Isocyanate exposure and hypersensitivity pneumonitis--report of a probable case and prevalence of specific immunoglobulin G antibodies among exposed individuals.

Isocyanate exposure and hypersensitivity pneumonitis - report of a probable case and prevalence of specific immunoglobulin G antibodiesamong exposed individuals. Scand J Work Environ Health 1989;15:234-237. A car painter experienced three episodes of a hypersensitivity pneumonitis-likediseaseafter exposure to two-component acrylic lacquers with hexamethylene diisocyanate (HDl) as the curing agent. High titers of HDl-specific immunoglobulin (Ig) G antibodies were found in the patient's serum by means of enzyme-linked immunosorbent assay (ELISA). In the ELISA, 5 to 10 010 of the sera from 455 isocyanate-exposed but asymptomatic workers were positive, depending on the criterion used for a positive test, whereas 0 Ufo of the sera from 157 unexposed referents was found to be positive. Among 10 subjects with isocyanate-induced asthma and isocyanate-specific IgE antibod ies, 50 % had specific IgG. It was concluded that the presence of isocyanate-specific IgG antibodies in serum is correlated with isocyanate exposure rather than with symptoms of isocyanate-induced disease.

However, the diagnosti c significance of isocyanatespecific IgG antibodies has not been systematicall y evaluated . An investigation of the prevalence of such antibodies in groups of isocyanate-exposed worker s, as well as in unexposed referents, was therefore included in this stud y.

Materials and methods
Case report. A 21-year-old man started to work as a car painter in February 1984. He was unskilled and consequently assigned to preparatory work (grinding , filling, and masking), but occasionally he spray-painted minor car detail s. He had no famil y histor y of atopy, but he had experienced allergic reactions in 1971-1972, ind uding one episode of urticaria of unknown origin . He had smoked 10-15 cigarettes a day since the age of 15 years.
On 23 May 1984 he performed his first major painting job, spraying a racing boat for about 1.5 h. Against regulations, this work was done outsid e the ventilated painting box. Within an hour of completin g the work, he had chills, dyspnea and chest pa in, followed by general malaise, fever , sweating , headache, and nonrotatory vertigo. He decided to seek medical attention at the local hospital, and upon admission tran sient pulmonary ronchi were noted. The patient 's pulse rate was elevated (108 beats/min) concordantly with his elevated body temperature (38.6°C). His blood pressure was 115/70 mm Hg (15.3/9.3 kPa) . He had a slight leucocytosis (10.5 .10 9 /1; normal 4-9.10 9 /1), but no differential count was obtained. Routine blood and urine tests were normal, ind uding an erythrocyte sedimentation rat e of 3 mm . Unfortunately, no pulmonar y radiography, blood gas analysis, or lung function test was performed in the acute phase of the disease. The following morning, the patient' s bod y temperature had dropped to 37.5°C. At this time , a pulmonary radiograph was normal , and the pa tient was discharged. He returned to work but did not paint until one day in August, when he spr ayed the interior of a small car. The same evening he experien ced a second episode of fever, chills, malaise , dyspnea, and chest pain. He felt alright again the next morning. A third and final episod e occurred early in November when he had been wat ching a workm ate spray painting for less than a minute.
On all three occasions, he had been exposed to a twocomponent acrylic lacquer with a curing agent containing 42 0J0 (by weight) prepolymerized HOI, includin g Absorba nce units 0 . 8 MOl (   Figure 1. Results of the ELISA (unbroken lines) when the patient's serum and con jugates prepared from the isocyanates hexamethylene diisocyanate (HOI), diphenylmethane diisocyanate (MOl), and toluene diisocyanate (TOI)were used and a typical result (broken line) when sera from unexposed referents were analyzed. The results for the referents were the same for all three lsocyanates.
had been washed as before, 100~I of an orthophenylenediamine solution in 0.1 mollI citrate buffer, pH 5.0, was added to each well. Five minutes later, 50~I of a 2.0 molll sulfur ic acid solution was added in order to stop the color reaction, and the absorbances were read at 450 nm by means of an MR 700 microplate reader (Dynatech Laboratories Inc) . Prevalence of immunoglobin antibodies. The prevalence of isocyanate-specific IgG antibodies among 455 isocyanate-exposed workers (Swedish polyurethane workers and spray painters), 157 unexposed referents (industrial workers), and 10 workers with isocyanateinduced asthma and isocyanate-specific IgE antibodies (16) was determined according to two different criteria for a positi ve ELISA. Criterion I required at least 0.5 ab sorbance units when serum was diluted 20 times and at least 50 0J0 inhibition at that dilution when 20 ug of an isocyanate conjugate was added to I ml of the dilut ed test serum prior to the analysis. Criterion II was defined as a positive test according to the first criterion plus at least 0.5 absorbance units when serum was diluted 100 times.

Results
No IgE antibodies against HOI, MOI, or TOI were detected in the patient's serum, and the total IgE concentration was 40 kU /1 (normal). No precipitating antibodies could be observed when the serum was used in an immunodiffusion test.
On the other hand, the serum from the patient gave a strong positive reaction to HOI in the ELISA. Positive, but weaker, reactions were also obtained with MOI and TOI (figure I). Approximately 75 0J0 of the reaction could be inhibited by the addition of 20 ug of isocyanate conjugate to I ml of the test serum prior Immunologic studies . The serological analyses started with HOI, MOI, and TOI conjugated to human serum albumin (HSA) and serum from the patient from December 1984. The conjugates had previously been optimized for the radioallergosorbent test (RAST) system with sera from isocyanate-exposed workers who had developed bronchial asthma, as well as isocyanatespecific IgE antibodies (13) . The number of isocyanate molecules per carrier molecule for the MDI-HSA, HOI-HSA, and TOI-HSA conjugate had been estimated to be 6,8, and 10, respectively. The RAST was performed according to standard procedures. The serum was also used in an immunodiffusion test.
The presence of specific IgG antibodies was investigated by means of the enzyme-linked immunosorbent assay (ELISA), employing the same conjugates as antigens as were used in the RAST. Briefly, the isocyanate-HSA conjugates were diluted in 3 mmolll phosphate-buffered saline (PBS) solution, pH 7.2, to a final concentration of 0.05 mg/rnl. The solution was added in 100~I amounts to a polystyrene microplate (Nunc-Immuno plate I, A/S Nunc, Roskilde, Denmark), and the plate was then incubated for at least 2 h at 4°C in a humidified chamber. After incubation, the plate was washed three times with 3 mmolll PBS with the use of a Dynatech Autowash 2000 washer/aspirator (Dynatech Laboratories Inc , Laboratory Design AB, Lidingo, Sweden). Sera were diluted with PBS containing 0.02 0J0 Tween ?'. Diluted sera were added in 100~I amounts to duplicate wells. The plate was then incubated for 1.5 h at room temperature in the humidified chamber , after which it was washed with 3 mmolll PBS containing 0.02 070 Tween, and 100~I of peroxidase-conjugated rabbit immunoglobulins to human IgG (Dako-Immunoglobulins A/S, Copenhagen, Denmark), diluted I : 500 with PBS containing 0.02 0J0 Tween, was then added to the plate, and the plate was incubated for 2 h . After the plate a maximum of 0.4 % HDI monomer. He had been wearing a half-ma sk respirator with a charcoal filter part of the time, but he admitted that it had not been properly adjusted . On the third occasion, he used no respiratory protection at all. In addition, the exhaust ventilation had been clearly insufficient. Consequently, vapors and possibly microdroplets of the paint might have been inhaled.
The patient connected his outbreaks of illness with his work, and reported this observation to the Social Insurance Office after the second episode. When he was fir st seen by one of us (AS), in December 1984, he had been on sick-leave for one month, and a routine physical examination was normal. A spirometric recording (Vitalograph'P) showed a vital capacity (VC) and forced expiratory volume in I s (FEV 1.0) within normal limits. The results were not significantl y different from a similar spirometric examination performed at a health surve y the day before the first episode of hypersensitivity pneumonitis-like symptoms.
to the analysis, which was considered to support the specificity of the test system .
The RAST and ELISA analyses were later repeated with a serum sample obtained from the patient in September 1987. He had not been exposed to isocyanates during the three intervening years. Once again, the RAST was negative, and the total IgE concentration was low, 10 kU/I. A borderline positive ELISA reaction was observed for HOI but not for the other two isocyanates tested.
The antibody prevalence in the various groups is shown in table I, where the group of exposed individuals has been subdivided according to occupation.

Discussion
In this case report, there is a lack of certain medical information to support a definite diagnosis of hypersensitivity pneumonitis (15). Specifically , there is no evidence of an impaired diffusion capacity or radiographic signs of alveolitis in the acute stage of the disease. A pulmonary radiograph with normal findings obtained some 12 h after the patient's admittance to the hospital does not preclude the diagnosis. We find the history and the symptoms, the clinical course, and the relapses upon repeated exposure to be highly suggestive of hypersensitivity pneumonitis. The offending agent seemed to be related to the somewhat uncontrolled exposure to a two-component acrylic lacquer with an isocyanate (HOI) hardener. During such work operations, concentrations of isocyanates (especially of the HOI oligomer, usually named "HOI biuret-trimer") several times higher than the Swedish occupational exposure limit have been recorded (16)(17)(18)(19).
Attempts were made to collect immunologic evidence in support of the diagnosis. The absence of IgE antibodies against HOI was not an unexpected finding insofar as the clinical picture did not suggest an immediate type of allergic reaction . Moreover, in isocyanate-induced asthma, IgE antibodies are an uncommon finding, although hapten sensitization of the airways by occupational exposure to various chemicals has been increasingly described.
The presence of serum IgG antibodies to an HDI-HSA conjugate followed by specific inhibition after preincubation was considered to support the conclusion that HOI was the cause of the disease. The significance of such antibodies for the case validation remained unclear, however, and we decided to investigate their presence in various groups of individuals.
As can be seen from table I, the occurrence of specific IgG antibodies is clearly related to isocyanate exposure. The prevalence, however, is dependent on the definition of a positive test. Among the isocyanateexposed groups, a serum dilution of only 20 in the ELISA gave positive results for up to 33 070 (mean 10 0J0) of the samples. The analysis of serum diluted lOa-fold reduced the number of positives in the exposed (but healthy) groups.
The proportion of positive tests remained unaffected among isocyanate asthmatics with specific IgE antibodies since these individuals generally had high IgG antibody titers. The high prevalence of specific IgG antibodies among these asthmatics is probably due to common factors controlling both IgE and IgG production. It should be noted that no symptoms of hypersensitivity pneumonitis had occurred in this group .
The patient was found to be positive according to both criteria when the December 1984serum was used but only positive according to the first criterion when the serum from September 1987 was used.
The findings in the evaluation of this ELISA system are rather similar to observations on alveolitis such as farmer's lung, for which IgG antibodies are supposed to play only a minor immunopathological role (20,21). Nevertheless, IgG antibody assays are valuable as evidence of accumulated exposure and may thus be considered supportive proof in clinically clear cases of hypersensitivity pneumonitis. From the low inci- dence of thisdisease among isocyanate-exposed workers, it follows that the sensit ivity of the test syst em needs to be fu rther evaluated. It sho uld be noted that a specif ic challenge test was not performed. The Swedi sh rules for workmen's compensation are comparatively liberal, and the Social Insurance Office accepted our patient's disease as being of occupational origin, and he recei ved some compensatio n . Thus, as the patient would not in any respect benefit from renewed exposure to a potentially harmful agent (ie, HDI), a challenge test was considered unethical.
We conclude that the demonstration of IgG antibodies in sera from patients with possible isocyanateinduced hypersensitivity pneumonitis furnishes evidence of immunostimulating hapten exposu re. The co rr ela tio n of the immunologic findings with the d isease process seems to be of low significance, however. It is not justified to look upon these antibodies as evidence of isoc yanate-induced hypersensitivity pneumonitis.