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Please be aware that this old REACH registration data factsheet is no longer maintained; it remains frozen as of 19th May 2023.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

Key value for chemical safety assessment

Skin sensitisation

Endpoint conclusion
Endpoint conclusion:
adverse effect observed (sensitising)
Additional information:

The test substance is covered by the category approach of methylenediphenyl diisocyanates (MDI). Hence, data of the category substances can be used to cover this endpoint. The read-across category justification document is attached in IUCLID section 13. It is important to note that the MDI category approach for read-across of environmental and human hazards between the MDI substances belonging to the MDI category is work in progress under REACH. Therefore the document should be considered a draft.

Some animal data indicate that MDI is a strong skin sensitiser, however the observations in animal studies are not fully consistent with human workplace experience, which gives limited findings of dermal sensitisation in workers and in case reports. This discordance is particularly evident for the mouse LLNA that indicates MDI is a strong sensitiser while occupational experience shows that MDI and other diisocyanates are relatively weak inducers of allergic contact dermatitis (Schlede et al., 2003). The LLNA is an induction assay determining lymphocyte proliferative responses induced in the regional lymph nodes of mice exposed topically to the test substance. Irritation can likewise contribute to a mild unspecific stimulation of lymphocyte proliferation (McGarry, 2007)1.

Nevertheless, it is generally agreed that MDI is a skin sensitiser.


The study marked as key study under REACH was performed according to the Buehler method in guinea pigs. This study demonstrated that at epicutaneous occlusive induction doses of 0.25%, 0.75%, and 2.5%, some animals exhibited marginal responses to high challenge concentrations of the test material (challenge concentrations: 0.00075%, 0.0075%, 0.075%, 0.25%, and 0.75%). However, only one definitive response (score of 1) was seen at a low induction dose to the mid-high challenge concentration. Based on the lack of dose-response and the overall marginal and inconsistent results the no-effect level of this study was defined as 100 mM (2.5%) (Davis et al., 1984), which can be seen as threshold under the test conditions used. Studies marked as supporting studies under REACH, two mouse local lymph node assays (LLNA) demonstrated a sensitisation potential of MDI. In the LLNA, exposure to MDI resulted in a stimulation index of 3.45 at a dose of 0.03% MDI (Hilton et al., 1995; Dearman et al., 1992).


The human data (together with the animal data) for skin sensitisation were summarized and evaluated by Schlede et al. (2003). For MDI this joint and systematic approach for potency ranking indicated that in relation to a relatively high exposure, a relatively low but substantial incidence of allergic contact dermatitis exists. MDI has been in global, large scale production for over twenty years. While good hygiene practice has improved over the years there has been ample opportunity for human exposure through poor practice and accidents. Nevertheless, reports of skin sensitisation in humans are very rare. Although comprehensive, reliable and good quality evidence from human cases or epidemiological studies exists, the data are not generated in controlled volunteer experiments. So, whilst the frequency of reported cases may, at least partly, reflect factors such as the exposure situation and non-reporting, it is clear from long- term human workplace experience that MDI does not exhibit marked skin sensitising activity. Therefore in consideration of the available human data the results of the animal studies, namely the LLNA, seems to be over-predictive for the endpoint skin sensitisation.



Additional literature not reported as endpoint study record in the IUCLID dataset:

(1) McGarry, H. F. (2007). The murine local lymph node assay: Regulatory and potency considerations under REACH. Toxicology 238:71-89.

Migrated from Short description of key information:

MDI is a skin sensitiser based on animal and human data. In consideration of human case reports and epidemiological studies the skin sensitising potential of MDI can be regarded as not strong.

Justification for selection of skin sensitisation endpoint:

Study with quality assurance and reliability 2, conducted according to the Buehler Method in guinea pigs

Respiratory sensitisation

Endpoint conclusion
Endpoint conclusion:
adverse effect observed (sensitising)
Additional information:

The test substance is covered by the category approach of methylenediphenyl diisocyanates (MDI). Hence, data of the category substances can be used to cover this endpoint. The read-across category justification document is attached in IUCLID section 13. It is important to note that the MDI category approach for read-across of environmental and human hazards between the MDI substances belonging to the MDI category is work in progress under REACH. Therefore the document should be considered a draft.

The Brown Norway (BN) rat model for respiratory sensitisation has used both dermal and inhalation routes for induction. The elicitation phase was typically four inhalation exposures, spaced at approximately 15-day intervals. After the last challenge the measurements made included: respiratory responses using whole body plethysmography, analysis of inflammatory endpoints using lung lavage fluid and IgE levels in lung lavage fluid or serum. Lung and lymph node weights were measured post mortem and the respiratory tract was used for histopathological investigations. Controls did not receive the induction treatment but were challenged by inhalation as for the test groups, to control for short term irritation effects. On occasion respiratory methacholine challenge was performed the day following MDI challenge to test for non-specific respiratory hyperreactivity. The test material in these studies was polymeric MDI.

Both inhalation and dermal routes of induction were investigated in a study in BN rats, with challenge by 4 x 30 min exposures to 15.7 mg/m3 MDI spaced at 2-week intervals (Pauluhn, Woolhiser and Bloemen, 2005). By topical induction (150μl neat MDI on day- 0 and 75 µl neat MDI on day 7) and inhalation challenge there were signs of progressive pulmonary inflammation in the bronchoalveolar lavage fluid (BALF), and increased weights of lungs and lymph nodes. However inhalation induction at 25-30 mg/m3 (5 x 3 hours/day) led to a respiratory response on the initial challenge which was not seen subsequently and was assessed as non-specific pulmonary reaction. It was concluded that an allergic respiratory response was seen with topical induction and respiratory challenge. Subsequent studies used higher challenge concentrations to elicit a greater response.

The dose-response for the topical induction by MDI was investigated using induction applications on day 0 and 7 in BN rats (Pauluhn, 2008). Doses of neat MDI were applied to shaved flank skin by means of applying the MDI as spots on aluminium foil. Spot size was controlled and the amount (volume) varied. The foil was pressed to the skin and removed, and the applied dose checked by weighing the foil. Inhalation challenges were made at days20, 35, 50 and 65, using 38mg MDI/m3 for 30 minutes. Groups of eight rats were exposed to the induction doses 2.5, 10 and 40μl/rat. Each dose group consisted of three subgroups with dosed surface areas of 3.1-12.6 cm2, 0.8-3.1 cm2, and 0.4-0.8 cm2, respectively. Controls included animals neither induced nor challenged, and animals not induced but challenged four times. Results from BALF analysis and from breathing pattern analysis indicated an asthmagenic response. Despite the large range in induction doses and area-doses, the elicited responses were essentially of the same magnitude across the groups. Most changes were statistically significantly different compared to the non-induced but four times challenged group. The induced group challenged once only after three months displayed a borderline asthmagenic response. In summary, the model successfully identified MDI as an asthmagen by cellular and respiratory responses using moderate induction doses and repeated, mildly irritating challenges. However dose and area-dose are interrelated and it was not been possible to establish the contribution of each in the induction process.

The dose-response for the inhalation induction by MDI was investigated using inhalation induction at 3 different C x T values in BN rats, either at higher-concentration-shorter-duration or lower-concentration-longer-duration (Pauluhn and Poole, 2011). Challenge at 41 mg/m3 elicited both cellular and respiratory responses. A no-effect level for induction was determined at about 5000 mg/m3.mins for both the induction regimens (i.e. 5 x 10-min/day to 97.1mg/m3 and 5 x 360-min/day to 2.9 mg/m3), although the higher exposure concentration seemed more effective. The dose-response for the inhalation elicitation by MDI was investigated using a high induction inhalation dose (968 mg/m3 for 10 mins on 5 consecutive days) and three elicitation challenges (of 40 mg/m3 for 30 mins) at two week intervals followed by a final elicitation challenge at either 5.1 or 14.5 or 39.8 mg/m3 . The lowest challenge concentration did not elicit a response, thus defining a no-effect level for the elicitation of 5 mg/m3. Higher challenges resulted in respiratory responses and an increase in polymorphonuclear neutrophils (PMNs).

In addition to the studies in BN rats, data are available in guinea pigs:

In another key study, groups of guinea pigs were exposed to MDI by intradermal injection, topical application, or inhalation exposure for induction and by inhalation to 25-44 mg/m3 MDI for the challenge exposure. Attempts to sensitize guinea pigs by inhalation exposure to MDI were unsuccessful; no animals exhibited pulmonary responses following challenge with MDI. Intradermal injection or topical application of MDI induced specific antibody responses and pulmonary responses in 12-65% of guinea pigs depending upon induction dose and route. The differences in immunogenicity observed clearly reflect variation in exposure route rather than the concentration of MDI used for sensitization (Rattray et al., 1994a,b).

In a study, using intradermal induction, 12.5% of the MDI sensitized guinea pigs showed a marked non-specific pulmonary reaction after an inhalation challenge of 35 mg/m3 MDI which was not observed in sham or vehicle controls. An association between increased airway responsiveness, increased airway eosinophilia, and increased levels of IgG1 anti-MDI antibody titers could not be established (Pauluhn, 1994).

In a study in guinea pigs using three intradermal and a single, nose only, 15 minute inhalation induction exposure followed on day 21 by a 30 min, nose only inhalation challenge to 3.4-60 mg/m3 MDI or 35mg/m3 MDI-guinea pig serum albumin conjugate (MDI-GPSA), only MDI concentrations above the irritant threshold evoked a pronounced respiratory response. An increase in non-specific airway hyper-responsiveness was observed during acetylcholine-challenge in animals challenged previously with 60 mg/m³ MDI. Conclusive delayed onset respiratory responses were neither observed following the MDI nor the MDI-GPSA conjugate challenge. Respiratory responses were only provoked in animals challenged with overtly irritant MDI-concentrations (Pauluhn and Mohr, 1994). Another guinea pig study used an inhalation induction to 132 mg/m3 MDI, nose only for 15 minutes and an inhalation challenge to 3, 15, and 35 mg/m3 MDI for 20 min/concentration for a total of 60 minutes on day 21. Only a borderline response was seen with this protocol. Mild MDI-specific immediate-onset responses were observed mainly during challenge to slightly irritant concentrations (35 mg/m³). A marked increase of neutrophilic or eosinophilic granulocytes could not be established. Animals sensitized to high concentrations of aerosolized MDI showed a mild airway hypersensitivity without concomitant influx of inflammatory cells (Pauluhn, 1995). An inter-laboratory study for the evaluation and validation of an animal model for low molecular weight chemicals to exhibit respiratory allergy in guinea pigs used intradermal induction of 0.0003 to 1% and an inhalation challenge on day 22 to 18-55 mg/m3 MDI. The results demonstrated that the intradermal injection of MDI was able to induce high titre antigen-specific antibodies in guinea-pigs. Inhalation exposure to MDI, approx. 3 weeks after sensitisation with MDI, induced a pulmonary response characterized by changes in respiratory rate (Blaikie et al., 1995).

In summary, although there are no guidelines available for respiratory sensitization studies in animals, several researchers have shown respiratory changes and/or antibody production in animals after induction exposure and subsequent challenge with MDI. The studies showed some type of respiratory response (alterations in respiratory rate, non-specific hyperreactivity, influx of inflammatory cells), however differences in immunogenicity observed clearly reflect variation in induction exposure route. Attempts to sensitize guinea pigs by single inhalation exposure only to MDI showed borderline responses at best as compared with potent respiratory sensitizers such as ova albumin and trimellitic anhydride. Respiratory responses were only provoked in animals challenged with overtly irritant MDI-concentrations.

The BN rat MDI respiratory sensitisation studies demonstrate the existence of thresholds for the induction of respiratory hypersensitivity by both the dermal and inhalation routes, and a threshold for elicitation of respiratory sensitization following induction and subsequent multiple challenges. The derived elicitation threshold C×t appears to be plausible relative to human evidence. The close association of C×t products triggering an elicitation response in asthmatic rats with the acute pulmonary irritation threshold C×t is intriguing and supports the view that for this class of chemicals portal of entry related allergic responses appear to be linked with pulmonary and/ or lower airway irritation. Accordingly, high concentrations delivered to the respiratory tract during short exposure periods appear to bear a higher sensitizing potency than equal C×t products during longer exposure periods (Pauluhn and Poole, 2011). Pauluhn (2008) examined if the results from this animal model could be applied to the human workplace experience and concluded that “Taking into account the current occupational exposure level of MDI (0.05 mg/m3), the conclusions derived from this bioassay of respiratory irritation and allergy are not at variance with existing human evidence and current occupational workplace standards.

An extensive literature study was performed to evaluate dose-response relationships and related no-effect levels for sensitisation and elicitation in skin and respiratory allergy (Arts et al., 2006). A more recent literature study with a specific focus on the sensitisation of the respiratory tract confirmed the existence of such thresholds for chemical respiratory allergy (Cochrane et al., 2015)1.

With respect to the respiratory tract, dose-response relationships and no-effect levels for induction were found in several human as well as animal studies. Reported results suggest that relatively high peak concentrations can induce sensitisation, and that prevention of such concentrations will prevent workers from developing respiratory allergy.

There are clinical case reports of occupational diisocyanate asthma after initial exposure to presumably high concentrations of MDI. Occupational challenge tests or specific inhalation challenges have demonstrated asthmatic responses to low levels of MDI in sensitized individuals. Although antibody testing is appealing as a diagnostic tool, unlike high molecular weight agents, these serologic markers are insensitive and non-specific for disease detection (Ott et al., 2007).

An epidemiological study reported a reduction of absolute number of diisocyanate asthma cases more recently, compared to previous data in Canada (30 ISO claims/yr during 1980 to 1993 as compared to 7.4 ISO claims/yr in 1998 to 2002) (Buyantseva et al., 2011, cfr study record 7.10.2e).

A more recent study in France has shown a significant decrease of work related asthma over the period 2001–2009 for cases related with isocyanate exposures (Paris et al., 2012, cfr study record 7.10.2f). In addition, a majority of diisocyanate asthma cases reported at least an improvement in respiratory symptoms at their last assessment; and 46% reported clearing of all of their symptoms (Buyantseva et al., 2011). The isocyanate-related asthma incidence in Europe was reviewed by Poole (2013), who concluded that there was some indication of a downward trend in isocyanate associated occupational asthma in some countries, (i.e. France and Belgium), the number of cases in other countries remain steady, or possibly rising (Czech Republic). While tests are available to identify specific agents responsible for inducing the hypersensitive state, such testing is not applied in the majority of cases, with diagnosis relying on clinical examination and labour anamnesis. There is some indication that isocyanate occupational asthma might be associated with certain occupations/jobs using spray applications, but information is sparse. A cross-sectional study was performed with 243 employees exposed to MDI in a polyurethane processing facility. The 8h time weighted average exposures did not exceed 5ppb, and all three cases diagnosed for work-related asthma appeared to have been induced as a result of intermittent high exposures during non-routine work activities (Bernstein et al., 1993). Overall, various recent data on isocyanate related asthma incidence indicates a reduction in cases in the last decade. Where controls and current exposure standards are met, new asthma cases can be minimized.


Additional literature not reported as endpoint study record in the IUCLID dataset:

(1)  Cochrane S. A. et al.(2015). Thresholds in chemical respiratory sensitization, Toxicology 333:

Migrated from Short description of key information:

MDI is a respiratory sensitiser based on animal and human data. The sensitisation of the respiratory tract can be induced by the dermal or inhalation route. Both induction and elicitation of respiratory sensitisation are clearly threshold dependent. Respiratory sensitisation requires recurrent irritant inhalation exposures to induce asthma. Thus, protection from irritation protects also from sensitising the respiratory tract.

Justification for selection of respiratory sensitisation endpoint:

GLP compliant study with reliability 2, conducted according to the OECD-GD 39 (inhalation exposure methodology).

Justification for classification or non-classification

Skin sensitisation

According to Regulation (EC) No 1272/2008 (CLP) MDI is classified as skin sensitiser with Category 1 (H317: May cause an allergic skin reaction).


In principle and depending on the available data taking into account animal studies as well as human cases or epidemiological studies a reliable assessment of the skin sensitisation potency with allocation to a sub-category is possible according to Regulation (EU) No 286/2011 ( Where data are sufficient a refined evaluation according to section allows the allocation of skin sensitisers into sub-category 1A, strong sensitisers, or sub-category 1B for other skin sensitisers).


For MDI, the joint assessment of all relevant animal assays on skin sensitisation does not result in a consistent and scientifically reliable allocation to one of the two sub-categories 1A or 1B. A skin sensitising potential was identified in animal assays, but apparently the resulting sub-category is dependent on the study type and/or study conduct. The strict application of the criteria for sub-categorisation for a LLNA may lead to an over-estimation of the sensitisation potency for substances with irritant properties. Although the LLNA is a reliable predictor of sensitisation induction, the confounding role of non-specific irritation in local lymph node proliferation limits its use for potency categorisation of irritant sensitisers. A similar conclusion was published by the Interagency Coordinating Committee on the Validation of Alternative Methods in its Test Method Evaluation Report (ICCVAM, 2010)1…LLNA cannot be considered a stand-alone assay to determine skin sensitisation potency categories. … Among the 21 substances that produced a LLNA EC3 ≤ 2 %, 67 % (14/21) were correctly identified as strong sensitizers, but 33 % (7/21) were incorrectly over-classified as strong skin sensitizers based on available human test data.


Human experience, e.g. a review of skin sensitisation potency by the Federal Institute of Risk Assessment in Germany (BfR; Schlede et al., 2003) came to the conclusion, that MDI, TDI, HDI, H12MDI and IPDI would fall into Category B ("Solid-based indication for contact allergenic effects because of: (1) less frequently proven contact allergenic effects in humans taking into account existing positive animal data..."), which fits best to sub-category 1B compared to the classification criteria of Regulation (EU) No 286/2011 ( Human evidence for sub-category 1B can include: (b) diagnostic patch test data where there is a relatively low but substantial incidence of reactions in a defined population in relation to relatively high exposure).


In conclusion, while animal and human data indicate MDI is a skin sensitiser, the animal data are equivocal on sub-categorization. In consideration of the available human data the LLNA therefore seems to be over-predictive for the endpoint skin sensitisation with diisocyanates. However, the human data clearly indicate that MDI is not a strong skin sensitiser and in accordance with Regulation (EU) No 286/2011 MDI shall be classified as Category 1B skin sensitiser.



Respiratory sensitisation


According to Regulation (EC) No 1272/2008 (CLP) MDI is classified as respiratory sensitiser with Category 1 (H334: May cause allergy or asthma symptoms or breathing difficulties if inhaled).


Potency categorization (sub-categorization) for respiratory sensitisation according to the GHS criteria is based on information on frequency of occurrence in humans. No clear guidance is given to distinguish Category 1A from 1B. MDI respiratory sensitisation is well known. Compared to the large number of potentially exposed individuals, the incidence of reported cases is very low, as e.g. indicated by a recent review of available publications and national or state/province-based registries, and it can be anticipated that the low frequency/incidence of cases may, at least partly, reflect factors such as the well-controlled exposure situation. In absence of clear guidance it has to be concluded, that the currently available data on respiratory sensitisation of MDI are not sufficient for sub-categorization.

Therefore, according to Regulation (EU) No 286/2011 ( Respiratory sensitisers shall be classified in Category 1 where data are not sufficient for sub-categorization) MDI should be currently remain in Category 1, without further sub-categorization.


Additional literature not reported as endpoint study record in the IUCLID dataset:

(1) ICCVAM (2010). ICCVAM Test method evaluation report on using the LLNA for testing pesticide
     formulations, metals, substances in aqueous solutions, and other products.NIH Publication No.
     10-7512,National Institute of Environmental Health Sciences, Research Triangle Park, NC.

     Available from:

    < >