We read with interest the recently published article "Measurements of Airborne Methylene Diphenyl Diisocyanate (MDI) Concentration in the U.S.Workplace"(1) that addresses an important and widespread workplace exposure. MDI, essential to the production of numerous polyurethane foams and other products, is a potent sensitizer that can cause isocyanate asthma. This article presents a unique exposure data set, including extensive personal and area samples (over 8,000) collected from over 300 different MDI production facilities. However, we have concerns related to three issues: (1) the sampling and analytical methods used to measureMDI airborne exposures, (2) the presentation and interpretation of the airborne MDI data, and (3) the underlying premise that maintaining airborne isocyanate levels below monomer occupational exposure limits ensures a safe workplace, with little mention of skin as a potential route of exposure and sensitization. (1) Sampling and Analytical Methods: As the authors note, several technical factors could have contributed to the underestimation of MDI airborne exposures in these manufacturing facilities. Importantly, the majority of samples (>90%) were collected using filters, which are more likely to underestimate MDI exposure than impingers, especially for spray applications.(2-3) For example, Lesage et al.,(4) reported that the filter-based Iso-Check method measured on average 25% (median 20%) of the impinger values. Additionally, measurement of only monomeric MDI and not also oligomers of MDI would have resulted in an underestimation of total biologically active NCO exposures in some environments.(5) In contrast to the statements in the article about the unavailability of methods and standards to measure oligomers of MDI, methods existed as far back as 1987 that do not require analytical standards for measuring isocyanate oligomers.(6) Although there continue to be limitations to the measurement of oligomeric species in MDI-based products, the contribution of such species to the exposure hazard should not be overlooked, particularly when their contribution to the total isocyanate is substantial relative to that of the MDI monomer. We also question the authors' statement that "there is no objective definition of what is meant by 'total isocyanate' by those methods claiming to analyze for total NCO functional groups." Total free isocyanate group is routinely measured in bulk products by the titration with di-n-butylamine.(7) When using this procedure, species that react with the amine are effectively defined as free isocyanates. Air sampling methods for total isocyanate rely on the same operational definition, although all species thus derivatized may not prove to be measurable using chromatographic methods. Finally, other factors could have contributed to an underestimation of MDI exposures, including the type of derivatizing agent and the use of older, less sensitive quantitative laboratory methods.(2) While the authors point out issues with the different sampling and analytical methods used, they do not discuss or estimate the impact of the analytical methods on their own data. (2) Presentation and Interpretation of the Airborne MDI Data: Despite the rich data set of over 8,000 samples collected from 1984 to 1999, the authors provide relatively limited information regarding sampling strategies, task-specific exposure levels, or potential changes in the manufacturing operations or products over time. Further analysis of the data, including potential exposure determinants, may have yielded additional valuable information. The authors conclude, "workplace airborne MDI concentrations are extremely low in a majority of the manufacturing operations." However, without additional information, it is difficult to know how relevant these exposure data are to the many different current MDI end-user settings or what work conditions or factors produce higher exposure levels. The authors highlight spray and heating operations as having higher exposure potential. It would have been helpful if the authors compared their exposure data to other published data, such as spray-on applications of truck bed liners and rigid foams for insulation,(4,8,9) which have documented notably high MDI airborne exposures. The authors also do not discuss their unexpected finding that over 30% of area and personal samples taken in an area where cured rigid foam was "cut to length, stacked, and made ready for shipment" were above the TLV for MDI. (3) Occupational Exposure Limits: The MDI exposure data are presented in comparison to the OSHA PEL and ACGIH R TLV R for MDI monomer, rather than using a total NCO metric, and without including a reference to the lower occupational exposure limits (OELs) used by a number of countries, including the U.K., Australia, and Sweden.(5) Comparison of airborne MDI data to MDI OELs perpetuates an outdated premise that maintaining airborne levels below thesemonomer mass standards will prevent isocyanate asthma. Isocyanate sensitization and/or asthma has occurred where measured airborne MDI monomer levels are below OELs or below the limits of detection with the methods used, or where similarMDI levels would be expected, butMDI airmonitoring data are not available.(10-13) Potential MDI skin exposure is frequently noted in such cases but has been difficult to quantify due to limited isocyanate skin sampling and analytical methodologies.(11-14) Although the authors recommend gloves "where dermal contact to MDIor PMDIis likely to occur," they largely ignore the important role that MDI skin exposure likely plays in inducing systemic sensitization that can develop into asthma following subsequent respiratory tract exposure.(14-16) Thus, for the reasons noted, we are concerned that the data from this publication may underestimate the risk to workers and industries that use MDI, and that the authors do not adequately highlight the importance of targeting prevention efforts at both skin and airborne exposures. Hopefully, future collaborative efforts among industry, academia, and government will lead to a greater understanding of MDI exposure determinants and risks and more effective strategies to minimize such exposures.
Carrie A. Redlich, Yale University, School of Medicine, Deptartment of Medicine Yale Occupational & Environmental Medical Program, New Haven, CT 06510