non-exposed individuals.

No differences could be observed in the formation of MetHb between men and women or between slow and fast acetylators after exposure to aniline.

In conclusion, the results on controlled exposure to airborne aniline and the formation of Met-Hb at current permissible exposure levels in Germany (2 ppm) show maximal individual increases of Met-Hb up to 2.07% and of aniline in urine up to 418.3 µg/L thus well below the current German guidance levels of Met-Hb (5%, DFG 2007) and the current German threshold limit of aniline in urine (1,000 µg/L, BMAS 2013). Overall, the results directly contribute to the risk assessment of occupational exposure to aniline vapors at maximally tolerable vapor concentrations. The chamber exposure of human volunteers in the experimental setup is taking into account a possible uptake via the skin and/or the respiratory tract, and MetHb as an integrative internal biomarker resulted in concentrations well below toxic levels.

Thus, based on the results of the human inhalation study of IPA (2012) and Käfferlein (2013, 2015) the OEL of 2 ppm (7.7 mg/m³; 8 hour exposure) has been confirmed as no adverse effect concentration (NOAEC) under workplace conditions.

6. DNEL derivation for aniline in workers

Aniline is not used in the public domain thus the only relevant human populations for aniline DNEL derivation are workers. Aniline is handled under strictly controlled conditions and exposure to workers is low.

For many years the data base for aniline induced toxicity in humans was limited to experiences with aniline poisoning and to old and poorly documented studies on human volunteers. The most prominent human study is the one of Jenkins et al. (1972) in which volunteers were exposed to oral aniline bolus doses (for further information see chapter 5.1). However, besides all technical limitations of such an early study, the oral bolus exposure scenario has proven to be not relevant for workplace conditions with airborne aniline exposure. The second prominent study is the one of Dutkiewicz and Piotrowski of 1961, in which human volunteers were exposed to aniline vapor to determine lung and skin absorption (for further information see chapter 5.2). However, the deficient description of study design and results does not allow a definite calculation of dermal absorption. Many studies on inhalation toxicity of airborne aniline have been performed in laboratory animals, mainly rats and dogs; however, species differences exist in aniline metabolism, in MetHb formation and MetHb reduction back to hemoglobin. Therefore, derivations of national and international Occupational Exposure Limits for aniline by e.g. the German MAK committee (2007) and SCOEL (2010) were based on the available human data for the time being and on the general commitment that a MetHb level of up to 5% is accepted as tolerable and safe.

The German MAK committee basically followed in their updated OEL derivation for aniline (MAK, 2007) the approach of 1992 (MAK, 1992). In short, considering an airborne concentration in the height of the actual OEL of 2 ppm aniline (about 8 mg/m³), the standard ventilation volume of 10 m³/working day, and an absorption of aniline via the respiratory tract of 90% and via skin of about 25 % (both values taken from Dutkiewicz and Piotrowski, 1961) MAK calculated a total of 92 mg human aniline exposure for an 8 hour working day. Taking the results of the Jenkins study (Jenkins et al., 1972) for comparison, a single oral dose of 45 mg/person was considered as the lower limit at which healthy persons show no symptoms induced by methemoglobinemia. MAK further took into account the short half-life of aniline in humans of 3.5 hours, recent results on dogs (oral bolus aniline doses were shown to be 5-fold more potent in MetHb formation than exposures via inhalation of the same systemically available dose, Pauluhn, 2002; for further information see chapter 1) and occupational experience that did not show health effects if the OEL is kept, to confirm the OEL of 2 ppm or 7.7 mg/m³. The excursion factor of 2 for peak limits (according to 4 ppm or 15.4 mg/m³) has also been retained.

The EU Scientific Committee on Occupational Exposure Limits (SCOEL, 2010) used the same literature as MAK as basis for their OEL derivation. Like MAK, SCOEL considered an airborne aniline concentration of 2 ppm aniline (about 8 mg/m³), the standard ventilation volume of 10 m³/working day, and an absorption of aniline via the respiratory tract of 90%. However, the dermal absorption rate of aniline from the study of Dutkiewicz and Piotrowski (1961) was interpreted differently by SCOEL. The committee assumed a quantitatively equal absorption via skin and respiratory tract (for further information see chapter 5.2) as worst case, resulting in a total intake of about 140 mg aniline per shift. On the other hand, SCOEL took the oral bolus dose of 35 mg aniline/person of the Jenkins study (Jenkins et al., 1972) as NOAEL, which induced a maximal increase of 3.7% MetHb and, added to the physiological background level of about 1% MetHb, is below the 5% level of tolerability. Consequently, SCOEL proposed an OEL of 0.5 ppm (1.94 mg/m³) and, taking into account the short half-life period of aniline in the human body, a Short-Term Exposure Limit (STEL) of 1 ppm (3.87 mg/m³) was derived.

The SCOEL values have not been transferred yet into the official Consolidated Indicative Occupational Exposure Limit Values (IOELVs) and in the meantime industry announced a new study on airborne aniline exposure of humans under work place conditions. SCOEL agreed to reassess their OEL in view of the experiences gained. The results of this human study are now available (for detailed information see chapter 5.2) and have been considered by SCOEL in September 2014 as addendum (SCOEL Recommendation 2010 and Addendum 2014; SCOEL/SUM/153).

The new study with airborne exposure of human volunteers to aniline under worst-case workplace conditions gives an insight into the time-dependency of potential aniline induced MetHb formation in workers (IPA, 2012; Käfferlein, 2013, 2015; for detailed information see chapter 5.2). The study design represents a worst-case scenario because potentially susceptible individuals were exposed (all volunteers were non-smokers and 19 of 24 were slow acetylators) to the maximal allowed concentration (OEL of 2 ppm or 7.7 mg/m³) for a complete work shift duration (6-8 hours) via inhalation and by allowing dermal contact with the aniline airflow (no skin or respiratory protection measures in terms of gloves or breathing masks were used). Additionally, the volunteers were exercising temporarily on a cycle ergometer to mimic physical activity at work.

Exposure to aniline at 2 ppm resulted in a slight increase in MetHb levels in all volunteers, reaching maximal levels of 2.07%. Thus, all maximal levels reached were clearly below the threshold limit of adversity of 5% MetHb (BAT, 2007). Thus, no clinical adverse health effects in terms of irritative effects on skin, eyes, respiratory tract or cyanosis could be observed in any of the volunteers. A plateau of MetHb was achieved after approximately 6 hours of exposure most likely based on equilibrium between MetHb formation and its degradation via MetHb reductase. Met-Hb concentrations consistently dropped after end of exposure and after about 24 h the levels were in the range of those observed for non-exposed individuals.

In conclusion, the results of this new human volunteer study clearly support the use of the German OEL derived by the MAK committee as DNEL for aniline. Exposure to airborne aniline at 2 ppm (7.7 mg/m³) under worst-case work place conditions for 8 hours is safe, does not induce any clinical signs and the maximal individual MetHb level of 2.07% (mean 1.21% including susceptible persons, i.e. slow acetylators) is well below the general threshold level for adversity of 5% MetHb. The potentially vulnerable subgroup ‘slow acetylators’ is adequately covered by this DNEL since 19 of the 23 volunteers in the human study carried this phenotype. Additionally, the estimated percentage of > 50 % slow acetylators in the human population guarantees that the volunteer groups in the old human studies also must have included this phenotype although not determined in these days.

As described above in detail (see chapter 4.2) aniline producing industry in Europe is going to establish a systematic and comprehensive offer of G6PD analyses for all employees working in aniline handling and producing plants. Thus, workers with a potentially higher risk in the extreme unlikely case of an emergency, e.g. aniline-induced cyanosis in terms of Met-Hb levels > 15%, may be excluded from aniline handling.

The long-term DNEL worker – systemic effects (inhalation) for aniline is therefore established with 2 ppm or 7.7 mg/m³ and confirmed as safe by the new human volunteer study.

Taking into account the results of the new human volunteer study (Käfferlein et al., 2015) SCOEL published a re-assessment of the former recommendation in September 2014 (SCOEL Recommendation, Addendum 2014; SCOEL/SUM/153). It says that the new experimental human exposure study by Käfferlein et al (2015) supersedes the old human data on aniline exposure, which was the the basis of the former recommendation of SCOEL. ‘According to the new study, an exposure to 2 ppm aniline for 8 hours will not lead to elevation of methaemoglobin beyond critical levels, even if a moderate workload is considered’. Therefore, SCOEL recommends an airborne level of 2 ppm aniline as OEL (8h-TWA).

- Since reliable data for body burdens after inhalative and dermal exposure via airborne aniline are not yet available, the dermal DNEL for aniline is calculated conservatively, based on the results of the old oral and dermal human studies published. The no observed adverse effect level (NOAEL) in humans of 35 mg aniline/person corresponds to an inhalation DNEL long-term for systemic effects of 7.7 mg/m³ (2 ppm) aniline vapor over an 8 hour shift. Using the default body weight of 70 kg for workers, the NOAEL in humans of the internal dose of 35 mg aniline/person and the dermal absorption considered by MAK of ca. 25 % a dermal DNEL long-term of 2 mg aniline/kg body weight can be calculated.

The long-term DNEL worker – systemic effects (dermal) for aniline is therefore established with 2 mg aniline/kg bw.

- The establishment of an acute toxicity DNEL is considered unnecessary as the long-term DNEL is derived on the toxicity to erythrocytes in humans after oral bolus exposure to aniline and on the new human volunteer study with airborne aniline (Käfferlein, 2013, 2015). However, the German MAK committee (MAK, 2007) has been agreed on that an excursion factor of two, leading to a DNEL for acute inhalation of 15.4 mg/m³ (4 ppm) and a DNEL for acute dermal exposure of 4 mg/kg bw, would still provide adequate protection against elevated MetHb levels.

The acute DNEL worker – systemic effects (inhalation) for aniline is therefore established with 4 ppm or 15.4 mg/m³.

The acute DNEL worker – systemic effects (dermal) for aniline is therefore established with 4 mg aniline/kg bw.

General Population - Hazard via inhalation route

Systemic effects

Acute/short term exposure

DNEL related information

Local effects

Acute/short term exposure

DNEL related information

General Population - Hazard via dermal route

Systemic effects

Acute/short term exposure

DNEL related information

General Population - Hazard via oral route

Systemic effects

Acute/short term exposure

DNEL related information

Additional information - General Population

According to the ECHA 'Guidance on information requirements and chemical safety assessment', Chapter R8.7.3 DNELs for the general population must be set if the substance is present in consumer-available products and is released to the environment and present as an environmental contaminant.Aniline is not used in the public domain thus the only relevant human population for aniline DNEL derivation is the worker. Investigation of the effects of aniline in the general population is thus not included in the scope of this risk assessment.