NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Research Council (US). Multiple Chemical Sensitivities: A Workshop. Washington (DC): National Academies Press (US); 1992.

Cover of Multiple Chemical Sensitivities

Multiple Chemical Sensitivities: A Workshop.

Show details

Diagnostic Markers of Multiple Chemical Sensitivity

G. Heuser, A. Wojdani, and S. Heuser

One hundred thirty five patients (75% females) were evaluated for complaints of often disabling sensitivity to small concentrations of multiple chemicals after chemical exposure in the recent or distant past.

A comprehensive evaluation of all subjectively involved systems showed a high yield of abnormal objective findings on random (unrelated to time of exposure) testing. When properly timed, certain immune function tests (TA1 cells and chemical antibody levels) became abnormal, or more abnormal, after unintentional self-reported acute exposure and were thus shown to be potential markers of multiple chemical sensitivity (MCS).

We suggest that appropriate tests of the central nervous system, peripheral nervous system, nose and sinuses, pulmonary function, T-cell subsets, chemical antibodies and autoimmunity be performed. If four of these seven systems show abnormality, the diagnosis of MCS is supported. If certain functions become abnormal or more abnormal after unintentional significant exposure, the diagnosis h confirmed.


The senior author has directed a headache and chronic pain clinic for more than ten years. As clinical histories were taken in increasing detail over the years, it became apparent that in many patients, headaches were triggered by chemicals in the environment. What appeared to be very small concentrations of perfumes, fumes, smoke and mist, would trigger headaches in some patients but not in friends and family members who were present in the same environment. Eventually, the senior author realized that small amounts of chemicals cannot only cause headaches but also a multitude of other complaints in some patients. This realization became a starting point for his increasing interest in chemical sensitivity.

One of the co-authors (A.V.) is also chemically sensitive. This is why, as an immunologist, he became interested in developing appropriate tests for chemical sensitivity. He made chemical antibody testing commercially available to our patients.

The other co-author (S.H.) specializes in computer assisted medical information retrieval. She works one day a week in a one story office which was built more than ten years ago. Furniture, equipment, carpets, and paint are also about ten years old. Never-the-less, she recently developed headaches and nose bleeds. These regularly developed the morning after she had worked in the office. An investigation showed that the manually set air intake valve for that office building had fallen shut. As a result, all air in the office was now being recirculated. No fresh air could enter anymore. When this situation was corrected, both headaches and nose bleeds disappeared. No other person in the office developed headaches and nose bleeds. The conclusion was reached that S.H. is chemically sensitive.

The three of us combined forces in order to develop a team approach to multiple chemical sensitivity (MCS). Consultants in varying specialties are complementing the efforts of our core group.

Initially, we became interested in complaints of chemical sensitivity in our headache patients. Now, we attract an increasing number of environmentally ill patients (''EIs''). Many of them claim disability and request help so that they can get State or Social Security. Others find themselves in litigation (personal injury, workers' compensation). Therefore, an objective evaluation of all patients became mandatory.

Patients with MCS present to their physicians with a startling spectrum of complaints. These complaints are often presented in a manner highly suggestive of a psychiatric disorder. None of them easily lend themselves to measurement. General malaise and weakness, fatigue, headaches, irritability, depression, memory problems, itching, numbness, "creepy and crawly" sensations, burning sensations in the nose, sore throat, hoarseness, shortness of breath, cough, abdominal distress, are all "soft" complaints and therefore not easily documented.

Patients may arrive in the office equipped with oxygen tank, air filter, and mask. Typically, patients describe either one single or repeated exposures to one or more chemicals in the past and the development of chemical sensitivity thereafter. It is striking how similar the history is, regardless of education or walk of life or background. This consistency from one patient to another certainly suggests that we are dealing with a disease entity.

On physical examination there is striking paucity of abnormal findings. Small erythematous lesions over the skin areas exposed to chemicals come and go. On chest examination, wheezing is occasionally found. Soft signs may or may not be present on neurological examination.

Routine laboratory tests (CBC, blood chemistry) and EKGs are also usually normal. We quickly learned that chemically sensitive patients almost regularly refuse to deliberately expose themselves to chemicals. We thus abandoned the idea of exposing our patients to chemicals in a controlled environmental chamber. Instead, we waited for unintentional exposure to occur and suggested testing within a given interval thereafter. The results were then compared with base line values obtained either before or a longer time after acute exposure.

Our results led us to conclude that patients who claim MCS could and should be used as their own controls. Obviously, whenever research funds are available, this patient population should be compared to non-exposed and non-symptomatic matched controls.

A patient with multiple complaints is difficult to evaluate. Most physicians have learned that such a patient usually turn out to suffer from "psychosomatic illness". This is why even a well trained physician is easily persuaded to assign a psychiatric diagnosis to patients with multiple chemical sensitivity. By contrast, it became our goal to look for objective markers of multiple chemical sensitivity. We believe that this paper shows initial steps in this direction.

Materials and Methods

Control Population. One hundred sixty healthy volunteers were examined. Their ages ranged from 22-55 years. They had no known disease and denied drug use and smoking. 60% were caucasian, 20% african-americans, 20% were hispanics. 62% were females, 38% were males. Their bloods were examined and normal ranges were established for all immune function tests reported in this paper.

Experimental subjects. One hundred thirty five patients (75% females) were selected from our private patient population. All complained of sensitivity to small amounts of chemicals in their work, home, or other (commuting, shopping, hobbies) environments. This sensitivity had diminished the quality of life in most patients many of whom claimed disability. In the extreme they now lived in self-imposed isolation from society. Some had gone to extraordinary efforts to live in an environmentally "clean" home. Some lived in tents on the beach, some in remote locations in California or out of state. Some had held well paying jobs, had made significant contributions to society but had to give it all up because of their environmental illness.

All patients were personally examined by the senior author. Their work or home environments could not be examined by the authors who had to rely on the patients' reports. Material safety data sheets (MSDS) were examined whenever available from the work site.

Table I lists the settings in which patients claimed to have been exposed. Please note that most exposures were due to groups of chemicals rather than a single chemical.

TABLE I. Group of 135 Patients Exposed to Chemicals.


Group of 135 Patients Exposed to Chemicals.

All laboratory tests were initially obtained at random i.e. regardless of time of chemical exposure or severity of symptoms. Whenever possible, tests were thereupon specifically timed so as to be obtained a given number of days after exposure and resultant symptoms. Patients were thus instructed to wait for follow-up tests until they were involuntarily exposed and developed symptoms as a result.

Whenever symptoms suggested impairment of neurological or psychological function, we suggested an electroencephalogram (EEG) and evoked response studies. MRI of the brain was usually ordered only if these electrophysiological studies were borderline or abnormal. More recently, we started ordering brain mapping and SPECT studies in a select group of patients.

Whenever symptoms suggested a peripheral neuropathy, we ordered nerve conduction and/or neurometric (perception threshold) studies.

If patients specifically complained of nasal or throat symptoms, consultation was obtained from a board certified ENT-specialist. Studies then frequently included sinus x-rays.

Shortness of breath and related complaints led to pulmonary function testing (PFT) and if indicated, chest x-rays.

Patients underwent immunological studies whenever possible.

We routinely obtained CBC and blood chemistry. As indicated, additional studies (viral, bacterial, endocrine and others) were done.

Neurological testing. Most EEGs were performed in the office of G.H. and included spontaneous sleep, hyperventilation, and photic stimulation. A 16 channel Beckman instrument was used. EEGs had at times been performed elsewhere.

Most MRI brain scans were performed at Medical Imaging Center of Southern California and read by a neuroradiologist.

Perception thresholds were tested with a Neurometer and evaluated by a neurologist with special experience in the use of this instrument. The results are indicators of peripheral nerve (A to C fibers) function.

Evoked response and conduction studies were done and evaluated by board certified neurologists.

SPECT scanning was done at Harbor General Hospital (UCLA affiliated).

Nasal and sinus evaluation. All symptomatic patients were referred to a board certified specialist.

Pulmonary evaluation. Pulmonary function tests (PFT) were done with a Brentwood Spirometer in our office. Some patients had undergone testing elsewhere.

Immune testing. Tests were performed at Immunosciences Laboratories under the direction of A. Wojdani, Ph.D.. In view of the importance of these tests in the context of this evaluation, the methodology is described in detail.

Preparation of formaldehyde-human serum albumin (F-HSA) and Formaldehyde-bovine serum albumin (F-BSA) conjugates:

F-HSA and F-BSA was prepared by the method of Patterson et. al. (1985). Briefly, one mg. of HSA or BSA (Biocell, Carson, Ca.) in PBS (Phosphate Buffered Saline) PH 7.4, each separately, were exposed to 270 of formaldehyde (Fisher Scientific, Fairlawn, N.J.). The mixture was incubated for 30 minutes at 37 C and was then extensively dialyzed against PBS. The F-HSA or F-BSA was sterilized with a 0.2 m filter (Milliport Corp., Bedford, Mass.).

Electrophoretic and immunoelectrophoretic comparison of HSA, BSA With F-HSA and F-BSA was performed to determine conjugation occurrence. Conjugation was evidenced by altered mobility of F-HSA, F-BSA when it was compared with HSA or BSA respectively. Moreover, the number of free amino add groups present in the F-HSA or F-BSA was determined by the method of Snyder and Sobocinski (1975) and was used to assess the amount of substitution. The number of amino groups bound to formaldehyde was 26 for HSA and 31 for BSA. In this calculation, the formation of intermolecular cross-linking was considered.

Preparation of toluene diisocyanate-human serum albumin (TDI-HSA) and Bovine serum albumin (TDI-BSA) conjugates:

This preparation was similar to the methods of Dewar and Baur (1982). According to this method, 1g HSA or 1g BSA was dissolved in 100 ml of a buffer solution containing potassium chloride (0.05 mol/l), sodium borate (0.05 mol/l), PH 9.4 and cooled to 4 C. Dioxane (10 ml) containing 0.15 ml of toluene diisocyanate was then added dropwise while stirring over a period of 3 hours, followed by addition of 2 ml of ethanolamine, centrifugation, dialysis filtration and lyophilization. Similar to F-HSA and F-BSA, conjugation was confirmed by electrophoresis and determination of free amino groups present in the conjugate. The number of amino groups bound to TDI was 37 for HSA and 43 for BSA. In addition, spectrographic analysis of the conjugate was undertaken according to Zeiss et. al. (1980). There was a marked increase in absorption from 230 to 260 nm which indicated that TDI had become covalently linked to the protein carrier. This increase in absorption did agree with NH2 group determination only 76% for HSA and 81% for BSA

Preparation of trimellitic anhydride-human serum albumin (TMA-HSA) and Trimellitic anhydride bovine serum albumin (TMA-BSA):

To prepare these conjugates 25 mg. of TMA was dissolved in 0.5 ml of dioxane and added dropwise either to 25 mg of HSA or BSA dissolved in 5 ml of cold 7% NaHCO3 in water. After stirring for 60 minutes at 4 C the conjugates were dialyzed against four changes of 0.1 M NaHCO3 and one change of buffer. Finally the conjugates were filtered and kept at -20 C until used. OD analyses of TMA-HSA, and TMA-BSA were done to determine the number of TMA residues linked to the corresponding carrier protein. The concentration of the carrier protein was converted to molar concentration with the molecular weight of HSA and BSA. From the ratio of the molar concentration of the TMA ligand and the protein carrier, the ratio of TMA residues per molecules of carrier was calculated. TMA-HSA was estimated to contain 5 TMA residues per HSA molecules and for TMA-BSA seven residues per albumin molecule (Pien et al., 1988).

Preparation of phthalic anhydride-human serum albumin (PA-HSA) and Phthalic anhydride bovine serum albumin (PA-BSA) conjugates:

These hapten-conjugates were prepared by adding 75 mg of PA to a cooled solution of 300 mg of HSA or BSA in 100 ml of H2O. The reaction mixture was stirred overnight, dialyzed against 0.1 M PBS using tubings with a cutoff of 8000 dalton. Using the method of Zeiss et. al. (1977) the molar ratios were calculated. The molar ratios were found to be 22/28 for PA/HSA and 25/30 for PA/BSA.

Preparation of benzene ring HSA (B-HSA) and Benzene ring BSA (B-BSA) conjugates:

For these preparations, 40 mg. of P-aminobenzoic acid was dissolved in 2 ml of 1 N HCL and cooled by immersion in an ice bath. A cold solution of 14 mg/ml was added dropwise. After each addition, the mixture was stirred for 30 seconds. In parallel, one gram of HSA or BSA was dissolved in boric acid 0.16 M sodium chloride (0-15 M) buffer PH 9.0 (PH was raised with NaOH). The beakers containing the solutions of albumins were surrounded by an ice bath on magnetic stirrer. The solution of diazonium salt was added dropwise, with rapid stirring to the cold protein solution. After addition of each drop the PH is readjusted to 9.0 to 9.5 with one normal NaOH. When all the solution had been added, the reaction was allowed to continue with slow stirring for at least an hour with further additions of NaOH solution and maintaining the PH at the range of 9.0 to 9.5. Unreacted small molecules were removed by extensive dialysis or by passage through a column of sephadex G-25 in the cold room, with an isotonic salt solution as the eluting buffer. OD analyses of orange color development of B-HSA and B-BSA were done to determine the number of B residues linked to the corresponding carrier protein. The amount of B substitution for HSA were approximately 41 and for BSA 53 (Migrdichian, 1957).

Antibody Determination:

Specific antibodies against F-HSA, TDI-HSA, TMA-HSA, PA-HSA and B-HSA were analyzed by a noncompetitive ELISA assay. Wells of microtiter plates (Dynatech, Alexandria, VA) were coated with 100 1 of antigen solutions (100 g/ml) in 0.1 M PBS PH 7.2 overnight at 4 C. Plates were washed 4 times with 0.1 M PBS containing 0.05% tween 20 between each step. Free absorption sites were blocked with 2% protease free bovine serum albumin at room temperature for 4 hours and stored at -20 C until used.

Analytical Procedure:

The procedure included the following: (1) washing four times, (2) addition of 100 1 of diluted serum (1:2 for IgE and 1:100 for IgM and IgG) in PBS tween-20 with 1% BSA (3) incubation for 4 hours at 20 C, followed by washing 4 times, (4) addition of 100 1 of an optimal dilution of alkaline phosphatase labeled affinity purified goat anti-human IgE ( ) (1:200), IgM (1:500) and anti IgG (1:1000), purchased from KPI (Maryland) (5) incubation for 120 minutes at 20 C, (6) washing 6 times, (7) addition of 100 1 of P-nitrophenyl phosphate (Sigma Chemical Co.) (8) incubation for 60 minutes at 20 C (9) addition of 50 1 of 3 N sodium hydroxide solution, and (10) duplicate reading. The results were calculated based on absorbances of duplicate samples of 405 nm using microtiter reader. All samples were read against an HSA antigen as a control of binding not specific to F-HSA, TDI-HSA, TMA-HSA, PA-HSA and B-HSA. Results were expressed as titer. Titer is being the last dilution of serum giving absorbance twice of HSA control.

Specificity And Cross-Inhibition Studies:

For determination of antibody specificity a cross-inhibition study was undertaken. Positive sera for each hapten-protein conjugate were run after appropriate incubation and precipitation with tenfold increasing increments of hapten bound HSA or BSA as inhibitors to cover the range of antibody to antigen excess. This range was between 50 g to 1000 g for hapten-BSA and 80 g to 1000 g for hapten-RSA. After incubation at 37 C and removal of precipitate by centrifugation, the samples from before and after cross-inhibition study were then placed on plates with wells coated with the specific conjugate. The subsequent steps were followed as described above for the ELISA study.

IgG and IgM antibody binding to different conjugates was inhibited by hapten-HSA or hapten-BSA from 36-85%. At a given concentration, both hapten-HSA and hapten-BSA inhibited the antibody level in similar manners.

Partial inhibition of IgE antibody binding to different hapten conjugates was observed when serum was pre-incubated with hapten-HSA or hapten-BSA. This incomplete observation of inhibition of IgE antibody was mainly related to nonavailability of serum with high IgE titers against different chemicals in our laboratory.

Determination of Normal Levels of Antibodies (Controls):

Based on the above procedures, 160 blood donor samples of healthy individuals, of both sexes, between the ages of 22-55, were examined for antibody levels against F-HSA, TD-HSA, TMA-HSA, PA-HSA, and B-HSA. The average titer was 1:800 400 for IgG, 1:3200 1600 for IgM and 1:8 4 for IgE. Thus, in our laboratory titers greater than 1:1600 for IgG, 1:6400 for IgM and 1:16 for IgE are considered positive.

In a given patient, rises or falls in antibody titers by more than one dilution were considered significant (see Tables).

Lymphocyte Subset Enumeration:

A single laser flow cytometer (Epics Profile: Coulter Epics, Inc., Hialeah, FL) which discriminates forward and right angle light scatter, as well as two colors, was used with a software package (Quad Stat: Coulter). Mononuclear cell populations were determined by two-color direct immunofluorescence by using a whole-blood staining technique with the appropriate monoclonal antibody and flow cytometry (Fletcher et al., 1989) The following pairs of fluorescein isothiocyanate (FITC), or phycoerythrin (PE)-conjugated monoclonal antibodies (Coulter immunology) were selected: T11-RDI/B4-FITC, T4-RDI/T8-FITC, T3-FITC/NKH-1-RDI and T11-FITC/Tal-PE for determination of T-cell/B cell, T-helper/T-suppressor, NKHT3+ /NKHY3 and for alternate pathway of lymphocyte activation respectively.

To monitor lymphocyte markers, bat maps were set on the lymphocyte population of the forward-angle light scatter versus a 90 light scatter histogram. The percentage of positively stained cells for each marker pair, as well as the percentage of doubly stained cells was determined. Estimates of absolute numbers of lymphocytes positive for the respective surface markers were determined by multiplying peripheral lymphocyte cell count by the percentage of positively stained cells for each marker pair. Also, the percentage of doubly stained cells was determined. Estimates of absolute numbers of lymphocytes positive for the respective surface markers were determined by multiplying peripheral lymphocyte cell count by the percentage of positive cells for each surface marker.

Measurement of anti-myelin basic protein antibodies:

Human myelin basic protein (HMBP) was prepared by the method of Diebler et al. (1972) and checked for purity by polyacrylamine gel electrophoresis. Antiserum to HMBP was induced in rabbits by repeated injection of HMBP in complete Freund's adjuvant. Antibody activity in the rabbit sera and patient's samples was detected by adding different dilutions (1:100 to 1:10,000) of sera to wells of a microtiter plate previously coated with HMBP as follows: HMBP 250 g/ml was dissolved in carbonate buffer, PH 9.6 and 200 l of this solution were added to each well. After incubation, washing and blocking as above, 200 1 of either diluted rabbit or human serum were added to the wells. After incubation for I hour at 37 C the sera were shaken out of the wells and then were washed 5 times with wash solution. 200 1 of peroxidase-conjugated goat anti-rabbit or goat anti human IgG, IgM or IgA (optimal dilution) were added to the appropriate well. After incubation and repeated washing 200 1 of ABTS substrate were added to each well. Plates were incubated for one hour at room temperature and read in a microtiter reader at 405 tun wavelength. Using rabbit antisera; a titration curve was plotted and patient's sera were compared to this standard curve. Based on more than 200 controls and patients' sample determinations, liters greater than 1:2000 for IgA, 1:5000 for IgM, and 1:8000 for IgG were considered positive.


Table II lists the neurological tests done in patients who had complaints of headaches, irritability, memory loss, depression, numbness, tingling, crawling sensation etc. Some studies were done in sufficient number to be suggestive of a significant trend. EEGs were abnormal in 45% of tested patients. They showed mild rather than severe abnormalities, with mostly unilateral (at times bilateral) intermittent slowing, dysrhythmia, and occasional single sharp waves and spies in the temporal and adjacent leads. MRI scans were also abnormal in a high percentage of cases. In some scans there was a definite impression of atrophy (13%) or demyelinating disease (7%). Others (8%) had more ill-defined non-diagnostic lesions. Visual evoked (VER) and brain stem auditory evoked (BAER) responses were also abnormal in a high percentage of cases.

TABLE II. Neurological Tests in Patients with Multiple Chemical Sensitivity.


Neurological Tests in Patients with Multiple Chemical Sensitivity.

The number of patients who underwent single photon emission computerized tomography (SPECT) studies of brain perfusion and metabolism as well as computerized analysis of their EEG activity (BEAM) was small but initial results suggest further studies in a greater number of patients.

Current perception threshold studies by Neurometer were also performed in only a small number of patients. Never-the-less, they hold promise as potential markers of peripheral neuropathy, just as conduction studies do.

Table III shows that a thorough ENT exam will show abnormalities in a high percentage of patients. The consistent findings were atrophic rhinitis in patients with severe nasal complaints. Sinusitis or at least thickened mucous membranes were found on sinus examination.

TABLE III. Examination of Nose, Sinuses and Pulmonary Function in Patients with Multiple Chemical Sensitivity.


Examination of Nose, Sinuses and Pulmonary Function in Patients with Multiple Chemical Sensitivity.

The typical abnormality on PFT was a decrease of FEF 25-75% to below 70% of predicted value, indicating small airway disease.

The tests done in table II and III were obtained at random i.e. unrelated to time of exposure.

Table IV shows that, again on random testing, increase in TA1 cell count and percentage is the most frequent abnormality in patients with MCS. Helper/suppressor (H/S) ratios can be increased (50%), unchanged, or decreased upon random testing. Suppressor cells were decreased in 27% of 110 patients. Whether their continued decrease leads to auto-immune disease is not yet apparent from our initial data. Mitogenesis was abnormal in 42% of 12 patients. Normal ranges in our control group were as follows: TA1 Cells 0-432/mm3 or 0-10%; H/S ratio 1-2.2; Helper Cells 336-2,376/mm3 or 35-55%; Suppressor Cells 192-1598/mm3 or 20-37%; Lymphocytes 960-4,320/mm3 or 20-40%; B Cells 48-648/mm3 or 5 – 15%.

TABLE IV. Immune Cell Population in Patients with Multiple Chemical Sensitivity.


Immune Cell Population in Patients with Multiple Chemical Sensitivity.

Table V shows abnormal levels of chemical antibodies in a high percentage of patients with MCS. By contrast "normal" (patients unaware of symptoms from chemical exposure) individuals remained in the normal range. (see Materials and Methods section).

TABLE V. Elevation of Antibodies to Chemicals in Patients with Multiple Chemical Sensitivity.


Elevation of Antibodies to Chemicals in Patients with Multiple Chemical Sensitivity.

Table VI illustrates elevated levels of chemical antibodies in a symptomatic patient. There was a significant change in IgG (benzene ring) and IgM (isocyanate) levels which decreased after exposure ceased and the patient became "asymptomatic" (while however still chemically sensitive).



K. W., 42, female, was symptomatic from exposures to auto mechanic repair shop in 3/90. No more exposures after early April 1990.

We also examined for IgE chemical antibodies which were not elevated in any of our studies.

Table VII shows a significant decrease to normal levels of chemical antibodies in a patient who traveled out of state and stayed in a "non-contaminated" environment where she slowly became "asymptomatic". TA1 cells were not a good marker of MCS in this case.



M. D., 53, female, with MCS before and after leaving ''toxic'' environment at her home.

Table VIII illustrates a case in which chemical antibodies were a poor marker of chemical sensitivity. By contrast, cells of the immune system were significantly abnormal and slowly approached normal as the patient stayed away from her home for several months. While the patient was originally bed ridden from exposure at her home, she was ambulating and much improved after four months away from her home. By contrast, her husband denied chemical sensitivity (he actually had some minor symptoms on detailed questioning) and could only be persuaded to be tested after he was away from his home for four months. It should be noted that the couple re-entered their home every few weeks to fetch some of their belongings. This led to intermittent exposure and may account for the slow recovery of the wife. While the patient claimed exposure to malathion, her home had possibly become contaminated by other pesticides.



Married couple, (male, 45 and female, 56) exposed to malathion with symptoms far more severe in the wife.

Table IX illustrates rapid increase in TA1 cells and decrease in immunocompetent natural killer cells (NKHT3) in a student with MCS who entered an anatomy laboratory for sufficient length of time to become severely symptomatic. By contrast, T-cells and helper-suppressor ratios did not change within that same time interval.



Patient L. W., 34, female with MCS, reacting to exposure in an anatomy laboratory on 8-12-90.

Table X illustrates increase in IgM antibodies to TMA, phthalic a-hydride and compounds with a benzene ring, and also in IgG antibodies to the benzene ring, approximately two weeks after significant exposure. By contrast changes in TA1 cells were seen within only one day! Note that other parameters were unchanged when studied during this short time interval.

TABLE X. Chemical Antibodies and Subpopulations Before and After Exposure.


Chemical Antibodies and Subpopulations Before and After Exposure.

Table XI illustrates that antibodies to TMA can be the only antibodies elevated after exposure. In table XII only two antibodies (isocyanate and chemicals containing the benzene ring) are elevated after significant exposure. Table XIII illustrates increase in benzene related antibodies in the wife who showed more evidence of chemical sensitivity than the husband. Note that other antibodies were not elevated. Results in tables XI, XII and XIII suggest specificity in the respective chemical antibody tests.



Patient, L. W., 34, female, with MCS, reacting to exposure in a "Sick Building".



Patient L. C., 45, male, with MCS before and after exposure to "noxious" environment on 9-25-90.



Married couple (male, 43 and female, 35) with MCS after intermittent exposure (ongoing) to gasoline fumes.

Table XIV shows increased antibodies in husband and wife. Both claimed total disability from severe MCS with the wife showing more symptoms. Testing was done at random and years after the acute exposure. Table XV also illustrates the long term effects of exposure in a mobile home which the couple had moved out of three years earlier. They both claimed MCS and were symptomatic when seen in our office. Note elevated antibodies to formaldehyde and TMA together with high TA1 counts in both husband and wife.



Married couple (male, 38 and female, 34) with MCS after exposure to roofing materials approximately 2 years earlier.



Married couple (male, 60 and female, 58) with MCS after exposure to Formaldehyde in 1986. Tests done in 1989.

Table XVI depicts the presence of auto-antibodies in our patient population.

TABLE XVI. Autoimmunity in Patients with Multiple Chemical Sensitivity (N=92).


Autoimmunity in Patients with Multiple Chemical Sensitivity (N=92).

Not shown in table XVI are additional results with respect to autoimmunity: ANA titers were positive in 17% of 96 patients. The highest percentage of elevated antibodies was seen when anti-myelin antibodies were studied. They were positive in 80% of 50 patients studied. Typically the elevation was in the IgM and/or IgA rather than IgG antibodies.

In contrast to our findings, "normals' have been reported to have a very low incidence of positive auto-antibodies: ANA 3-4%, parietal approximately 2%, smooth muscle approximately 3%, mitochondrial approximately 1%.


We embarked on our studies with the hope that patients with a claim of MCS could eventually be objectively evaluated. We felt that an approach should be found which could accommodate a great number of patients and could be supervised by a primary physician.

Our studies were done by specialists and with equipment available in most dries in the US. The immunological studies described in this paper are sophisticated but can be executed on blood which is mailed over-night to an appropriate laboratory.

The growing number of patients claiming MCS will in our opinion make it impossible to study them all in environmental chambers under controlled conditions. These chambers will however be needed for further research.

Table XVII shows a list of chemicals brought to the senior author by a patient with MCS. She was exposed to some or all of these chemicals at work on an everyday basis. This table illustrates the dilemma for both patient and physician in trying to attempt to disentangle a complex problem such as this. An environmental chamber approach might not be practical in such a case as it would take too long and would be too expensive to study this matter in great detail

TABLE XVII. Chemicals Used at Work.


Chemicals Used at Work.

While exacting research requires well selected controls, these are not easily found in our polluted urban environment. For example, most people are more or less aware of and effected by pollution in Los Angeles. Thus, there seems to be a whole spectrum of sensitivity, with our patient population being at one extreme.

Nevertheless we, have accumulated in excess of one hundred patients who did not complain of MCS and had no elevation of chemical antibodies (also see Material and Methods section for discussion of controls) nor of TA1 cells.

In view of the above we feel that a patient should serve as his/her own control and that therefore all studies should be longitudinal. Our results show that properly timed studies can bring about significant changes in certain parameters after self-reported exposure. We are not certain at this time when TA1 cell counts reach a peak after exposure. We now know however, that these cells are elevated one to two days after exposure. We are also uncertain when chemical antibody levels reach a peak value. All we have shown so far is an elevation about two weeks after exposure.

Chemical exposure and its effects on the immune system has recently become the subject of discussion by leading allergists (Salvaggio, 1990). Changes in immune cell populations, specifically TA1 cells, alter exposure to chemicals were recently described by another group (Thrasher et al., 1989; Thrasher et al., 1990).

Chemical antibody measurements in chemically exposed patients recently became commercially available. Appropriate immunological procedures were originally developed by Dr. Wojdani and used in patients of the senior author (Thrasher et al., 1987). The original procedures were then expanded by Dr. Wojdani to include additional chemicals (this paper) and adopted and verified by another laboratory (Antibody Assay Laboratories) where additional research was done and published (Thrasher et al., 1989; Thrasher et al., 1990).

At this time, it should be noted that benzene is not per se antigenic. However, our data suggests that some chemical compounds containing the benzene ring are antigenic. Further studies are needed to determine which of these compounds cause antibody formation.

Our data suggest that chemical exposure eau push some patients in the direction of autoimmune disease. Multiple sclerosis is an example. A number of our patients are suspected of having that disease on the basis of not only their clinical presentation but also abnormal MRI and evoked response studies together with high anti-myelin antibodies. This was previously discussed (Gard and Heuser, 1990).

Studies other than immune tests should also be done in a longitudinal fashion. Cost containment, a lack of research funding and other factors made this impossible in our patient population. However, PFT were at times studied immediately after exposure and became abnormal. It is possible that some neurological parameters (EEG, BEAM, and SPECT) may also show some significant changes.

While we have come to expect a high percentage of abnormal immune function tests, we were surprised at the high percentage of abnormal neurological tests. This indicates that the "psychiatric" presentation by many of these patients may well have a neurological basis.

The high number of abnormal test results in our patient population is probably explained by the fact that many patients were disabled with MCS and therefore quite sick.

Our EEG and SPECT studies point toward the limbic system as being involved in MCS. This system's possible role is aptly discussed in this conference by Dr. Miller. It, together with the role of the olfactory system deserves further study (also see Dr. Bell's presentation at this conference). Early studies by Russian authors (Bokina et al., 1976) pointed in this same direction.

While EEG studies showed mostly mild abnormalities in the temporal and adjacent leads (see results), our youngest patients (sister and brother, ages two and four respectively) developed actual clinical seizures with grossly abnormal EEGs about three weeks after moving into a new home and playing on the brand new carpet. For several months thereafter the mother observed MCS in both her children. Seizures and MCS slowly abated after the family moved out of the new home.

Diagnostic criteria for MCS. The diagnosis of MCS should be suspected if a patient reports impaired well-being whenever exposed to more than one chemical in concentrations which do not effect the general population. In the extreme, the concentrations are very low and the patient is very sick and claims disability.

A comprehensive evaluation of seven systems should then be undertaken. It should be understood that not all seven systems (central nervous system, peripheral nervous system, nose and sinuses, PFT, T-cells subsets, chemical antibodies, autoimmune panel) are always affected. However, we suggest that abnormalities in four out of these seven systems strengthen the suspicion of MCS. If parameters become abnormal or become more abnormal following self-reported acute exposure, the diagnosis is basically confirmed. Studies in environmentally controlled chambers will be necessary to further advance the field.

The suggestion to use four out of seven criteria is taken from the diagnostic criteria for systemic lupus erythematosus (SLE), where four out of eleven criteria have to be present to make the diagnosis (see table XVIII). It should be noted that psychiatric features are seen in a significant number of patients with SLE. Thus, the same should not be unexpected in patients with MCS.


Patients who present with complaints of MCS deserve a comprehensive objective evaluation. If this is performed, a high percentage will be shown to have abnormal test results. This is true if the central and peripheral nervous systems as well as pulmonary and immune functions are tested. Also, anatomical changes are frequently found in the nasal passages on close inspection. By contrast, CBC and blood chemistry are usually within normal limits. So are findings on general physical examination.

Whenever possible, longitudinal studies should be performed in which the patient is used as his/her own control. Increases in TA1 cells and chemical antibodies can then be seen following self-reported unintentional exposure and are therefore suggested as markers of MCS.

Our results suggest diagnostic criteria for MCS. These are sorely needed as the number of patients who claim disability as a result of MCS is growing. Millions of dollars are potentially at stake as claims increase. Patients who are truly sick deserve attention and help from industry, housing authorities and government agencies as well as physicians. Patients who make unjustified claims should be quickly identified.

Patients and industry and government are all in need of a practical approach to the diagnosis of MCS. We believe that our findings are pointing the way to such an approach.

Acknowledgements: We thank Dr. D. Alessi for ENT evaluations, Dr. R. Holgate (neuroradiology) for evaluation of MRI brain scans, Dr. R. Lawrence (neurology) for interpreting the neurometer studies and Dr. I. Mena (nuclear medicine) for evaluation of SPECT brain scans .


  • Bokina, A.I., N.D. Eksler, A.D. Semenenko, and R.V. Merkur'yeva. 1976. Investigation of the mechanism of action of atmospheric pollutants on the central nervous system and comparative evaluation of methods of study. Environ. Health Perspect. 13:37-42. [PMC free article: PMC1475010] [PubMed: 1269506]
  • Dewar, M.A., and X. Baur. 1982. Studies on antigens useful for detection of IgE antibodies in isocyanate-sensitized workers. J.Clin. Chem. Clin. Biochem. 20:337-340. [PubMed: 6288831]
  • Diebler, G.E., R.E. Martenson, and M.W. Kies. 1972. Large scale preparation of myelin basic protein from central nervous tissue of several mamalian species, Prep. Biochem. 2:139-165. [PubMed: 4623901]
  • Fletcher, M.A., S. Azen, B. Adelsberg, G. Gjerset, J. Hassett, J. Kaplan, J. Niland, T. Maryhon, J. Parker, D. Stites, and J. Mosley. 1989. Immunophenotyping in a multicenter study. The transfusion safety study experience. Clin. Immunol. Immunopathol. 52:38-47. [PubMed: 2656018]
  • Gard, Z. and G. Heuser. 1990. Re: Multiple sclerosis, solvents and pets. Letter to the Editor. Arch. Neurol. 47:138.
  • Migrdichin, V. 1957. Conjugated and synthetic antigens in organic synthesis. Reinhold, New York.
  • Patterson, R., K.E. Harris, and L.C. Grammer. 1985. Canine antibodies against formaldehyde-dog serum albumin conjugates: induction, measurement, and specificity. J. Lab. Clin. Med. 106:93-100. [PubMed: 4009027]
  • Pien, L.C., C.R. Zeiss, C.L. Leach, N.S. Hatoum, D. Levitz, P.J. Garvin, and R.Patterson. 1988. Antibody response to trimellitic hemoglobin in trimellitic anhydride induced lung injury. J. Allergy Clin. Immunol. 82:1098-1103. [PubMed: 3204253]
  • Salvaggio, J.E. 1990. The impact of allergy and immunology on our expanding industrial environment. J. Allergy Clinical Immunol. 85:689-699. [PubMed: 2182694]
  • Snyder, S.L. and P.Z. Sobocinski. 1975. An improved 2,4,6-Trinitrobenzosulfonic acid method for the determination of amines. Ann. Biochem. 64:285-288.
  • Thrasher, J.D., A. Wojdani, G. Heuser, and G. Cheung. 1987. Evidence for formaldehyde antibodies and altered cellular immunity in subjects exposed to formaldehyde in mobile homes—a brief communication. Arch. of Environ. Health. 42:347-350. [PubMed: 3439811]
  • Thrasher, J.D., R. Madison, A. Broughton, and Z. Gard. 1989. Building-related illness and antibodies to albumin conjugates of formaldehyde, toluene diisocyanate, and trimellitic anhydride. Am. J. Industrial Med. 15:187-195. [PubMed: 2543217]
  • Thrasher, J.D., A. Broughton, and R. Madison. 1990. Immune activation and auto-antibodies in humans with long-term inhalation exposure to formaldehyde. Arch. Environ. Health. 45:217-223. [PubMed: 2400243]
  • Zeiss, C.R., R. Patterson, R., J.J. Pruzansky, M.M. Miller, M. Rosenberg, and D. Levitz. 1977. Trimellitic anhydride-induced airway syndromes. Clinical and immunologic studies. J. Allergy Clin. Immunol. 60:96-103. [PubMed: 68964]
  • Zeiss, C.R., T.M. Kanellakes, J.D. Bellone, D. Levitz, J.J. Pruzansky, and R. Patterson. 1980. Immunoglobulin E-mediated asthma and hypersensitivity pneumonitis with precipitating anti-hapten antibodies due to diphenylmethane diisocyanate exposure. J. Allergy Clin. Immunol. 65:346-352. [PubMed: 6246152]
Copyright © National Academy of Sciences.
Bookshelf ID: NBK234795


Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...