Proceedings: Indoor Air 2002
ACUTE TOXICITY OF MARKING PEN EMISSIONS
R Anderson and J Anderson*
Anderson Laboratories, Inc., West Hartford, Vermont, USA
ABSTRACT
Groups of mice were exposed for one hour to the emissions of 8 brands of felt tip markers (AH). A modification of ASTM-E 981 test method was used to measure changes in respiratory
cycle parameters and diagnose Sensory Irritation (SI), Pulmonary Irritation (PI), and Airflow
Limitation (AFL). A Functional Observational Battery was used to evaluate neurotoxicity.
Pens A through F produced respiratory toxicity and neurobehavioral changes including
abnormal posture and gait, tremors, falling, and/or hyperactivity, facial swelling, severe
lacrimation, and gasping. Pens G and H were much less toxic. The TVOC concentrations used
in the chamber tests were similar to those near marking pens in actual use. These results
document that some felt tip markers emit mixtures of chemicals which can cause respiratory
and neurobehavioral abnormalities. These toxicity results explain human complaints
concerning respiratory and neurological reactions to marking pen emissions.
INDEX TERMS
Marking pens, VOCs, toxicity, mice, ASTM E 981
INTRODUCTION
Numerous complaints have been received by our laboratory concerning adverse reactions to
the emissions of marking pens. One student experienced blurred vision, severe nasal
congestion, ringing in the ears, headache, dizziness, difficulty concentrating, drowsiness,
memory deterioration, decreased fine motor coordination, difficulty breathing, nausea, and
swollen ankles after sitting in lecture halls where many students were using marking pens to
highlight lecture notes. Many of the marking pens which are commercially available are
labeled "nontoxic." This discrepancy between the labels ("nontoxic") and the human
symptoms (which suggest toxicity) led us to systematically study the acute toxic potential of
marking pen emissions.
The study design was to expose mice for one hour to the mixtures of volatile chemicals
emitted by various marking pens. The mice were monitored during the exposure with
pneumotachographs to determine whether there were any toxic effects on the upper or lower
airways. After these exposures the mice were examined and scored by a Functional
Observational Battery to screen for evidence of neurotoxicity.
METHODS
Eight brands of marking pens were purchased in local stores. All the marking pens were
labeled "nontoxic." A and B were sold for use on dry erase boards. C and G were general
purpose markers. D, E, and F were made for use in art projects. H was marketed for use on
vinyl transparencies. One or more marking pens were uncapped and placed in a 40 liter glass
chamber and allowed to equilibrate for one hr before use.
*
Contact author email: jharca@hotmail.com
641
Proceedings: Indoor Air 2002
Animal Exposures
Four male Swiss-Webster mice were positioned in a glass manifold as previously described
(Anderson and Anderson, 2000). The head of each mouse extended into the central exposure
area with the body in a side arm which served as a whole body plethysmograph. During a 15
min "baseline period" the animal exposure chamber was continuously ventilated with
charcoal-filtered air. Then the mice breathed marking pen emissions for 55 min, carried by
charcoal-filtered air passed at 6 l/min through the sample chamber, through the animal
exposure chamber, and then out of the building. Thirty minutes after the end of the exposure,
the mice were scored using a Functional Observational Battery (FOB) (Anderson and
Anderson, 1998).
12 or 16 mice were exposed to each dose of each brand of pen. Each mouse was exposed to
only one brand/dose combination. As controls 220 mice were subjected to the entire
procedure with no pen in the sample chamber (sham exposures).
Bioassay ASTM-E-981
ASTM-E-981 is a standardized toxicological test method for measuring acute biological
effects of airborne irritant chemicals (ASTM, 1984). Alarie and associates (Vijayaraghavan,
Schaper, Thompson et al., 1993, 1994; Boylstein, Anderson, Thompson et al., 1995) added
pneumotachographs to continuously measure airflow velocity during each breath of each
mouse. Digital computer programs determine the duration of pause after inspiration (TB, time
of break), the duration of pause after expiration (TP, time of pause), and the mid-expiratory
airflow velocity (VD), the respiratory rate, and the tidal volume. A diagnosis of Sensory
Irritation (SI) requires an increase of TB by more than 2 times the standard deviation (S.D.)
from the baseline mean value for that mouse that day; a diagnosis of Pulmonary Irritation
(PI) requires an increase in the TP to greater than 2 S.D. from the baseline mean for that
mouse that day; airflow limitations are diagnosed when VD falls 1.5 S.D. below the
baseline mean value for that mouse that day.
FOB
Each mouse was individually observed and scored for 25 parameters including overall level of
activity, posture, gait, tremors, balancing on the lip of the jar, facial swelling, lacrimation,
gasping, reach reflex, grip strength, and righting reflex. Statistical significance was
determined using chi-square analysis of 2x2 contingency tables with the Yates correction
factor (Zar, 1984). Frequency data was subsequently converted to percentages for tabular
presentation.
Room air TVOCs
One or four markers were uncapped and used to draw 10 cm circles on a piece of tablet paper.
The sampling probe for a flame ionization detector (Rosemont 400A, calibrated with 100 ppm
methane) was positioned 26 cm above this drawing.
Chemistry
Matrix Analytical Laboratories of Addington, Texas, used gas chromatography to analyze
head space gasses and water extracts of the felt tips for volatile organic chemicals.
642
Proceedings: Indoor Air 2002
RESULTS
Exposure TVOCs
The pens varied considerably in offgassing rates, resulting in TVOC values between 68 and
32,000 ppm in the exposure chamber (Table 1). Charcoal filtered air had a TVOC of 23 ppm.
Actual use of one or four pens in simulated art projects generated TVOC values in the same
range as those used in the mouse chamber experiments (Table 1). Chemical analysis revealed
various combinations of acetone, butanol, methanol, ethanol, isopropanol, 1-propanol, methyl
isobutyl ketone, and several other ketones, esters, and acetates.
Table 1. TVOC Values
TVOC (ppm)
Number of pens open
A
B
C
D
E
F
G
H
Mouse chamber
1
4
3,800
17,000
5,700
26,000
8,900
32,000
1,800
4,000
530
2,400
5,100
15,000
70
380
68
250
Room air near art project
1
4
2,400
11,000
2,800
14,000
4,700
25,000
1,500
5,600
420
2,100
4,200
15,000
48
240
46
160
Respiratory effects
Mice given sham exposures averaged 7% SI. Mice exposed to marking pen emissions
developed various amounts of SI (Table 2). Pen F caused the largest SI (72% of the breaths),
while Pens G and H caused almost none. As a single unit B was the most potent.
Table 2. Sensory Irritation
Percent of breaths with SI
Number of pens open
1
4
9
A
4
35
46
B
63
69
57
C
59
63
33
D
13
30
70
E
5
21
F
34
50
72
G
4
6
H
4
7
12
There were 220 shams and 12 or 16 mice in the other groups. Values are means (SE mean
varied from 1 to 10). Bold type indicates p < 0.05 compared to Shams. The Shams averaged
7% SI with SE of 1.
Relatively little PI was seen with any of the pens. Single units of pens D, E, and F caused
AFL in 10 to 17% of the breaths at the peak effects. The lowest VD recorded was 31% below
baseline in experiments with Pen D.
FOB abnormalities
High frequencies of abnormalities occurred with posture, gait, tremors, hyperactivity, facial
swelling, and gasping (Table 3). Severe lacrimination was seen with Pens A, B, and E. In
643
Proceedings: Indoor Air 2002
addition, avoidance behavior was observed in 50% of the mice exposed to Pen B, and the
hyperactivity with Pen D was so severe that 75% of these mice were recorded as "explosive."
Decreased grip strength was observed in several mice exposed to Pens D and F, and the
righting reflex was slow in several mice exposed to Pen B. Pens G and H caused the least
neurotoxicity.
Repeat exposures and FOB
Two pens were studied with repeat exposures one hour a day for three days. The FOB score
(number of abnormalities observed) resulting from Pen B increased from 12 to 14 to 18 with
repeat exposures. Increases with Pen E were similar.
Table 3. FOB: behavioral and appearance abnormalities
Percent of mice with abnormality
Pen
Sham
A
B
C
D
E
F
G
H
Posture
5
17
0
67
75
42
33
100
83
Gait
8
0
0
42
75
58
75
56
42
Tremor
17
13
92
94
50
67
75
75
50
Disorient.
2
8
6
17
12
0
0
0
33
Falling
0
6
0
8
0
25
33
31
13
Hyperact.
3
0
0
0
0
0
0
83
56
Face swell.
0
19
0
0
67
31
17
33
67
Sev. Lacrim.
0
0
0
0
0
17
37
17
33
Gasping
0
0
83
88
58
42
44
92
67
There were 64 shams and 12 or 16 mice in the other groups. Bold type indicates p < 0.05
compared to shams (chi square test). All data obtained using one pen open. Disorient. =
disorientation. Hyperact. = hyperactivity. Swell. = swelling. Sev. Lacrim. = severe
lacrimation.
DISCUSSION
All the pens tested were labeled "nontoxic;" the current data demonstrate that this is clearly
not correct. It is not known which of the chemicals present in the emission mixtures were
most responsible for the adverse effects measured; the results may be due in part to the
combined effects of the solvents in these mixtures, but other volatiles probably also contribute
to the overall effects. These acute toxic effects occurred at concentrations similar to those a
human would encounter near an art project involving one to four marking pens. These results
provide a toxicological basis for some of the complaints we have received concerning acute
toxicity of marking pen emissions.
SI results from chemical irritation of trigeminal nerves in the conjunctivae, facial skin, and
nasal mucosa (Alarie, 1973; Neilsen, 1991). VOCs which cause SI in mice cause humans to
experience burning of eyes, throat, skin, nose, or chest; conjunctivitis, lacrimation; coughing;
and/or gagging (Alarie, 1973). Thus the SI observed in mice exposed to marking pens
emissions probably explains the complaints of burning face and eyes in some humans exposed
to marking pen emissions.
Both the VD measurements used with mice and the peak flow measurements used to evaluate
human asthmatics reflect expiratory airflow velocity. The VD values in these experiments
decreased by 15 to 30% during this exposures; this is similar to the decreases in peak flow
seen in human asthmatics during bronchoconstriction episodes. Because AFL occurred
during the first hour of encounter between mice and marking pen emissions, these airflow
644
Proceedings: Indoor Air 2002
reductions probably represent acute toxic effects rather than allergic phenomena. Toxic
airflow reductions are involved in occupational asthma (Chan-Yeung and Malo, 1995). The
AFL observed in mice exposed to marking pen emissions may explain the sensation of
difficult breathing in some humans exposed to marking pen emissions.
The FOB scores are based on gross changes in behavior or appearance. Without more
invasive experimentation one can only speculate about the mechanisms of these toxic effects.
Some of the abnormalities (e.g. abnormal posture, abnormal gait, tremors, and falling)
presumably reflect neurological or neuromuscular effects. Some of the other findings (e.g.
facial swelling and severe lacrimation) might reflect local neurogenic inflammation.
Regardless of mechanisms, the FOB results in the mice demonstrate there is a toxicological
basis for the neurological complaints of some humans exposed to marker emissions.
CONCLUSIONS
Several marking pens emitted mixtures of volatile organic chemicals which caused acute
adverse effects including Sensory Irritation, airflow reductions, and neurobehavioral
abnormalities in normal mice. Use of felt tip markers can create high local concentrations of
VOCs which might be toxic to the user and to nearby observers.
REFERENCES
Alarie Y. 1973. Sensory irritation by airborne chemicals. CRC Crit Rev Toxicol 2: 299363.
Anderson RC and Anderson JH. 1998. Acute toxic effects of fragrance products. Arch
Environ Health 53: 138-146.
Anderson RC and Anderson JH. 2000. Respiratory toxicity of fabric softener emissions.
J Toxicology and Environmental Health. Part A. 60: 121-136.
ASTM (American Society for Testing and Materials). 1984. Standard test method for
estimating sensory irritancy of airborne chemicals. Philadelphia PA. ASTM, 1984.
Boylstein LA, Anderson SJ, Thompson R, and Alarie, Y. 1995. Characterization of the
effects of an airborne mixture of chemicals on the respiratory tract and smoothing
polynomial spline analysis of the data. Arch Toxicol 69: 579-589.
Chan-Yeung M and Malo JL. 1995. Occupational asthma. New Eng J Med 333: 107112.
Neilsen GD. 1991. Mechanism of activation of the sensory irritant receptor by airborne
chemicals. CRC Crit Rev Toxicol 21: 183-208.
Vijayaraghavan R, Schaper M, Thompson R, Stock MF, and Alarie Y. 1993.
Characteristic modification of the breathing pattern of mice to evaluate the effects of
airborne chemical on the respiratory tract. Arch Toxicol 67: 478-490.
Vijayaraghavan R, Schaper M, Thompson R, Stock MF, Boylstein LA, Luo JE, and
Alarie Y. 1994. Computer assisted recognition and quantification of the effects of
airborne chemicals acting at different areas of the respiratory tract in mice. Arch Toxicol
68: 490-499.
Zar JH. 1984. Biostatistical Analysis (2nd edition). Prentice Hall, Englewood Cliffs, NJ
pp 64-70
645