EPA 560/6-77-018
EPIDEMIOLOGY STUDIES
SELECTED NON-CARCINOGENIC EFFECTS
OF
INDUSTRIAL EXPOSURE TO INORGANIC ARSENIC
October 1977
FINAL REPORT
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF TOXIC SUBSTANCES
WASHINGTON, D.C. 20460
-------
EPA 560/6-77-018
July 1977
SELECTED NON-CARCINOGENIC EFFECTS
of
INDUSTRIAL EXPOSURE TO INORGANIC ARSENIC
EMANUEL LANDAU*
DONOVAN J. THOMPSON °
ROBERT G. FELDMAN+
GUY J. GOBLE°
WILFRID J. DIXON'
AMERICAN PUBLIC HEALTH ASSOCIATION*
WASHINGTON, D.C.
UNIVERSITY OF WASHINGTON"
SCHOOL OF PUBLIC HEALTH
SEATTLE, WASHINGTON
BOSTON UNIVERSITY+
MEDICAL SCHOOL
BOSTON, MASSACHUSETTS
UNIVERSITY OF CALIFORNIA'
Los ANGELES, CALIFORNIA
PROJECT OFFICER
JOSEPH SEIFTER
OFFICE OF Toxic SUBSTANCES
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C.
-------
DISCLAIMER
This project has been funded with Federal funds from the Environmental Protection Agency under contract number
68-01-2490. The content of this publication does not necessarily reflect the views or policies of the U.S. Environ-
mental Protection Agency, nor does mention of trade names, commercial products, or organizations imply endorse-
ment by the U.S. Government.
-------
Abstract
In June 1976, a study team representing four Universities and a research institution carried out a study of possible
neurological effects of long-term exposure to airborne arsenic trioxide in a Western copper smelter. The study
involved comparing active working men heavily exposed to arsenic in the work force with workers not so exposed.
Its purpose was to determine whether nerve conduction velocity can be utilized as a biological indicator of the
subtle health effects of chronic exposure to inorganic arsenic in a community setting.
The double-blind electrophysiologic and clinical study was based on 111 volunteers recruited from a selected set
of smelter workers with at least five years of high urinary arsenic levels (37 men) and five years of low urinary
levels (33 men). The control population consisted of members of the same union who worked at an aluminum
plant in the same city (13 men) as well as male municipal employees with no industrial exposure (28 men).
Demographic and other characteristics, including smoking, were obtained through use of a pre-tested questionnaire.
The parameters studied included clinical examination (history and neurologic examination) and electrophysiologic
tests of the sensory and motor nerve conduction in the upper and lower limbs. Levels of arsenic and eight other
metals were measured in the blood, hair,, urine and nail specimens collected. The study team screened the workers'
blood for carboxyhemoglobin levels, diabetes mellitus and anemia. The neurological status of the men was categor-
ized both on the clinical judgement of the examining neurologist and the objective measurements from the electro-
physiologic examination. Care was taken in the analysis not to confuse age effects with other effects.
After the study population was classified into clinical neuropathy, sub-clinical neuropathy, and normal categories,
a statistically significant association was demonstrated with the sum of the logarithms of arsenic levels in urine,
hair and nails. This association remained after adjustment for smoking, "excuses" and average age differences in
the three categories.
The conclusions of the study are that: (1) chronic arsenic exposure in an industrial setting affects the peripheral
nervous system, and (2) the neurologic parameters used in this study appear to be suitable for use in screening
a community population for changes related to arsenical neuropathies.
-------
Acknowledgments
The authors wish to thank all the workers who constituted the study population and without whose participation
the study could not have been conducted. Our thanks also go to the management officials of the Tacoma Smelter
of the American Smelting and Refining Company, particularly Armand Labbe of the Tacoma Smelter, and to
K. W. Nelson of its parent organization for their complete cooperation. The local of the International Steelworkers
Unions was of significant help and we wish to thank George Becker, National Steelworkers Union and Robert
Guadiana, local USWA Union for assisting us in getting this study started. Tom Hughes, Kaiser Aluminum
Company and Jeanne Barzar, City of Tacoma were also helpful.
Our special thanks go to the many people who helped us carry out this study. First, there is Margaret Farwell,
R.N., of the Fred Hutchinson Cancer Research Center and Tom Rawson, a doctoral candidate of the University
of Washington, School of Public Health. Then, we have the Boston University staff consisting of Daniel Sax, M.D.,
Margaret Hayes, R.N., M.S., Norman Leifer, Daniel Melbar, M.D., and Clyde Niles, M.D. Assisting in the
statistical analysis at UCLA were V. K. Murthy, Ph.D. and Coralee Yale, M.S., Vivian Henderson, M.S. of
the University of Pittsburgh was responsible for obtaining the historical data for the workers under the direction
of Phillip Enterline, Ph.D. R. M. Orheim assisted in the laboratory analysis at the University of Washington.
Ralph Wands, M.S., of the National Academy of Sciences performed an on-site visit of the study on behalf of the
APHA Technical Panel on Environmental Hazards. F. Irene Williamson, B.S., of APHA was responsible for the
timely completion of the study and participated actively in the preparation of the study and this report. The
encouragement and guidance of three EPA scientists, William Coniglio, M.S. and Robert Carton, Ph.D., formerly
with the Office of Toxic Substances, and James Everts, Ph.D. of Region X, should be acknowledged. Finally,
without the invaluable typing assistance of Shelia Mason of the Communications Center of APHA, this report
would not have materialized in its present form.
-------
Selected Non-Carcinogenic Effects of Industrial Exposure
to Inorganic Arsenic
EMANUEL LANDAU, DONOVAN J. THOMPSON, ROBERT G. FELDMAN, GUY J. GOBLE,
AND WILFRID J. DIXON
Introduction
Within the past decade, scientists have come to
accept the preeminent role of the workplace as well as
the non-occupational environment in the induction of
cancer. Accordingly, while attention has been focused in
large measure on new chemicals and compounds, at-
tempts have been made to review and update knowledge
on common substances. One of these is arsenic. In this
report, we furnish a brief review of arsenic's carcinogenic
properties emphasizing airborne arsenic, since we do not
believe they are well-known. The non-carcinogenic ef-
fects of long-term exposure to atmospheric arsenic are
virtually unknown at this time though relevant informa-
tion is available from episodes of arsenic-contaminated
water and food.
Arsenic, known and used since antiquity, is one
substance which is truly ubiquitous both from natural
and industrial sources. A current NIOSH estimate in-
dicates that in the United States alone 1,500,000 workers
in dozens of occupations are potentially exposed to in-
organic arsenic.1 The list of arsenical compounds present
in and discharged in the environment is extensive and
includes both organic and inorganic compounds. Accord-
ing to Frost in his review of arsenic and biology, in-
organic arsenicals are more toxic than the organic, and
trivalent arsenic is more toxic than the pentavalent, but
he also pointed out that for any such generalization
exceptions can be found.2
Recently, we have witnessed a significant expansion
of knowledge of the health effects of exposure to arsenic.
This has occurred principally in the area of carcinogen-
esis. The advance in our state of knowledge may be seen
by comparing the summary of the chapter on arsenic in
Volume II of the International Agency for Research on
Cancer (IARC) series with that of the National Academy
of Sciences (NAS). The IARC is an agency of the World
Health Organization (WHO).
As recently as 1973, the IARC said: "The available
stradjes point consistently to a causal relationship be-
tween skin cancer and heavy exposure to inorganic arse-
nic in drugs, in drinking-water with a high arsenic con-
tent, or in the occupational environment. The risk of
lung cancer is clearly increased in certain smelter workers
who inhale high levels of arsenic trioxide. However, the
causative role of arsenic is uncertain, since the influence
of other constituents of the working atmosphere cannot
be determined. An increased relative frequency of deaths
from lung cancer has been found in other occupational
groups exposed to high levels of inorganic arsenic com-
pounds (e.g., sheep-dip workers, certain mining and
vineyard workers). Cases of lung cancer occurring after
the medicinal use of inorganic compounds, and of liver
haemangioendothelioma following various kinds of ex-
posure to arsenic have been reported, but these may be
chance associations. No evidence exists that other forms
of cancer occur excessively with heavy arsenic ex-
posure."3
The NAS document stated: "There is some evidence
that arsenicals can be mutagenic in humans. There is
strong epidemiologic evidence that inorganic arsenic is a
skin and lung carcinogen in man."4
So, in a period of a few years, we have progressed
from a statement indicating that the main effect of ex-
posure to arsenic was an increase in skin cancer to a
statement that says there is strong epidemiological evi-
dence that inorganic arsenic is a skin and lung carcino-
gen to man.
In every case, each of the epidemiologic studies
reviewed in the NAS report (or in the critique on it)5 has
valid limitations per se. Taken together, however, they
demonstrate that occupational exposure to inorganic ar-
senicals is carcinogenic in two different human tissues.
There is some evidence at hand of excess cancer in
another human tissue. The recent study performed by
Baetjer et al., compared the mortality experience of pes-
ticide plant retirees with that of the general population of
Baltimore, Maryland, the location of the pesticide facil-
ity.8 Baetjer found even greater differences between ob-
served and expected deaths from all cancer, as well as
respiratory and leukemia-lymphatic cancer.
Unfortunately, our knowledge of the non-carcino-
-------
genie effects of chronic exposure to airborne arsenic to-
day is much more limited. Yet, significant information
about the effects of the use of arsenical inductions and
high arsenic food and drink is at hand. In the NAS
report, the effect of exposure to inorganic arsenic con-
tamination of food has been thoroughly described. Rey-
nolds in 1901 reported the clinical expression of toxicity
among 500 people consuming beer contaminated with
arsenic.7 Initial symptoms were digestive disturbances.
Complaints of conjuctivitis, rhinitis, laryngitis, bronchitis
and skin eruptions followed. Nervous system in-
volvement appeared before classical skin lesions and ef-
fects on the heart and blood capillary changes were also
noted. Implications of depressed immunological capacity
are suggested by Reynold's observations on herpes zos-
ter. The findings of Mizuta et al, reporting on accidential
poisoning of individuals using contaminated soy sauce,
confirm Reynold's neurological findings.8
The health impact of chronic community exposure
to arsenic water has been addressed through studies con-
ducted in areas where geological pecularities have
created elevated levels of arsenic in the drinking water
used for a protracted period of time by the inhabitants.
The findings are striking in that many diseases similar to
those found in occupational exposure, were found in-
volving skin changes and skin cancer. Bronchopul-
monary involvement parallels the findings with medici-
nal arsenic.
In 1967, Tseng et al. reported on a survey of 40,421
inhabitants of Taiwan, where artesian water containing
up to 1.8 ppm arsenic had been used as a source of
drinking water for 45 years.9 Hyperpigmentation, kera-
tosis and skin cancer were all reported among the popu-
lation as was a peripheral vascular disorder resulting in
gangrene known locally as Blackfoot disease. Individuals
using drinking water containing above 0.6 ppm arsenic
experienced skin cancer rates approximating five times
the rate for individuals of the same age and sex living in
communities where the arsenic content was 0.3-0.6 ppm.
In 1974 Chuang, reporting on the relationship of Black-
foot disease and arsenic levels in drinking water, pre-
sented indications that this disease manifestation had
occurred in individuals consuming drinking water con-
taining arsenic levels as low as 0.35 ppm.10
In 1964, Bergoglio reported on the cancer mortality
experience of inhabitants of a province of Cordoba in
Argentina where natural soil arsenic levels resulted in a
contaminated drinking water supply." Although it had
long been recognized that individuals living in this area
experienced a high rate of skin lesions on the palms and
soles as well as skin cancer, these observers noted that
increased mortality from respiratory and visceral cancer
was also occurring.
In 1971, Borgono and Greiber published the results
of studies for the city of Antofagasta, Chile where the
inhabitants had been consuming drinking water contain-
ing 0.6-0.8 ppm arsenic for a 12 year period.12 Com-
parative studies with neighboring populations revealed
that persons from Antofagasta having elevated levels of
arsenic in their hair experienced an increased frequency
of abnormal skin pigmentation, keratosis, chronic herpes,
bronchopulmonary disease, chronic cough, abdominal
pain, and chronic diarrhea, as well as cardiovascular
manifestations including Raynaud syndrome, and aero-
cyanosis. One striking finding was the incidence of vascu-
lar disease and repeated episodes of pneumonia with
bronchiectasis in children. Children may respond differ-
ently to arsenic exposure than adults.
Another recent drinking water study was that con-
ducted in Yellowknife, Northwest Territory, Canada.18
There did not appear to be a striking increase in the
incidence of skin cancer. However, the report stated: "A
comparatively large number of skin lesions were found
on examination. These were notable for the number of
cases of psoriasis, scaly dermatitic changes, eczematous
dermatitis and a number of rashes around the naso-labial
folds." It also said that clinical examination uncovered a
large number of neurological findings. These neurologi-
cal findings include "demonstrable loss of sensation,
weakness, etc., and only exclude cases of simple
"Tremor" when unaccompanied by other findings." The
number appeared to be excessive but their significance
could not be appraised as no control population was
available. We should also like to point to the toxicologic
work on chronoxie by Rozenshtein reported in 1969, as
an example of the limited research on the neurological
properties of arsenic.14
In order to shed some additional light on this issue,
we undertook a study of possible neurologic effects re-
sulting from exposure to arsenic trioxide in June 1976. In
addition, the May 1977 Interim Report of the Canadian
Public Health Association Task Force on Arsenic made
reference to the possible need for "other areas of medical
surveillance of workers."15
Selection of Study Site and Study Protocol
A significant point source has been identified in the
continuing dissemination of arsenic in the environment,
namely a copper smelter using high arsenic ores. How-
ever, while the carcinogenic effects on retired workers at
the smelter have been demonstrated,18 no similar data
are as yet available on the morbidity patterns of the
community residents though the burden of arsenic as
measured in urine and hair has been found to be ele-
vated.17 The stigmata of high arsenic exposure such as
dermatitis, perforation of the nasal septum, con-
junctivitis, pigmentation and keratosis have not been
reported in the community thus far despite newspaper
and other publicity given to the emissions of arsenic from
its smelter. Accordingly, it appeared necessary to rely on
a subtle measure of possible neurological damage.
This measure, the peroneal nerve conduction veloc-
ity test, had been performed on children age 5 to 9
exposed to lead in Kellogg, Idaho, with measurable dif-
ferences indicated between highly lead-exposed and less
exposed children.18 We believed that the finding of al-
tered nerve impulse conduction velocity (NICV) in con-
firmed cases of arsenical exposure would be an important
documentation of possible adverse effect on the nervous
system in such individuals.
-------
In 1975, we approached a group of faculty members
at the School of Public Health and Community Medi-
cine, University of Washington about their interest and
ability to carry out a study of the hazards arising from the
Tacoma Smelter of the American Smelting and Refining
Company (ASARCO) to the health of residents of Pierce
and King Counties. The community study would involve
sampling of a stratified random sample of clusters of
households in the two counties, and would also involve
measurement of arsenic (and possibly other metals) in
tissues of residents of these households and in the imme-
diate environment of their residence in soil, air, dust, etc.
A subsample of this group of residents identified by
questionnaire responses as having symptoms or health
conditions potentially ascribable to the exposure to arse-
nic were to be invited to have a careful physical examina-
tion by a physician trained to look for adverse manifesta-
tions of arsenic exposures. Among the possible health
effects of arsenic exposure to be looked for was reduced
nerve conduction velocity, an effect known to be associ-
ated with relatively heavy exposure to lead.
Before undertaking an extensive epidemiologic
study of the exposed community population it was de-
cided that a limited study of workers should be done,
including some with exposures higher than those of the
general population, in order to determine if NIVC was
adversely affected by arsenic exposure and if the tech-
nique were sensitive enough to reveal changes in the
exposed public. Accordingly, the same faculty group was
thus asked to plan a pilot study involving working men
from the Tacoma Smelter and a comparison group of
working men not exposed to arsenic in their workplace.
Through the cooperation of the management of the
Tacoma Smelter, the union representing its and other
employees and officials of the City of Tacoma, we pro-
posed a plan with the following general outline.
ASARCO would furnish rosters of approximately thirty
men in each of two categories: men working in plant
locations where the arsenic exposure was high and
roughly comparable men whose exposure was known to
be low. A third group of men for comparison would be
sought from members of the same union employed at an
aluminum plant in an area in which no arsenic exposure
existed. This third group was ultimately expanded to
include men employed by the City of Tacoma in non-
industrial settings.
Each volunteer from these groups was to have his
nerve conduction velocity tested for one or more nerves
and provide samples of blood, urine, hair and nails for
laboratory determination of content of arsenic (and other
metals). The resulting data were to be analyzed to deter-
mine the association (if any) of increased levels of arsenic
in the tissues studied and reduced nerve conduction ve-
locity. This study as stated previously, was conducted in
June of 1976.
An account of the study and the findings of the
resulting analyses constitute the substance of this report.
Selected sections of the report depend entirely upon the
contribution made to the study by two groups sub-
sequently separately contracted to do the neurological
investigation and the laboratory determinations; the
neurological group headed by Robert G. Feldman, M.D.,
Chairman of the Department of Neurology, Boston Uni-
versity, and the laboratory team headed by Guy J. Goble,
Ph.D., Department of Environmental Health, University
of Washington. The data processing and the preliminary
statistical analysis, as well as the supervision of the field
word, were the responsibility of Professor Donovan J.
Thompson and colleagues from the Department of
Biostatistics, University of Washington. Additional and
more detailed statistical analysis were the responsibility
of Professor Wilfrid J. Dixon and members of his staff
from the Department of Biomathematics, University of
California at Los Angeles. Phillip Enterline, Ph.D. of the
University of Pittsburgh was responsible for the work
history records used in the study.
Study Subjects
FIELD WORK
From preliminary meetings with representatives of
the plant management and the union local, it was agreed
that a group of approximately thirty men employed in
areas of the smelter with high exposure to arsenic would
be identified. These men and a comparison group of
about the same size, matched as nearly on age as pos-
sible, would be invited to participate. Subsequently, the
high exposure group was further defined to be men who
had had urinary arsenic levels of 200 ug/liter or more for
at least the past 5 years. Later, the management agreed
to increase the roster to fifty men in each category from
which 37 and 33 volunteered to participate, respectively.
In an attempt to ensure a comparison group with no
exposure to arsenic in the workplace, which yet would
have comparable other characteristics, efforts were made
to enlist volunteers from members of the same union that
represent the ASARCO employees who worked at an
aluminum plant in Tacoma. When a sufficient number of
volunteers from this source appeared to be unlikely, an
appeal for support was made to officials of the City of
Tacoma to canvass their male employees with no indus-
trial experience for volunteers for the study. The two
sources ultimately furnished 13 and 28 participants, re-
spectively. The total study group thus consisted of 70
men from ASARCO and 41 non-ASARCO men. An addi-
tional 6 persons participated in the study for various
reasons (5 males and 1 female) but have been excluded
from most analyses, since they were not recruited from
the groups to be compared (the study Director, an inter-
ested professor-observer, two EPA observers, the study
interviewer, a son of a planned participant who could not
attend due to illness). Men volunteering to participate
were furnished transportation to and from the study site
and paid $10 for their effort.
Study Facility and Procedures
The study was conducted on June 9, 10 and 11, at a
convenient central location, the Tacoma Labor Center in
downtown Tacoma. Figure 1 in the Appendix is a sketch
of the study space provided in this air conditioned, mod-
3
-------
1. Interview- Phyticttnind RictptionntTurn
2. Tiblt (or iptcitMn collfction
3. NiuralogiaKCIininDEnminition
4. Nem Conduclioii toting
S. FonrnChicted
FIGURE 1. Layout of Study Space
ern facility. Men were scheduled for their appointments
in a manner that ensured that ASARCO men were thor-
oughly intermingled with men from the other two
sources during each portion of the two study days. A
participant upon arrival was seated in the hall near Parlor
B until study personnel were ready to start him through
the routinized study procedures. Upon entering Parlor B
a receptionist-physician team interviewed the participant
using the questionnaire appended (Appendix A) and ac-
cepted the urine specimen he brought with him. This
specimen was from the first voiding upon arising in the
morning for most of the participants. A few participants
not on the day shift brought specimens obtained at other
times. Following the interview, the participant was asked
to disrobe to the waist and remove his shoes and stock-
ings. He next was sent to a station where a nurse took
samples of his toe and finger nails (approximately 100
mg.) and a swatch of hair from the back of his head
(approximately 50 mg.). Each participant carried with
him a sheet of computer labels with* his identification
number printed on each, one of which was affixed to his
questionnaire, his urine specimen bottle, and to each
subsequent specimen or form pertaining to him. Next, a
blood specimen of 25 cc's was drawn by a second nurse.
Three vacutainers were obtained for each man: 10 cc's
for metal determinations; 10 cc's for the SMA-18 battery
and 5 cc's for carboxyhemoglobin and hematocrit deter-
minations. Each container was labelled and taken to the
laboratory station (Board Room) where the bloods were
refrigerated and transported regularly, either to the labo-
ratory of Tacoma General Hospital (within a mile of the
study site) for the SMA-18 studies, or to the laboratory at
the Department of Environmental Health, University of
Washington, Seattle, for determination of content of the
nine metals under investigation: antimony, arsenic, cad-
mium, copper, lead, mercury, nickel, selenium and zinc.
Hematocrit determinations and specific gravity of the
urine specimens were made immediately by the labora-
tory team in the Board Room.
Neurological Studies
The neurological examination and the nerve con-
duction testing were conducted in Parlor A. To insure the
blinding of the place of employment, all volunteers were
asked not to mention where they worked during the
examination. Moreover, there was no indication on any
of the appointment rosters as to whether the man came
from ASARCO, the aluminum company, or the City of
Tacoma. Scheduling, as mentioned, interspersed men
from the three locations and was not discussed with the
neurologist. Men appeared for the examination in their
street clothes, barefoot and stripped to the waist. Two
neurologists alternated in conducting the clinical neuro-
logic examination. Again, to eliminate the possibility of
the clinical examination influencing the determination of
the nerve conduction velocities, the neurologist con-
ducting the clinical examination influencing the determi-
nation of the nerve conduction velocities, the neurologist
conducting the clinical examination did not assist in the
electrical studies for that man. Inadvertent information
(if any) acquired during the clinical examination through
conversation with the men could therefore not influence
the velocity or amplitude measurements. During the
nerve conduction testing no conversation between physi-
cian, technicians and the volunteers took place. From
observation of the entire procedure the blinding seemed
to be secure. The results of the clinical examination were
recorded on a standard neurological form (Appendix B).
A history of metabolic disturbances, nutritional distur-
bances, trauma, alcohol consumption and medication
was obtained and then a full neurological examination
was carried out.
When the clinical examination was completed the
participant passed to the nerve conduction testing station
near the exit door in Parlor A.
For the electrophysiological examination, sensory
and/or motor nerve conduction studies were carried out
on the following nerves: 1) the right ulnar nerve-sen-
sory/motor 2) the right common peroneal nerve-motor;
and 3) the right sural nerve-sensory. Both conductions
and amplitudes were measured for each nerve using
standard techniques, with surface electrodes.19"28
Two electromyographs (Teca 4), each equipped
with a digital averager, were used for the study. A supra-
maximal stimulus was delivered on each occasion with a
duration of 0.1-0.2 millisec. and a potential of 100-300 V.
The sensory action potentials for both the ulnar and sural
nerves were averaged 32 times in every case. The room
temperature was kept at 74-75° F for all recording ses-
sions. In addition, surface skin temperature was also
monitored in every case throughout the study, using a
specially adapted telethermometer, manufactured by the
Yellow Springs Instrument Company. The skin temper-
ature was maintained at an average of 33° C.
At the completion of the electrophysical examina-
tion, the participant dressed, reported with his com-
pleted forms to the kitchen where the completeness of all
materials relating to the study was checked. Payment for
his participation occurred at this time and transportation
back to work or home was offered.
On the first day of the study, participants were asked
if they would volunteer for a repeat nerve conduction
testing session on the following day. This invitation was
offered systematically to each participant until ten volun-
-------
teers had been identified. These men returned on Friday,
June 3, and were worked into the schedule for nerve
conduction testing only. The test re-test results are pre-
sented later in this report.
The field work was completed on schedule on Friday
afternoon, June 11, without incident.
LABORATORY PROCEDURES AND DATA
PREPARATION
Laboratory Procedures
One vacutainer containing 10 cc's of blood for each
participant was used for the SMA-18 battery which was
run in the laboratory of Tacoma General Hospital. The
results for this battery were reported on the appended
from (Appendix C). These data were edited, key
punched, and entered into the GA-16 computer of the
Department of Biostatistics, University of Washington.
The other bloods, the urine sample, and the hair and
nail samples were processed by the laboratory of the
Department of Environmental Health, University of
Washington. A brief description of their procedures fol-
lows:
Cleaning of Glassware
Urine specimen bottles were thoroughly washed and
rinsed then given a final rinse with purified nitric acid.
This final rinse was poured out and the bottle stoppered.
Blood specimens were collected in tubes which were
certified free of heavy metals, while hair and nails were
collected in plastic bags.
Cleaning of Samples
Nails and hair were weighed into a clean digestion
vessel. These specimens were washed with a 1% Triton
X-100 solution and placed in an ultrasonic bath for 10
minutes.24"26 These were then rinsed twice with distilled
water and drained.
Digestion
Specimens were digested for mercury analysis by
heating with concentrated sulfuric acid at 50-60° C for 30
minutes.28 These were then soaked overnight with 6%
potassium permanganate. The samples were taken to
volume with 1 % nitric acid.
Blood, urine, hair and nails used in the analysis of
the remaining metals were digested with 0.1% sulfuric
acid or perchloric acid at 50-60° C for 30 minutes.27 These
were then soaked overnight with 6% potassium per-
manganate. The samples were taken to volume with 1 %
nitric acid.
Extraction
\ This technique was applied to blood and urine spec-
imens for the analysis of nickel, antimony, selenium,
cadmium, zinc and lead.28"29 The digestates were ad-
justed to pH 2.7 with 10% NH4OH and extracted with
0.4% ammonium pyrrolidine dithicarbamate (APDC) in
10 ml of methyl isobutyl ketone (MIBK).
Analytical Methods
A. Arsenic in Blood and Urine30
The acidified digestate was treated with sodium
borohydride, the liberated arsine gas was swept with
argon into the heated graphite furnace of the
Atomic Absorption Spectometer (AAS).
B. Arsenic in Hair and Urine81"82
A 10:1 aliquot of the digestate was pipetted
into the graphite furnace for ASS analysis.
C. Mercury in all substrates38 *
A aliquot of the digestate was neutralized with
20% hydroxylamine, then treated with 1.0 ml of
20% stannous chloride. The reduced mercury was
swept into a quartz ended cell with a 1350 ml/min
air flow and measured by cold vapor technique in
AAS.
D. Nickel, Antimony, Selenium, Cadmium, Zinc, and
Lead in Blood and Urine81"82
A volume of 10:1 of APDC chelate in MIBK
was pipetted into the graphite furnace for AAS anal-
ysis.
E. Zinc and Copper in Hair and Nails
The final digestate was aspirated into a flame
for AAS analysis.
Quality Control
A. Blind Controls84
Ten percent of all samples were split into mul-
tiple samples and assayed as separate samples, not
known to the laboratory technician.
B. Reagent Blanks84
Two percent of all analytical samples were dis-
tilled water and treated in all respects (digestion,
extraction, etc.) as if they were regular samples.
These samples might be spiked with appropriate
metals.
C. Standard Curves.
Working standards were prepared fresh daily
and results compared to an existing analytical con-
centration: Absorbancy curve. Results must be
within a standard deviation of each working stan-
dard before analyses can proceed.
D. Analysis.
Standard curves were prepared in duplicate at
the beginning of each metal: substrate set and at the
beginning of each day; these were compared to the
working curve. Samples were run in groups of ten,
the tenth specimen was assayed twice, an appropri-
ate level standard was also run at this time. Standard
curves were repeated at the end of the working day
and at the end of a metal: substrate set.
E. Data Reduction.
-------
Instrument noise levels were determined, twice
this value was used to calculate the minimum de-
tectable level from the regression line curve. The
metal concentration of substrate was determined by
substracting the reagent blank from the concentra-
tion as derived peak height reading.
All calculations were rechecked by a dis-
interested party to minimize the likelihood of errors.
For each metal: substrate set the precision for both
the blind duplicates and sample replicates have
been summarized and are presented in Appendix D.
A summary of the "normal values" for the various
metal: substrate combinations gleaned from a search of
the literature together with the mean ± two standard
deviations is reported in Appendix E. Additional data
regarding the laboratory procedures and quality control
efforts of the laboratory director are also included as
Appendix F.
Data Preparation
The questionnaire results for each participant were
coded and entered into the G.A. Computer. The neuro-
logical examination resulted in each participant being
classified as normal, having a sensory neuropathy, a mo-
tor neuropathy or both a sensory and motor neuropathy
by criteria established by the consulting Boston neuro-
logical group. A copy of their report, case by case, is
included as Appendix G. Where a condition unrelated to
industrial exposure was thought by the neurologist to be
potentially implicated as the reason for neuropathy (or
slowed nerve conduction) it was indicated on this listing.
Again these data were coded and entered into the com-
puter.
The basic nerve conduction measurements for each
participant are included as Appendix H. These data also
have been entered in this form into the computer. For
reasons that are not clear the neurological information
for two participants was lost during the processing of the
material in Boston. Normal values for the electrical pa-
rameters from the experience of the consulting neurolog-
ical group are included as Appendix I.
Finally, through the joint efforts of Dr. Phillip En-
terline at the University of Pittsburgh School of Public
Health and ASARCO officials, a set of work history rec-
ords and past urinary arsenic and lead determinations
were furnished for all participants from ASARCO. These
records were summarized through a coding scheme that
recorded for each man the years of employment in the
smelter, year of first urinary arsenic determination, total
number of such determinations, minimum, and average
urinary arsenic value, and the maximum past urinary
lead value. A copy of the basic record is appended (Ap-
pendix J-a-b).
ANALYSIS
Characteristics of Participants
In planning the study, a number of issues were
raised relating either to non-industrial causes for reduced
nerve conduction (or neuropathies) or the exposures
(principally to arsenic) from sources other than the work
place. These issues resulted in framing a series of ques-
tions included on the questionnaire and/or asked during
the neurological examination.
An examination of the responses from the total
group of participants can eliminate a number of these
issues from further consideration. Ingestion of shellfish in
the 36 hours to prior examination was reported by only 2
of the 117 participants. No participants reported taking
anti-convulsants, anti-tuberculosis drugs, or medications
containing arsenic. Only 2 men taking medicine for dia-
betes, 2 taking hormones, 3 taking thyroid medication,
and 2 taking tranquilizers. As causes for any differences
in either exposure to arsenic or nerve conduction abnor-
malities among the groups of workers to be compared
these may safely be dismissed from further consideration.
Somewhat more frequent reports of other conditions
occurred: 15 cases of recent trauma, 12 men on anti-
hypertension medication; 15 with substantial weight loss
in past three months, 7 with history of liver disease, 27
who obtain their home drinking water from wells (poten-
tially a source of arsenic) and 13 who admitted to current
gardening activity involving spraying with insecticides
(again, a potential exposure to arsenic). Examinations of
the distribution of these occurrences among the four
worker groups revealed that there were no concentra-
tions in a single group that would be likely to disturb the
group comparisons.
Two other characteristics potentially affecting group
comparisons, alcohol consumption and cigarette smok-
ing, were far more frequently reported. Most of the
participants drink alcohol (102) and 46 admitted to hav-
ing consumed alcohol in the 48 hours prior to examina-
tion (despite being asked not to). Cigarettes are currently
smoked by 61 of the men. Examining the distribution of
men with these characteristics among the employer
groups showed association of these characteristics with
place of employment, a result that might have been
expected.
A somewhat similar situation prevails with respect to
age. Here, although the two ASARCO groups (high and
low arsenic exposure) are comparable on age—an aver-
age age of 47 years in both groups with range 26-60 in
the high and 22-62 in the low exposure group—they
differ markedly from both of the comparison groups
which average 39 years of age with a range from 22-58.
Since it is also known that [the amplitude of the response
to the electrical stimulus used in] the nerve conduction
testing is slightly age dependent, and some evidence
exists for velocities also being age dependent, this differ-
ence is troublesome in subsequent analyses. However, it
may be noted here that neurological differences between
the ASARCO employees and the other two groups can-
not be accounted for by the age difference above, and
that neurological pathology is not confined to the older
workers in either the ASARCO or the comparison groups.
This problem will receive further attention in subsequent
sections of the report.
A final variable whose potential impact would be to
narrow the average exposure to arsenic and other metals
-------
* ASARCO Hi
A ASARCO Lo
• Comparison Group
AUBURN
FIGURE 2. Residence Location of Study Participants
between the ASARCO employees and the comparison
group, is place of residence. If the micro-environment of
a person's residence were contaminated by effluent from
the smelter stack, presumably this would be detectable
by examining levels of arsenic in the tissues of men not
exposed in their workplace, and then relating average
levels in groups of men to distance and direction of their
residences from the smelter stack. The role of place of
residence and exposure to arsenic as measured by tissue
levels has been examined. In Appendix Figure 2 the
residence location of the 111 participants has been iden-
tified. The residence location of men from ASARCO Hi
and ASARCO Lo groups are indicated by solid dots while
those of the aluminum and Tacoma workers are open
Circles. Clearly, some of the comparison group men live
hear the smelter and may exhibit higher arsenic levels in
their tissues than those living at a distance from the
smelter stack.
TABLE 1—Classification of Men According to the Location of
their Residence Relative to the Smelter Stack
Place of Employment
Urinary/ppb)
Arsenic ASARCO Hi
Tacoma &
ASARCO Lo aluminum
Area
A&B
C&D
E&F
G
Hair(PPm)
A&B
C&D
E&F
G
12
9
12
4
12
9
12
4
867
1465
891
524
2380
1388
2658
719
9
10
10
4
9
10
10
4
152
223
85
170
76
68
207
108
8
15
9
9
8
15
9
9
33
3
19
118
9
4
5
5
-------
TABLE 2—Current Urinary Arsenic Values (PPb) by Place of Employment
Group
ASARCO Hi
ASARCO Lo
Comparison Group
Total
None
Detectable
1
14
33
48
<10
2
6
2
10
10-19
2
3
4
9
20-29
3
7
1
11
30-49
7
2
0
9
50-99
11
0
1
12
>100
11
1
0
12
TOTAL
37
33
41
111
In Appendix Table 1 we have classified the men
according to the location of their homes relative to the
smelter stack as indicated on Appendix Figure 2.
The number of the men in the cells of the table is
quite small. However, there is no indication of a pattern
of increasing arsenic levels in either hair or urine with
proximity to the smelter stack.
The conclusion from this examination is that, with
the possible exception of age, smoking, and residence,
those potentially confounding factors which were valid
concerns in planning the study seem not to be important
in considering arsenic exposure and neurologic deficit
associations.
Exposure Results
Principal interest in this study was directed at exam-
ining the relationship of exposure to arsenic and neuro-
logical abnormalities. Of necessity, the measurement of
this exposure had to be limited to the present for the non-
ASARCO workers since no historical records existed for
this group. For the ASARCO participants, company rec-
ords were utilized to obtain information on past arsenic
exposure and some limited information on lead. The
current information on exposure to metals was expanded
to include antimony, cadmium, copper, lead, mercury,
nickel, selenium and zinc. The amounts of these metals
51Ch
405-
^ 301-
I
a
209-
20 196 372 548
Micrograms per liter
724
900
FIGURE 3. Association of Current Urinary Arsenic Values (PPb)
and Average of Past Values (Micrograms per liter) for Asarco
Participants.
in urine, blood, hair and nails are the basis for classifying
the exposure status of each worker.
Probably the simplest way to indicate the relevance
of particular tissue-metal combinations for the purposes
of this study is to tabulate the number of cases for each
such combination in which a known non-zero quantity
was recorded by the laboratory.
Clearly, if a metal-tissue combination could not be
detected in more than half of the participants, and if
those with positive results are scattered throughout the
four employer'groups with no quantitative pattern, little
will be learned from more formal analysis. On this basis,
no further examination of antimony, nickel or selenium
in urine; the same three metals plus mercury, cadmium
and arsenic in blood; selenium in hair; and mercury,
selenium, cadmium, nickel, antimony and copper in nails
will be made.
Since principal interest relates to arsenic, a summary
of these findings will be presented first.
1. Arsenic Exposure. Current arsenic exposure as mea-
sured by urinary arsenic values for all participants is
summarized in Appendix Table 2 and Appendix Fig-
ure 3.
From Appendix Table 2 it seems clear, as ex-
pected, that workers with the higher urinary arsenic
values are concentrated in the ASARCO Hi group.
Only two men from the ASARCO Lo, and Kaiser
Aluminum groups had values greater than 50 PPb;
respectively, 64.2 and 167.0. The average value for
the ASARCO Hi group, on the other hand, was 98
PPb with 22 of the 37 men having values greater than
50 PPb. It should also be noted that no arsenic was
detected by the laboratory procedures employed in
the urines of 48 of the men tested.
Examining the urinary arsenic values from the
records furnished by ASARCO, we can summarize the
differences between the ASARCO Hi and ASARCO
Lo groups in several ways as shown in Appendix
Table 3.
TABLE 3—Historical Urinary Arsenic Values for ASARCO
Participants
Micrograms per liter
Group
Average
Average of Average of Years
Average Maxima Minima Employed
ASARCO Hi
ASARCO Lo
378.5
74.1
822.2
122.8
136.1
41.2
14.5
.16.4
-------
25.0-1
20.4-
15.3-
10.2-
5.1-
TABLE 5—Current Lead Values in Urine, Blood, Hair and Nails
by Place of Employment
20
30
40
50
60
70
Age
FIGURE 4. Ulnar Sensory Amplitude by Age (Control Croup Only).
Again, it is clear that these two groups differ
markedly in their past exposure to arsenic. Thus, we
may conclude that the volunteers for the study do
exhibit current and past differences in urinary arsenic
values, with men having high values belonging exclu-
sively to the ASARCO Hi group.
It is interesting to examine the correlation of the
historical and current urinary arsenic values. The
product-moment correlation coefficient in Appendix
Figure 4 is approximately one-half. If the two obvious
outliers are eliminated, this coefficient changes to .67
showing a reasonably strong association of past and
current values. On the other hand, either through
better protection for the men, modification of the
work environment, seasonal variation or systematic
laboratory differences in measuring these values, the
current levels of urinary arsenic are related to past
levels by the equation:
Current Level of Urinary Arsenic
= 5.9 + 0.227 (Past Level) of Urinary Arsenic
showing an apparent marked overall reduction in ex-
posure. Roughly speaking, current levels appear to be
only 23% of former levels.
No past arsenic values are available for other
than urine, so the comparison of arsenic in hair, nails
and blood must be limited to the current study results.
As indicated earlier, arsenic was detected by the labo-
ratory in the bloods of only seven of the 111 men.
Group
ASARCO Hi
ASARCO Lo
Comparison
Group
\
Urine
(PPb)
19.8
12.8
7.8
Blood
(PPb)
207.9
105.5
120.2
Hair
(PPm)
132.9
93.1
59.4
Nails
(PPm)
13.0
6.7
3.1
Four of these men were in the ASARCO Hi group,
two in the aluminum and one in the Tacoma group,
with the highest values being 80 PPb, occurring in a
man with no arsenic detectable in his urine and nails
and a value of 1.0 in his hair. Thus, blood arsenic is
not a useful measurement in this study. For the hair
and nail samples, we have the results shown in Ap-
pendix Table 4.
Again, these results clearly distinguish the groups
and add an element of exposure duration to the com-
parison, recognizing that an arsenic exposure today
would show up in hair and nail clippings some weeks
or months later.
2. Exposure to Other Metals.
Lead; Since lead exposures have been associated
with neurological deficits in other studies, it is of
interest to examine the differences among employee
groups for this metal. Appendix Table 5 summarizes
the findings.
The pattern of this table is high values of lead in
all the four tissues for the ASARCO Hi groups; inter-
mediate values for the ASARCO Lo; and lower values
for the comparison group except for blood.
Copper: The comparison of the employer groups
for the copper in urine and blood is shown in Appen-
dix Table 6. The conclusion from the values in this
table is that copper in blood and urine does not differ-
entiate the employer groups.
The hair specimens gave quite different results,
as shown in Appendix Table 7. The high recovery of
copper from the hair of the ASARCO Lo group should
be noted.
Other Metals: The results for the other metals in
the nail specimens are shown in Appendix Table 8.
Generally, these results are typical of what is
found for these metals in the other tissues; one or both
of the ASARCO groups have the highest values.
A summary of the results for metals found in the
tissues of the men from the four places of employment
might be that one or the other or both of the ASARCO
TABLE 4—Average Arsenic Values (PPm) In Nails and Hair by
Place of Employment
Group
Nails
ASARCO Hi
ASARCO Lo
Comparison Group
72.8
21.1
1.2
Hair
182.6
8.9
0.6
TABLE 6—'Current Copper Values (PPm) in Blood and Urine by
Place of Employment
Group
ASARCO Hi
ASARCp Lo
Comparison Group
Blood Averages Urine Averages
0.99
0.93
0.96
.072
.054
.043
-------
TABLE 7—Current Copper Values (PPm) in Hair by Place of
Employment
TABLE 8—Current Values for Cadmium, Nickel, Zinc in Nails
by Place of Employment
Group
Hair
ASARCO Hi
ASARCO Lo
Comparison Group
102.6
94.5
10.2
groups tended to have the highest mean values. Also,
large differences between ASARCO Hi employees
and the other men exist for arsenic and lead.
Neurological Findings
As described in Section II, both a clinical neurological
examination and objective electrical measurement were
TABLE 9— Electrical Test-Retest Results for 10 Participants
NERVE'
Group
Cadmium
ASARCO Hi 0.51
ASARCO Lo 0.11
Comparison Group .019
Nickel
16.6
4.6
2.7
Zinc
105.6
108.2
92.8
used to assess the neurological status of each participant.
Since our interest focuses on the question of whether
there are neurological deficits for the participants em-
ployed in the ASARCO Hi group, it is necessary to
continue to examine both kinds of neurological measure-
ments i.e. clinical signs of neuropathy and elec-
trophysiological parameters in those at risk.
ULNAR
MOTOR
V
Parti- Wrist-
cipant elbow
1 57
1 56
,, 55.8
^ 52
o 58
3 55
4 48
4 58
5 54
S 58
c 56
6 63.5
7 62
' 61
8 64
8 56
9 54
9 57
10 S?
avgdiff -3.1
avg S 4.0
average 57.0
coeff. of
variation 7%
Elbow-
axilla
49
57
53
59.5
70
60
57
50
66
57
71
59
62
67
56
59
64
57
51
60.5
1.6
5.7
59.3
10%
MAP
DUR
13.2
12
9.6
12.5
13.4
13.2
15.5
13
12.9
10.7
14.6
12
12.7
9
12.8
10
12
12
14.3
13.9
1.3
1.6
12.5
12%
wrist
7.5
10
10
2.4
13.5
10.5
8.5
8.5
11.5 *
10.0
2.2
6.5
5
13
8.5
5.2
8
3
15.5
15.5
0.5
3.1
8.7
36%
A
elbow
7
10
10
2.4
12
9.2
8.2
8.5
10
9.5
1.8
6.7
3.4
12.5
8.7
4.4
6.5
3.2
15.5
15.5
0.1
3.2
8.2
39%
SENSORY
PERONEAL
MOTOR
SURAL
SENSORY
A
ankle
6.5
10
9.5
2.2
12
9.2
8.6
7.5
10
9
1.8
5
4
12
6.5
4.4
6.5
3.2
14.5
14.5
0.5
3.0
7.8
39%
V
54
52
57.8
47.5
52
30.5
45
42
61
63
56
70
55
47.5
49
55
47
40.5
43
48
4.2
8.2
50.8
16%
A
14
15
12
10
8
8
7.5
5
12
14
9
9
11
24
6
4
13
3.5
14
8
0.6
4.0
10.4
39%
V
46
50.5
45.8
48
44.6
46.8
47
48
43
46
47
48
50
49
55
52
38
34.5
45
43
-0.4
1.8
46.3
4%
ankle
8
2.5
5.4
4.8
2.0
1.5
4.6
4.6
4.4
9.5
3.7
1.8
3
5.6
9.2
5.6
3.4
2
10
8.8
0.7
2.0
5.0
40%
knee
6
2.5
5.4
4.4
2.6
2
5.3
6.0
3.4
7.5
3
1.2
3
5.6
9.5
5.2
2.6
3.4
9.5
7.8
0.5
1.8
4.8
38%
V
42
36.8
43.7
43
43
41
42
36
45
36.7
45
40
48
48
42
42.5
41
33.5
44
42
3.6
3.3
41.8
8%
A
8
5
6
18
10
15
9
5
12
21
9
9
18
14
6
22
11
9
7
13
-3.5
5.4
11.4
47%
V = Velocity
A = Amplitude
10
-------
TABLE 10—Revised Electrical Test-Retest Results for 10 Participants
ULNAR
MOTOR
V
Parti- Wrist-
cipant elbow
, 129 57.4
1 123 57.0
„ 30 57.0
^ 67 52.5
14 58.0
J 76 57.5
. 20 48.5
4 62 47.5
_ 42 53.9
5 85 58.3
R 58 57.0
° 84 55.6
78 59.5
' 97 58.5
„ 104 63.0
0 55 56.4
Q 105 54.5
s 53 53.9
in 45 55.0
10 26 56.7
Ave.
Diff. 1.0
Avg.s 2.1
Average 55.9
Coeff.
of
Variation 4%
Elbow-
axilla
57.0
57.2
56.5
57.0
70.0
60.0
59.0
48.9
64.0
58.0
71.5
59.1
64.5
61.5
50.0
59.4
60.5
60.5
52.5
64.0
1.1
5.6
59.6
9%
MAP
DUR
13.2
12.0
9.6
12.5
13.4
13.2
15.5
13.0
12.9
10.7
14.6
12.0
12.7
9.0
12.8
10.0
12.0
12.0
14.3
13.9
1.2
1.5
12.5
12%
wrist
7.5
10.0
10.0
12.0
13.5
10.5
8.5
8.5
11.5
10.0
20.0
10.0
12.5
13.0
8.5
13.5
8.0
5.0
15.5
15.5
0.8
2.8
11.1
25%
A
elbow
7.0
10.0
10.0
12.0
12.0
10.0
8.5
8.5
10.0
9.5
18.0
13.0
10.0
12.5
8.5
11.0
6.5
4.8
14.5
15.5
-0.2
1.3
10.6
12%
axilla
6.5
10.0
9.5
11.0
12.0
9.5
8.5
7.5
10.0
9.0
18.0
13.0
10.0
12.5
8.5
11.0
6.5
4.8
14.5
15.5
0.0
1.8
10.4
17%
SENSORY
V
56.5
52.4
54.5
52.4
52.0
48.0
48.1
46.0
61.0
60.0
55.5
70.0
50.0
47.7
53.0
54.0
48.0
46.5
44.6
48.2
-0.2
3.7
52.4
7%
A
14.
15.
12.
11.
8.
8.
12.
7.
12.
14.
10.
8.
11.
12.
6.
4.
13.
5.
14.
8.
2.0
2.6
10.2
26%
V
45.0
43.0
44.5
41.4
44.6
43.7
45.5
49.5
43.5
42.7
45.0
48.9
50.0
49.5
47.5
51.5
39.6
39.2
42.0
43.0
-0.5
1.8
45.0
4%
PERONEAL
MOTOR
ankle
3.4
2.5
5.4
4.8
2.0
3.0
4.8
4.6
4.4
9.0
3.6
3.2
3.0
5.6
9.5
13.5
3.4
5.0
10.0
8.8
-1.0
1.6
5.5
29%
A
knee
2.6
2.5
5.4
4.6
2.6
3.6
5.4
5.0
5.4
7.5
3.2
4.0
3.0
5.6
9.5
14.0
3.0
5.0
9.5
7.8
-1.3
1.4
5.5
26%
SURAL
SENSORY
V
44.0
40.0
42.5
41.3
43.0
41.4
42.5
38.0
45.3
38.0
45.2
41.3
48.4
48.4
43.7
43.7
41.2
35.0
42.5
43.8
2.7
2.7
42.5
6%
A
8.
5.
12.
17.
10.
15.
14.
14.
12.
20.
9.
9.
18.
28.
12.
22.
11.
9.
8.
13.
-3.8
4.2
13.3
32%
V = Velocity
A = Amplitude
Electrical Measurements. With any measurement de-
signed to distinguish differences among groups, there
is a need for some understanding of its reproduc-
ibility. In this instance, ten participants volunteered
to submit to a second measurement of the electrical
parameters. The results for these ten men are given in
Appendix Tables 9 and 10.
Appendix Table 9 results were the first computed
and were done blindly in that the identity of the
pairing was unknown to the person taking the mea-
surements on the tracing. Since varying degrees of
precision in the recording were evident in the results,
the neurologists agreed to recalculate the entries in
the table. The result is Appendix Table 10. In this
TABLE 11—Average Nerve Conduction Velocities (Meters per Second) by Place of Employment
Group
ASARCO Hi
ASARCO Lo
Aluminium
Tacoma
Number
36
33
13
27
Ulnar-Motor
Wrist-Elbow
55.6
54.4
56.1
53.9
Ulnar-Motor
Elbow-Axilla
63.5
63.3
61.5
62.2
Ulnar
Sensory
50.4
50.6
52.2
49.4
Peroneal
Motor
44.9
44.8
47.1
47.4
Sural
Sensory
41.4
41.3
43.5
41.6
11
-------
TABLE 12—Average Amplitudes (/iV) and Duration (ms) by Place of Employment
Ulnar Motor
Group
ASARCO Hi
ASARCO Lo
Aluminum
Tacoma
Dur.
12.4
12.5
13.3
12.8
Ap-W
9.7
9.6
11.0
10.2
Ap-E
8.9
9.1
10.7
9.9
Ap-A
8.5
8.6
10.3
9.5
Ulnar
Sensory
Ap
6.4
8.2
8.4
10.3
Peroneal
Ankle
4.7
4.3
6.9
5.4
Amplitude Sural
Knee
4.4
3.9
6.6
5.1
Amplitude
10.7
12.7
11.4
10.9
case, too, the revisions for these 10 individuals were
done while the data were still blind to the laboratory.
Revisions for the entire study population were done in
the same way.
It is difficult to summarize precisely the information
about repeatability of the measurements on a participant
from one day to the next day and the evidence con-
cerning calculation errors from the tracing. The coeffi-
cients of variation suggests that the velocities are more
repeatable than the amplitudes. Among the velocities,
the peroneal motor nerve determination appears to be
the most repeatable which fits with a priori consid-
erations regarding source of error in the electrical mea-
surements in that this nerve is the longest nerve studied
and the error in measuring its length with a steel tape
should have the smallest percentage effect.
The second calculations show perhaps a little
smaller average difference and their signs are more
evenly distributed, six negative and seven positive, ver-
sus ten and three on the first occasion. These facts should
be borne in mind in interpreting subsequent results from
the electrophysical examination of the participants.
If the calculation differences are used to estimate
the error of the measuring process, an average standard
deviation of approximately three meters per second for
the velocity determinations results. FoYtunately, the re-
calculations do not suggest any large average bias in the
initial readings but rather suggest only residual random
error; 57 of 90 non-zero first minus second reading differ-
ences were negative, with an average difference of 0.02
meters per second for the 100 differences.
Average conduction velocities by employer groups
are shown in Appendix Table 11. No consistent patterns
by employer group are apparent with the possible ex-
ception of the differences for the peroneal nerve. Pooling
the two ASARCO groups and the two comparison groups
yields a mean difference in velocity for the peroneal
nerve of 2.45 meters per second. This difference is sig-
nificantly different from zero at the p .005 probability
level. However, the anticipated difference between the
ASARCO Hi and the ASARCO Lo groups, on the
assumption that a higher arsenic exposure might be
associated with an increased velocity deficit was not
observed.
It is also of some interest that for the velocity mea-
surement wjjich is most repeatable, the peroneal nerve
velocity, the two ASARCO groups have standard devia-
tions of 5.5 and 5.9, as compared to 3.2 and 3.4 for the
two other groups. Two possible reasons for the increased
variability in the ASARCO participants may be ad-
vanced, a greater age span in these groups and the possi-
bility of more varied exposure that may influence these
measurements.
The situation for the amplitude and duration mea-
surements is shown in Appendix Table 12.
The ordering of each of the average values is of
interest. With the exception of the sural amplitude, the
two lowest values are always in the ASARCO groups.
Except for the sural and the ulnar sensory amplitudes,
the values for the ASARCO Hi and Lo groups are quite
similar. Thus, amplitude and duration means are more
consistently related to workplace than are the velocities.
Many of the values in Appendix Tables 11 and 12
appear unusual when judged against the corresponding
normal values (Appendix I) from the consulting Neuro-
logical Group's experience. As an example, the average
velocity for the ulnar sensory nerve for the entire group
of 109 participants with this measurement is. 50.4 m/s
which is essentially the value given for the lower end of
the normal range for this velocity, 50 m/s. Only the
participants from the aluminum company have an aver-
age clearly within the normal range. Somewhat similar
remarks can also be made about the sural sensory veloci-
ties. For the amplitudes, it is apparent that low values are
generally present in the entire group. As an example, 58
of the 109 men tested had an amplitude measurement
TABLE 13—Clinical Neuropathies by Employer Group
Neuropathies
ASARCO Hi ASARCO Lo Aluminum
Tacoma
TOTAL
None
Sensory Only
Motor Only
Both Sensory and Motor
21
10
2
4
57%
27%
1R%
19
11
1
2
58%
33%
go/
8
5
0
0
62%
38%
22
5
0
0
81%
19%
70
31
3
6
64%
28%
3%
5%
TOTAL:
37 100% 33 100% 13 100% 27 100% 110 100%
12
-------
TABLE 14—Clinical Neuropathies by Employer Group: Eliminating
Participants with Potential Reasons for Neuropathy
Neuropathies
ASARCO Hi ASARCO Lo Aluminum
Tacoma
TOTAL
None
Sensory Only
Motor Only
Both Sensory and Motor
18
10
2
3
55%
30%
15%
17
8
1
2
61%
29%
10%
7
2
0
0
78%
22%
22
2
0
0
92%
8%
64
22
3
5
68%
23%
9%
TOTAL:
33
28
24
94
less than the lower limit of the normal range (8jiv) for the
ulnar sensory nerve. For the benefit of readers who are
interested in these phenomena, the entire estimated cu-
mulative probability distribution, the probability density
function, and a histogram with "normal value" end-
points for each of these measurements are available on
request. A final note on the electrical measurements can
be made, again concerning the values for the peroneal
velocities. Of the twelve men who have velocities slower
than the lower limit of the normal range (40 m/s), all are
employed at ASARCO; six in the high exposure group
and six in the low group. They range in age from a low of
32 to a high of 64 years. The failure to have a concentra-
tion of these men in the high arsenic exposure group is
again puzzling, suggesting at least that a dose response
mechanism is not clearly evident.
2. Clinical Findings. If the participants are classified by
the outcome of the neurological examination as hav-
ing no evidence of a neuropathy or having a sensory, a
motor or both a sensory and motor neuropathy, the
results are as shown in Appendix Tables 13 and 14.
A potential reason for exhibiting a neuropathy
was recorded by the neurologist during the examina-
tion; e.g., diabetes, trauma, alcohol consumption, etc.
If the 16 participants with such "excuses" are re-
moved from Appendix Table 13, a slightly different
impression of the association of neuropathies with
place of employment is created in Appendix Table 14.
No interpretation is made of these findings at this
point as a discussion of the results of the multivariate
analysis is presented in a later section will include
interpretation of these results.
3. Relation of Clinical Findings and Electrical Results.
Two ways of determining whether the clinical exami-
nation and the electrical measurements are identi-
fying the same men as those having neurological ab-
normalities are next examined. Since there were only
three participants among the thirty-one who were
classified as having a neuropathy whose neuropathy
was exclusively motor, the classification in Appendix
Tables 13 and 14 has been collapsed to two categories,
"neuropathy" and "no neuropathy" in this com-
parison. Also, for economy in presentation, the ulnar
motor electrical values relate only to the forearm. For
this comparison, we have included all persons exam-
ined and tested who did not have an excuse listed by
the neurologist for a potential neuropathy.
From Appendix Table 15, it seems possible to con-
clude that there is essentially no association between the
clinical classification and the electrical measurements for
the two motor nerves — approximately the same propor-
tion of each neuropathy group has the various electrical
abnormalities. For the sensory nerves, on the other hand,
there is a positive association between the two (the pro-
portions of electrical abnormalities do vary by class),
although it is not very strong. A statistic appropriate for
measuring the degree of this association on a scale from 0
(no association) to 1 (perfect association) is the tetra-
choric correlation coefficient. For calculating this statistic
the electrical-classification is further collapsed to "veloc-
ity and amplitude normal" and "one or both abnormal".
The degree of association measured in this way for the
ulnar sensory nerve is 0.46 and for the sural sensory it is
0.27, neither very high.
TABLE 15—Comparison of Clinical Classifications and Normal Values for Electrophysical
Examination
Electrophysical Examination
Nerve
Peroneal Motor
Ulnar Motor
Ulnar Sensory
Sural Sensory
Velocity &
Amplitude
Clinical Normal
Neuropathy n %
Yes
No
Yes
No
Yes
No
Yes
No
19
42
25
53
5
26
19
49
63
67
83
84
17
41
63
78
Velocity
Reduced
n %
2
7
2
1
2
13
5
10
7
11
7
2
7
21
17
16
Amplitude
Reduced
n %
7
13
3
6
11
16
4
2
23
21
10
10
37
25
13
3
Both
Reduced
n %
2
1
0
3
12
8
2
2
7
1
0
5
40
13
7
3
13
-------
TABLE 16—Average Conduction Velocities (M/s) and Amplitudes (uV) for Participants Classified
by Neuropathy Status
VELOCITIES
Neuropathy N
None 69
Sensory Only 23
Motor Only 3
Both 5
Ulnar Motor
Wrist-elbow
55.1
53.8
56.0
56.1
Dinar-Motor
Elbow-Axilla
63.2
63.0
66.2
57.2
Ulnar
Sensory
51.1
49.3
53.4
46.9
Peroneal
Motor
46.3
46.3
45.5
41.0
Sural
Sensory
42.3
41.8
41.9
39.1
Total
100
AMPLITUDES
Neuropathy
None
Sensory Only
Motor Only
Both
Dur.
12.9
12.3
12.5
12.9
Ulnar-Motor
Wrist
10.0
9.0
10.3
9.9
Dinar-Motor
Elbow-Axilla
9.5
8.2
10.0
10.0
9.3
7.7
9.5
8.6
Dinar
Sensory
9.1
7.0
4.7
4.8
Peroneal Motor
Ankle
5.0
4.9
4.8
2.7
Knee
4.7
4.4
3.5
3.3
Surai
Sensory
12.6
9.7
8.7
9.2
Dur. = Duration
Attention may also be called to the fact that the
proportion of participants with normal electrical values
varies from a high of 84% for the ulnar motor nerve to a
low of 31 % for the ulnar sensory nerve. If the information
for the four nerves is used simultaneously to classify a
participant, only 18 of 100 men in this excuse-free group
have normal values for all four nerves.
A second way of examining whether the two mea-
surements are associated is to compute the average val-
ues of the electrical measurements for men grouped by
the clinical classification. These results are included in
Appendix Table 16.
Some consistent results are exhibited for the two
sensory nerves measured. The "motor only" and "both"
groups are apparently too small for stability of the mean
results in these categories for velocities and amplitudes
for the motor nerves. »
65-
60-
55-
50-
45-
40-
35-
30-
20
30
40
50
60
70
Age
FIGURE 5. Peroneal Motor Conduction Velocity by Age (Control
Croup Only).
Summarizing the association between the clinical
and the objective measurements of the neurological
status of each man thus far, we conclude that the associa-
tion exists but it is weak. In consequence, subsequent
analyses of the relation of neurological deficits to ex-
posure to various metals will either require an examina-
tion of each of the methods for describing the neurologi-
cal status of a participant or the development of a new
classification embodying elements of both methods.
The analyses in this section could possibly be im-
proved if normal values specific for age were available.
There is evidence in these data that both velocity and
amplitude are age dependent. If anything, taken as a
group, the amplitudes seem to evidence more age de-
pendency than the velocities. If examination of the de-
pendency on age is examined only in the non-ASARCO
participants to reduce the potential confounding influ-
ences of wide variation in exposure, the results in Appen-
dix Figures 4 and 5 are typical of the age dependency.
For the ulnar sensory nerve, the reduction in amplitude
for each increase of one year in age is estimated at 0.19
microvolts. The product moment correlation, r, of the
amplitude with age is —0.44, indicating that only about
19% (r8) of the variability in this amplitude is. explained
by age. Factors other than age, including measurement
error, apparently are strongly involved in determining
the outcome of the amplitude measurement. Figure 5
shows a reduction of 0.16 meters per second in the per-
oneal velocity for each increase of one year in age with an
r = 0.48.
Neuropathy as diagnosed by the neurologist also
seems to exhibit a weak association with age as shown in
appendix Table 17. While the association does not reach
statistical significance (p .05), the increasing frequency of
neuropathy with age in the group of participants without
potential excuses in quite orderly and suggestive of an
age effect.
Association of Neurological Abnormalities and Ex-
posure to Arsenic and Other Metals: A Multivariate
Analysis. The descriptive data and the preliminary analy-
14
-------
TABLE 17—Number and Percent of Participants with Clinical Neuropathies by Age
GROUP
TOTAL
AGE
20-29
30-39
40-49
50+
10
31
13
39
TOTAL
No Neuropathy
Neuropathy
8
2
80%
20%
24
7
77%
23%
9
4
69%
31%
22
17
56%
44%
63
30
93
sis were predominately done at the University of Wash-
ington, and have been presented above. These were
supplemented by a sophisticated multivariate statistical
analysis.
Since statistics moves from the simpler techniques to
the more complex analyses to uncover relationships, we
shall now deal with a limited multivariate analysis of the
neurological responses in relation to the categorical de-
mographic groups, arsenic levels and covariates.
As the entire pilot investigation was directed to-
wards determining if the "high" exposure to AS2O3 (ar-
senic trioxide) of employees of the ASARCO Copper
Smelter had adverse neurological responses as compared
with their "low" exposed counterparts, the main cate-
gorical demographic classification was employer type.
The ASARCO management furnished subjects with
"high" exposure and "low" exposure as follows:
A. Men working in plant locations by way of job classi-
fication, where the arsenic exposure was "high".
B. Comparable men (age wise) where exposure was
"low" by way of job classification.
Since the "high" exposure group was defined to be
200 Mg/Hter or more for at least the past five years, this
automatically defines the "low" exposure group also.
Some of our observations on available variables follow:
Demographic variables:
1. Four employer groups:
ASARCO High exposure
ASARCO Low exposure
ASARCO
Number'
Total number of volunteers
37
33
13
_28
111
(32)
(28)
( 9)
(24)
(93)
2. Age
The median ages in ASARCO High and Low
are both around 51 years. The mean ages in
ASARCO High and Low are both around 47 years.
This is to be expected as the "volunteers" are
matched by age!
* Throughout the report numbers in parentheses indicate results
omitting data for excuses, e.g., in the ASARCO group (total of 69) there
were nine excuses, one case without ulnar sensory measures, two with-
out arsenic in nails measures, three without arsenic in hair measures
and one case without clinical findings. Therefore, the sample size will
vary according to the analysis done.
In Kaiser Aluminum the median and mean ages
are around 36 and 37 years respectively. Also, in
Tacoma, the median and mean ages are around 37
(39) and 38 years respectively.
As aging is generally accepted to have an ad-
verse effect on neurological responses, it is worth
noting the statistical significance of the difference in
the ages of the control (non-ASARCO) and ASARCO
High or Low with p = .001 (p = .011). This indicates
the necessary caution to be used in assigning the
basis for the differences in the groups.
3. Smoking; Amount Smoked
The smokers and nonsmokers' distribution as
well as the amount smoked by different employer
types is significant with p < .05 (p < .01) and should
be taken into consideration as related to neurological
responses.
4. Urine Specific Gravity
The distribution of this variable is significantly
different over the strata with unknown effect on the
neurological response with p = .0204 (p = .0237).
5. Urine Creatinine Mg%
The distribution of this variable is significantly
different over the strata with unknown effect on the
neurological response with p = . 0072 (p = .0017).
Range (22 to 408)
6. Carboxyhemoglobin as of % of Hemoglobin
Carboxyhemoglobin as a percentage of hemo-
globin is known to adversely affect the immunologi-
cal balance and perhaps will affect the neurological
response adversely. The distribution of this variable
is found to be significantly different over the strata of
employer type and high-low exposure group with
p = .009, (p = .007). Smoking frequency is highly
correlated with Carboxyhemoglobin (r = .73).
Comment: The smoking, urine specific gravity,
urine creatinine and Carboxyhemoglobin all indicate
the TACOMA group not to be comparable to the
other groups.
Significant Response Variables
7. Ulnar Sensory Amplitude
The mean of this neurological response variable
is significantly different over the employer types in-
cluding the High-Low exposure groups in ASARCO,
p = .0007 (p = .0004). There appears to be one
15
-------
outlier (#136) which, if elirriinated, would contribute
to an even higher level of significance.
8. Peroneal Amplitude Ankle
The mean of this neurological response variable
appears to vary, p = .06, over the high-low and other
employer strata. When "excuses" are omitted the p
value changes to .43. This is mostly due to the
change made in the KAISER group.
9. Peroneal Amplitude Knee
The mean of this neurological response variable
is significantly different over the groups, p = .05.
Removal of "excuses" with its resulting change for
the KAISER group brings p to .38.
Comment: The two ASARCO groups appear to
differ from the KAISER and TACOMA groups.
Since age correlates with each of the neurological
measures at about —.30 it is difficult but necessary to
separate the age effect and the group effect.
10. Urinary Arsenic
Some measure of arsenic is used in forming the
strata with the higher values corresponding to
ASARCO High. Therefore, we can expect the groups
to differ at a high level of significance.
Given that they have significantly high arsenic
levels in the ASARCO High group, we shall analyze
the relationship of this variable to others measured
in a later section.
Multivariate Analysis
The following are selected comments on multi-
variate procedures which should be helpful to the reader.
Factor analysis is a technique for discovering linear
combinations of variables which "explain" the major
factors or directions of variability in a group of individ-
uals. The factors are estimated by least squares tech-
niques and are uncorrelated with each*other. Improve-
ments in this and other multivariate analyses can be
expected by using transformations that improve the lin-
earity of the interrelationships of the variables and that
equalize the variability of measurement error for differ-
ent individuals.
Discriminant analysis is a technique for discovering
linear combinations of the variables (similar to the factors
above) which best "separate" the groups under study.
The criterion is the ratio of the distance between the
groups to the standard error of the discriminating factors.
Again improvements in linearity or homogeneity of vari-
ances can improve the discrimination.
Canonical analysis is a technique designed to esti-
mate the pairs of linear functions of variables (one from
each of two groups) which are most highly correlated.
We thus obtain indications of the factors in one group
that are related to factors in the other group.
The tables discussed in this section of the report
present the results of the multivariate approach with the
corresponding interpretations, including appropriate
transforms on the response variables, to supplement the
various univariate analyses carried out in the preliminary
statistical analysis.
A comparative factor analysis on the response vari-
ables between the raw and transformed (logs and recip-
rocals) measures was undertaken. A Logi0 transformation
on the amplitudes and an inverse transformation on the
velocities and duration improved the separation of the
factors. These transformations also increased the discrim-
ination between the groups. Log transformations were
taken on the arsenic level measures to reduce the skew-
ness in their distribution.
The logarithmic transformation for amplitudes and
for dose are fairly standard as is the reciprocal for veloci-
ties. The subsequent reporting is for the transformed
variables.
Some observed associations on neurological
responses vs. arsenic
1. Lower ukiar sensory amplitude and lower peroneal
velocity are associated (p <- .01) with higher arsenic
levels in the urine, hair and nails.
2. Lower peroneal amplitude at ankle and knee are also
associated with higher arsenic levels in the urine, hair,
and nails. (The variables are appropriately trans-
formed. ) There are too few non-zero blood readings to
expect definitive results using that variable.
3. Discriminant Analysis of the Data: The discriminant
function separating the ASARCO group from the
non-ASARCO group gives 69.1 % correct classification
for the ASARCO group and 67.5% correct classifica-
tion for the non-ASARCO group respectively. The
variables (original readings) that constituted the dis-
criminant function were discovered to be the neuro-
logical response variable consisting of 1) logarithm of
the ulnar sensory amplitude and 2) reciprocal of the
peroneal motor conduction velocity (p H .0005). The
percentages reported are for "jackknifed" analyses.
This is a procedure to approximately adjust the per-
centage correct classification for the bias due to eval-
uating the results on the same observations used to
determine the discriminating function.
The discriminant analysis showed that the step-
wise results of separating ASARCO from non-
ASARCO employees is not improved by including
more than the three measures—log ulnar sensory am-
plitude, log peroneal amplitude, knee, and inverse
ulnar velocity, elbow to axilla. . . . The raw and trans-
formed data indicated the significance of the same
three measures, whether revised or original. Appendix
Table 18 gives the jackknifed classification results af-
ter three steps.
4. Factor Analysis of the Data: For a breakdown of the
employee groups, consisting of ASARCO High, Low.
Non-ASARCO and all employees combined, a factor
analysis placed neurological response variables among
the first three factors as shown in Appendix Table 19.
The first three factors explain between 50% and 60%
of the total variance of the neurological variables.
First-of all, amplitudes, whether they are action
potential amplitude or peroneal amplitudes at any
16
-------
TABLE 18—Discriminant Analysis Jackknifed Classification Tables after Three Steps
Original Readings
a. Raw measures, priors = .63, .
Group
ASARCO
CONTROLS
TOTALS
b. Transformed
Group
ASARCO
CONTROLS
TOTALS
c. Transformed
Group
ASARCO
CONTROLS
TOTALS
% Correct
85.3
37.5
67.6
measures, priors
% Correct
79.4
22.5
58.3
measures, priors
% Correct
69.1
67.5
68.5
37
ASARCO
58
25
83
= .63, .37
ASARCO
54
31
85
= .50, .50
ASARCO
47
13
60
Controls
10
15
25
Controls
14
9
23
Controls
21
27
48
Group
ASARCO
CONTROLS
TOTALS
Group
ASARCO
CONTROLS
TOTALS
Group
ASARCO
CONTROLS
TOTALS
Revised Readings
% Correct
88.4
26.8
65.5
% Correct
87.0
36.6
68.2
% Correct
56.5
65.9
60.0
ASARCO
61
30
91
ASARCO
60
26
86
ASARCO
39
14
53
Controls
8
11
19
Controls
9
15
24
Con/rote
30
27
57
joint in the body, seem to behave alike. In the same
way velocities seem to behave alike except for the
peroneal velocity. The factor analysis revealed that
the neurological response variables that are "af-
fected" in some way by playing a role in accounting
for the factors are the amplitude variables and the
only velocity variable was the peroneal.
The ulnar amplitudes at axilla, elbow and wrist
appear together in Factor 1 whatever the group. The
peroneal amplitudes (ankle and knee) appear in the
second factor whatever the group. Factor 3 appears to
be an assortment of velocities primarily. We do not
see any distinguishing differences in these factors
from one group to another.
Test-Retest Study
Let us now compare these factors with test-retest
loadings of original vs. revised "blind" sub-sample of
ten employees. We have used loadings (those greater
than or equal to .70) and found the distributions of
measures across the first three factors to be consistent
with the ASARCO Low and High exposure groups
from which they were drawn. The four sub-samples
also compare well with each other, although the order
of significance of the factors vary. See Appendix Table
20.
In addition to the factor analysis, we compared
the average differences of the response measures
taken not only between the test and retest, but also
between the original and the revised data we later
received. Appendix Table 21 shows that perhaps the
computational errors were larger than the measure-
ment errors.
TABLE 19—Rotated Factor Loadings (Patterns) > .50
A. For all employee groups combined, the distribution of response variables over the first three factors is:
Factor 1 Factor 2 Factors '
U. amp. elbow (.976) P. amp. ankle (.976) P. vel. (.748)
U. amp. axilla (.967) P. amp. knee (.973) U. sensory vel. (.673)
U. amp. wrist (.969) Suralvel. (.692)
B. For ASARCO High the distribution of variables over the factors is:
Factor 1 Factor 2 * Factors
U. amp. elbow
U. amp. axilla
U. amp. wrist
C. For ASARCO Low
Factor 1
U. amp. elbow
U. amp. axilla
U. amp. wrist
D. For Non-ASARCO
Factor 1
U. amp. elbow
U. amp. axilla
U. amp. wrist
(.968)
(.965)
(.929)
P. amp. ankle
P. amp. knee
the distribution of + variables over
(.958)
(.956)
(.957)
Factor 2
P. amp. ankle
P. amp. knee
(.889) U. vel. wrist-elbow
(.860) Sural amp.
the factors is:
Factors
(.964) P. vel.
(.953) Sural vel.
U. sensory vel.
U. vel. wrist-elbow
(.728)
(-.617)
(.735)
(.666)
(.598)
(.542)
the distribution of the variables over the significant factors is:
(.958)
(.961)
(.969)
Factor 2
P. amp. ankle
P. amp. knee
U. sensory vel.
U. vel. duration
Factors
(.958) P. vel.
(.963) U. sensory amp.
(-.554) Sural vel.
(.536)
(.865)
(-.615)
(.560)
17
-------
TABLE 20—Rotated Factor Loadings ^ .70 on Test-Retest Samples (Original and Revised)
A. Original Test
Factor 1
P. amp. ankle
P. amp. knee
U. vel. wrist-elbow
B. Original Retest
Factor 1
U.amp. elbow
U. amp. axilla
U.amp. wrist
C. Revised Test
Factor 1
U. amp. elbow
U.amp. axilla
U. amp. wrist
D. Revised Retest
Factor 1
U.amp. elbow
U.amp. axilla
U.amp. axilla
Sural vel.
(.927)
(.966)
(-.792)
(.973)
(.954)
(.954)
(-979)
(.984)
(.967)
(.975)
(.962)
(.962)
(.831)
Factor 2
U. amp. elbow
U.amp. axilla
U.amp. wrist
Factor 2
Sural vel.
U. vel. elbow-axilla
U. vel. duration
Factor 2
P. amp. ankle
P. amp. knee
U. vel. elbow-axilla
Factor 2
P. amp. ankle
P. amp. knee
Sural amp
(.777)
(.843)
(.822)
(.802)
(.785)
(-.775)
(.979)
(.949)
(-.843)
(.873)
(.930)
(.753)
Factor 3
U. vel. duration
Sensory vel.
Factors
P. amp. ankle
P. amp. knee
Factors
P, vel.
Sural vel.
U. vel. wrist-elbow
» Factors
U. vel. wrist-elbow
U. vel. elbow-axilla
(-.756)
(.712)
(.941)
(.920)
(.885)
(.738)
(.736)
(.906)
(.733)
In Table 22 we present the standard errors of the
various neurological measurements as obtained origi-
nally and as revised. Also, we have analyzed the data
in raw form as well as transformed. The important
findings are in the final two columns of the table. The
ratio for the raw data ranging from .53 up to 1.00
indicates the greater amount of information available
from the revised data; e.g., a ratio of .71 noted for the
peroneal amplitudes at ankle and knee indicates a
factor of 2.0 improvement in precision l/(.71f We
can seek even further improvement by taking the
transform; e.g., for peroneal amplitude knee the fac-
tor is 4.0 = 1/C50)2. These factors are so large (and for
some of the measurements even larger) that we could
expect a much more definitive analysis using revised
data for the whole study.
6. Canonical Analysis of the Data: While there is no
demonstrable separation between the strata, namely
the ASARCO High, ASARCO Low and non-
ASARCO, the neurological responses most related to
the combined arsenic levels in urine, hair and nails
are:
a. Peroneal Amplitude Ankle
b. Peroneal Amplitude Knee
c. Ulnar Sensory Amplitude, and
d. Peroneal Motor Conduction Velocitv.
TABLE 21—Ranks of Average Differences in Test-Retest Study. (Smallest Average Difference
Rank 1.)
Neurological Measures
A
Original
Test vs.
Retest
B
Revised
Test vs.
Retest
C
Test:
Original
vs. Revised
C
Retest:
Original
Revised
1. Ulnar Velocity, Wrist-Elbow 3
2. Ulnar Velocity, Elbow-Axilla 3
3. Ulnar Action Potential
Duration 3.5
4. Ulnar Amplitude, Wrist 1
5. Ulnar Amplitude, Elbow 1
6. Ulnar Amplitude, Axilla 2
7. Ulnar Sensory Velocity 3
8. Ulnar Sensory Amplitude 2
9. Peroneal Motor Conduction
Velocity 1
10. Peroneal Amplitude, Ankle 2
11. Peroneal Amplitude, Knee 2
12. Sural Velocity 4
13. Sural Amplitude 4
Average Rank 2.4
2
4
3.5
2
2
1
1
4
2
3
3
3
2
2.5
1
2
1.5
4
3
3
2
1
4
1
1
1
3
2.12
4
1
1.5
3
4
4
4
3
3
4
4
2
1
2.96
18
-------
TABLE 22—Standard Error of Differences, Test-Retest Study.
Neurological Measures
Orig. test-retest
Revised test-retest
Ratio Rev/Orig
1 . Ulnar Velocity, Wrist-Elbow
2. Ulnar Velocity, Elbow-Axilla
3. Ulnar Action Potential Duration
4. Ulnar Amplitude, Wrist
5. Ulnar Amplitude, Elbow
6. Ulnar Amplitude, Axilla
7. Ulnar Sensory Velocity
8. Ulnar Sensory Amplitude
9. Peroneal Motor Conduction Velocity
10. Peroneal Amplitude, Ankle
1 1 . Peroneal Amplitude, Knee
12. Sural Velocity
13. Sural Amplitude
raw
4.07
5.95
1.36
3.25
3.42
3.02
7.01
4.16
1.90
2.09
1.84
2.25
5.08
transform
.00128
.00172
.01080
.230
.264
.236
.00346
.168
.00100
.179
.179
.00149
.200
raw
2.12
5.69
1.36
2.89
1.80
1.89
3.90
2.36
1.84
1.48
1.31
2.03
5.09
transform
• .00065
.00166
.01080
.106
.073
.080
.00116
.111
.00086
.111
.090
.00125
.226
raw
.52
.96
1.00
.89
.53
.63
.56
.57
.97
.71
.71
.90
1.00
transform
.51
.96
1.00
.44
.28
.34
.34
.66
.86
.62
.50
.84
1.13
This finding is based on a canonical correlation
analysis taking the individual values of the arsenic
levels in blood, hair, nails and urine as one group and
the revised or original neurological readings as the
other group. We have significance p = .007 for the
first factor (originally p = .07) using the revised elec-
trical readings. See Appendix Table 23.
Using this analysis as an indication of the possi-
bly important variables from the sum of the log levels
of arsenic in urine, hair and nails and a "neurological"
variable from the sum of the log amplitudes for ulnar
sensory, peroneal ankle and peroneal knee reduced by
the reciprocal velocity, peroneal. These are the vari-
ables "suggested" by the canonical analysis. We
found these macro variables to be correlated at —.467
(p < .0005) for ASARCO employees. With "excuses"
omitted the correlation is —.430 (p ^- .005). Non-
ASARCO individuals were not included in this analy-
sis.
The nine tables and the two remaining figures in the
Appendix provide the data for the final sets of analysis
devoted to dealing with the effects of "excuses", smoking
and age. Thus, Table 24 not only includes the sets of
correlation and regression coefficients but also lists the
means, standard deviations and p-values for age, amount
smoked, MNV (the macro-neurological variable) and
MAL (the arsenic load macro-variable) for the four
groups of employees. The four groups differ significantly
on all four variables. We can see the need for the analysis
in Table 25 since the controls are younger and the KAI-
SER group smokes less. We can also note the difficulty in
showing a difference between the high and low exposure
groups at ASARCO since both have a high arsenic load
and the "excuses" show similar results. Table 26 shows
the correlations of age, amount smoked and arsenic load
with the macro-neurological variable for the four em-
ployee groups. Table 27 repeats some of the correlations
of Table 26 and includes also some additional partial
correlations which may clarify the fact that the relation-
ships to age and smoking do not appreciably affect the
relationship of arsenic to neurological response. There is
no evidence that including the "excuses" would vitiate
these findings when they are included in the analysis.
Table 28 shows how comparable they are with respect to
the macro-neurological variable generated. Table 29
shows a similar result for those above maximum arsenic
load. Table 30 reports the comparison for those who were
below normal-neurologically and above normal maxi-
mum arsenic load. The relationship between the neuro-
logical measure and arsenic is significant as shown by the
chi-square analysis of Table 31. The relationship be-
tween the neurological measure and arsenic remains sig-
nificant even when persons with excuses are removed.
(See Table 32).
Appendix Figure 6 presents a chart showing the
reduced neurological response related to increasing arse-
nic load of the groups under study. Appendix Figure 7 is
a similar presentation but is based on the data for indi-
vidual employees. There is no reduction in neurological
response for the controls, a lower response level for the
ASARCO low exposure group and dose-dependent fur-
ther response with increasing arsenic load for the high
exposure group and for the "excuses" group. The "ex-
cuses" group's neurological response is also arsenic
"dose" dependent at a smaller dose level than the non-
excuses group. As seen in Appendix Figure 7, the "ex-
cuses" slope parallels the high exposure non-excuses
slope but is shifted back to a lower arsenic load, i.e.
neurological loss is dose dependent and greater in
amount of loss.
TABLE 23—Canonical Variable Loadings. Restricted to Vari-
ables with Loadings. > .40 In any canonical variable.
FIRST CANONICAL VARIABLE
Original Data Revised Data
Group 1:
Group 2:
arsenic blood
arsenic urine
arsenic hair
arsenic nails
ulnar sensory amp.
peroneal amp. ankle
peroneal amp. knee
peroneal vel.
.331
.802
.855
.864
-.771
-.555
-.539
.410
.160
.815
.893
.688
-.685
-.585
-.602
.647
19
-------
TABLE 24—Correlation Results of Neurological Macro Variable against Arsenic Macro Variable, adjusting for age and smoking.
PEARSON CORRELATIONS
1. AGE vs. NEUROS
2. AMT. SMOKED vs. NEUROS
3. AGE vs. ARSENIC LOAD
4. AMT. SMOKED vs. ARSENIC
5. NEUROS vs. ARSENIC LOAD
PARTIAL CORRELATIONS
6. NEUROS vs. ARSENIC (age &
amt. smoked partialed out)
MULTIPLE CORRELATIONS
7. NEUROS vs. ARSENIC, AGE
AND AMOUNT SMOKED
EXCUSES REMOVED
ASARCO CONTROLS ALL
(n = 54) (n = 33) (n =87)
.0329
-.1873
.0991
.0906
-.2802
(p - .025)
-.2672
(p -.05)
,3240
(p - .02)
-.4079
(P -.02)
-.1946
-.2097
-.0062
.0578
-.0302
.4422
(p -.01)
-.1934
(p -.03)
-.2274
(p -.02)
.3035
(p - .002)
.1724
-.3695
(p - .001)
-.2916
(p - .005)
.4224
(P-..001)
EXCUSES
ALL ASARCO
(n=16) (n = 9)
-.4935
(P ~ .03)
-.3558
.5150
(p - .025)
.1529
-.7285
(p ~ .001)
-.6050
(P- .01)
.7962
(p ~ .001)
-.5043
-.2152
.1929
.0607
-.6918
(p - .020)
-.7533
(P -.01)
8884
(P - .002)
ALL
ASARCO
(n = 64)
-.0376
-.1837
.0757
.0988
-.430
(p ~.005)
-.3234
(p - .008)
.3327
(p - .005)
ALL
EMPLOYEES
(n = 103)
-.2559
(p -.005)
-.2477
(p -.007)
-.3475
(p - .005)
.1964
(p -.025)
-.4443
(p - .0001)
-.3444
(p ~ .005)
.4947
(p - .0001)
REGRESSION COEFFICIENTS
Standardized
INTERCEPT
AGE
AMT.SMOKED
ARSENIC LOAD
2.493
.004
-.162
-.266
3.196
-.403
-.169
-.028
2.756
-.128
-.192
-.298
3.724
-.247
-.303
-.555
8.559
-.618
-.470
-.544
2.799
-.075
-.178
-.319
2.892
-.156
-.196
-.351
F's
(2.89)
(5.01)
(14.27)
We may sum up our findings by stating that the
macro-neurological variable (MNV) is related to age in
general, but is not correlated to age in the ASARCO
employees. This is also true for the macro-variable for
arsenic load (MAL). The amount smoked is related to
MNV in general and should probably be adjusted for in
the analysis. The relationship is similar whether, or not
the employee was labeled "excuse". However, the
TABLE 25—Means and (Standard Deviations) and p-values.
,»
>
INCLUDING "EXCUSES"
High exposures, ASARCO
n = 37
Low exposures, ASARCO
n = 33
TACOMA
n = 13
KAISER
n = 28
ALL GROUPS
n= 111
AMT.SMOKED NEURO.S
AGE (Log Scale) (Macro Variable)
47.3( 9.7)
46.6(11.6)
37.8(11.9)
38.8( 9.4)
43.9(11.1)
p = .0012
.302(.267)
.314(.282)
.342(.258)
.133(.236)
.267(.272)
p = .0220
1.965(.697)
2.113(.578)
2.451(.528)
2.399(.425)
2.177(.606)
p = .0092
ARSENIC
LOAD
(Macro
Variable)
4.531(1.036)
2.186( .769)
.330( .325)
.395( .481)
2.237(1.907)
p = .0000
EXCLUDING "EXCUSES"
High exposures, ASARCO
n = 32
Low exposures, ASARCO
n = 28
TACOMA
n = 9
KAISER
n = 24
ALL GROUPS
n = 93
46.8(10.2)
45.4(12.0)
38.0(11.6)
38.4( 9.5)
43.3(11.3)
p = .0114
.296(.265)
.331(.281)
.441(.206)
.135(.241)
.296(.265)
p = .0093
2.004(.656)
2.138(.541)
2.361 (.584)
2.382(.448)
2.178(.578)
p = .0741
4.609( .975)
2.236( .753)
.305( .366)
.383( .473)
2.315(1.935)
p = .0000
20
-------
TABLE 26—Correlations with Macro-Neurological Variable.
ASARCO
HIGH LOW
KAISER TACOMA
ALL
INCLUDING "EXCUSES"
AGE
AMI. SMOKED
ARSENIC LOAD
EXCLUDING "EXCUSES"
AGE
AMI. SMOKED
ARSENIC LOAD
-.259
(n = 36)
-.191
(n = 36)
-.557
(n = 33)
-.182
(n-31)
-.226
(n-31)
-.489
(n = 28)
.134
(n = 33)
-.160
(n = 33)
-.105
(n-31)
.202
(n = 28)
-.133
(n = 28)
.028
(n = 26)
-.445
(n = 13)
-.670
(n = 13)
.598
(n = 13)
-.542
(n = 9)
-.704
(n = 9)
.623
(n-9)
-.282
(n = 28)
-.085
(n = 28)
-.069
(n = 28)
-.335
(n = 24)
-.031
(n = 24)
-.140
(n = 24)
-.246
(n= 110)
-.242
(n= 110)
-.430
(n = 105)
-.197
(n = 92)
-.215
(n = 92)
-.370
(n = 87)
amount smoked is poorly related, if at all, to arsenic load.
The macro variables for arsenic load and neurological
response are not related in the controls, but are related in
the ASARCO employees (even more strongly in the "ex-
cuse" group). There is evidence that all three factors,
age, smoking and arsenic reduce neurological function
when taken together. All groups show a significant over-
all effect from these three factors. If we remove by re-
gression the effect of smoking and age on neurological
response, we find very little reduction in the correlation
of MNV and MAL. We can state that the relation of the
neurological variable to arsenic load is not an artifact of
age or amount of smoking among ASARCO employees.
The control group continues to show a significant overall
effect from these three factors.
The conclusion of the analysis is that there is a
statistically significant relationship between arsenic load
and reduced neurological response in the smelter popu-
lation.
Discussion
There are a number of points that seem to merit
comment. First, the subjects were volunteers from a
selected set of ASARCO employees or volunteer partici-
pants following a general appeal and therefore it is hard
to assess the representativeness of the samples from their
respective strata. Then, the levels of arsenic measured in
the urine of the ASARCO high men are substantially
lower than those furnished by the company from records
dating back, in some instances, many years. There is also
some evidence in the historical data of a trend to lower
average values with the passage of time, but the data are
scanty. Whether this phenomenon is real or is a product
of collection, handling, and laboratory procedure differ-
ences with the urines cannot be determined from the
data at hand. Nevertheless, the ASARCO High group is
clearly distinguishable from the ASARCO Low in urinary
arsenic values as well as arsenic in hair and nails. The
levels of lead in the participants as a total group also
appear to be lower than that found in other studies. No
reason for the lower values has been determined by the
laboratory making the determinations. Also, the elec-
trophysical examination produces values for the various
velocities and amplitudes used in this analysis that have
relatively large components of variation due to day-to-
day variability within the same individual and to mea-
surements made on the tracing from the test equipment.
Recalculating the electrical parameters late in the study
TABLE 27—Correlations and Partial Correlations Associated with Age and Smoking.
EXCUSES REMOVED
Neuro vs. Arsenic
adj. (age)
adj. (smoke)
adj. (age + smoke)
Neuro vs. Smoke
adj. (age)
Neuro vs. Age
adj. (smoke)
ASARCO
(n = 54)
-.280
-.285
-.269
-.207
.091
-.187
.033
-.034
CONTROLS
(n = 33)
.058
-.031
-.021
-.030
-.006
-.185
-.408
-.374
ALL
(n = 87)
-.370
-.332
-.344
-.293
.172
-.259
-.193
-.231
EXCUSES
ALL
(n = 16)
-.727
-.636
-.730
-.605
.153
-.486
-.494
-.582
ASARCO
(n-9)
-.692
-.702
-.696
-.753
.061
-.589
-.504
-.699
ALL
ASARCO
(n = 64)
-.430
-.341
-.332
-.323
.099
-.209
-.038
-.110
21
-------
TABLE 28—Number Below Normal
Finding
Minimum Neuro-Clinical
No
Sen Motor Both Total
ASARCO high
low
Controls
Excuses
Total
2
2
2
1
7
3
1
0
2
6
1
1
0
0
2
2
0
0
1
3
8
4
2
4
18
TABLE 30—Number Below Normal Minimum Neuro and Above
Normal Maximum Arsenic Load-Clinical Finding
No
Sen Motor Both Total
ASARCO high
ASARCO low
Controls
Excuses
Total
2
0
1
1
3
3
0
0
1
4
1
1
0
0
2
2
0
0
1
3
8
1
0
3
12
and subsequent analysis of these new values appears not
to markedly change any earlier findings. The presence of
these variations only adds to the difficulty of detecting
other factors that may be associated with the electrical
variables. This may be the reason that no significant
regression of amplitudes or velocities on any of the
metal-tissue combinations taken singly could be found.
In many of the scatter plots there was a suggestion of a
decrement in nerve function with increasing levels of the
metals in the tissues, but the "noise" in the system made
detecting the decrement by statistical procedures ex-
tremely difficult. Transformations and the development
of an index of neurological status demonstrated an associ-
ation of arsenic in urine, hair and nails with neurological
deficits. The association of the macrovariables from the
multivariate analysis further strengthened this finding.
The multivariate analysis did show that some of the
neurological responses are associated with the arsenic
levels in the tissues of the subjects. Indications of this
relationship were present in the original data and were
strengthened when the revised neurological readings
were used. This should add confidence to the findings of
this study. The establishment of a dose-response relation-
ship is a first and necessary step in studying the possibly
harmful effects of heavy doses.
The issue of the biological importance of the associ-
ation between arsenic levels and neurological deficit is
difficult to assess. While the clinical neuropathies diag-
nosed were all mild, non-disabling, and generally not
recognized by the men with these diagnoses, any loss of
function that could be avoided by preventive measures
cannot be regarded as unimportant or trivial.
What is a neuropathy? A peripheral neuropathy is
any disorder that affects the peripheral nerves and inter-
feres with their normal functioning. We know the pe-
ripheral nerves are those parts of the nervous system that
carry the messages from the spinal cord to the muscles or
the sensory messages from the skin to the central nervous
system, namely the spinal cord, and then ultimately the
brain. Clinically, the peripheral nerve can be affected by
many processes, some metabolic as in diabetes, others
that are toxic as can be seen in association with alcohol-
ism or with heavy metals such as arsenic or lead. Arsenic
is known to cause a sensory motor neuropathy. In the
chronic low grade form of arsenical intoxication, it is
more likely that a sensory neuropathy will occur with
relatively minor motor signs, namely weakness. In acute
intoxications however, there is both loss of sensation and
profound weakness.
There is clinical and histologic evidence of changes
in nerves of some individuals who have been exposed to
arsenic. In an industrial environment, the mode of in-
toxication may possibly be predominately from ingestion
of airborne particles, a finding which has long been rec-
ognized as leading to a neuropathic picture. Petren
points out that with ingestion as major mode of entry, a
polyneuropathy can be expected.36 This may be the ma-
jor type of exposure and route of entrance into the body
in the workers in a smelter. The requirement of face
masks by the smelter in certain job categories may have
diminished the exposure of these workers.
As indicated above, a significant number of individ-
uals have both clinical as well as sub-clinical signs of
neuropathy. By sub-clinical neuropathy is meant an indi-
vidual who, on neurologic examination states that he has
no subjective complaints of numbness or pins and nee-
dles feelings in his fingers and there is no evidence of
decreased appreciation of sensation to pin touch, temper-
ature, position or vibration sense in the fingers or hands.
There is no weakness and there is no reflex loss. These
individuals, however, will show on nerve conduction
studies a slowing of conduction velocity both in motor
and in sensory nerves. (See attached ranges of normal
values.) These values represent, for example, that the
ulnar sensory was 50-70 meters per sec. and the action
potential amplitude was 8-28 microvolts. By recording
the evoked response on film from the muscle belly after
stimulating the appropriate nerve, one measures not only
TABLE 29—Number Above Normal Maximum Arsenic Load
No
Sen Motor Both Total
(n = 29) ASARCO high
(n = 26) ASARCO low
(n = 34) Controls
(n = 16) Excuses
Total
12
3
0
1
16
8
3
0
1
12
2
1
0
0
3
3
0
0
1
4
25
7
0
3
35
TABLE 31—All Cases
ARSENIC
>3.05
£3.05
S1.55
23
64
87
NEURO
<1.55
12
6
18
35
70
105
x"= 10.86
p < .005
22
-------
TABLE 32—ASARCO Cases Without Excuses
^1.55
>3.05 23
ARSENIC
£3.05 02
23
NEURO
£1.55
9
3,
12
32
3
35
x2 = 6.53
p<.01
how long it takes for that electrical signal to appear after
stimulating the appropriate nerve but also one assesses
the voltage and the wave form. These measures indicate
whether or not there is any abnormality.
Since this technique has become available, ranges of
normal have been established for men and women and
have to some extent been correlated with age. There is a
suggestion that there is some slowing of conduction ve-
locity with age though the effect of age does not seem to
be very marked.
Those individuals who have a clinical neuropathy
show both subjective and objective signs of decreased
sensation, usually in a stocking and glove distribution, as
well as having evidence of slowing of conduction by
measuring the ulnar sensory conduction velocity, the
motor conduction velocity of the ulnar nerve and the
motor conduction velocity of the common peroneal nerve
as well as the sensory conduction velocity of the sural
nerve. This latter nerve supplies the lateral aspect of the
3.30
Q CONTROL "EXCUSES"
IKAISER
DTACOMA
) CONTROLS EXCLUDING "EXCUSES"
O LOW EXPOSURE
HIGH EXPOSURE W/0 "EXCUSES" (ASARCO)
O ASARCO "EXCUSES"
(n - 110, r = -.444)
HAL
8.40
FIGURE 7. Macro Means
3.313
2.017
I
I
0 1.57 2.67
ARSENIC LOAD
FIGURE 6. Regression Lines by Groups.
leg. It should be pointed out that in this study all workers
did not consider themselves as being sick or patients and
the examiners did not know the source of employment of
each of the individuals they studied. Nonetheless, a num-
ber of individuals show signs of early neuropathy, either
sub-clinical or clinical. The report of the findings, there-
fore, are even more significant and show that these indi-
viduals exposed to an arsenic environment are showing
changes that have as yet not caused any functional defi-
cit, at least as measured by their ability to work and carry
out their occupations. Other causes of neuropathy i.e.
those individuals who might have another explanation
for their neuropathy such as diabetes, alcoholism, frost-
bite, traumatic events, etc. all have been eliminated from
this analysis and the only recognized etiologic factor
other than age and smoking is the exposure to arsenic.
Many questions need to be answered, clearly, the
percentage of individuals who have a neuropathy is
much higher than can be expected in a normal popu-
lation. Since this is only one study, one does not have the
information as to whether these findings represent a
static change or a progressive change. These findings
suggest that further investigations in a serial manner
might answer the question as to whether the chronic
exposure results in any progressive abnormality.
It would appear that neurologic changes have oc-
curred in individuals who are exposed to chronic arsenic
loads though the exact long-range significance of such an
arsenic burden cannot be determined at this point. Cer-
23
-------
12/60
6/60
Inor. Uric
T.P. Alb. Cat i- Phos. Choi. Aold
mt%f
iOj— BT-
Alk. SOOT/
Great. T. Bill. Phoa. CPK LDH 34O
mt% mg% mU.'ml mu./ini mU.>ml mU./ml
15-t— lOi— 350-p- 1200— 600r^ 3DO-|—
I I
t
01 CO.
meq/litei ineq/liter
T.P. Alb. CB +
MEDICARE #73
Inor. Choi. Urlo
Phoa. Aold
Lab. No.
-Tech..
Date
Lab. No.
_Teoh._
M.+
- 11001— 550: —
1000— 50&— 250-—
;; aoo~ 400— 200
700 J— 350; j-
— -- 600 j— 300— ISO- —
Craae. T. Bill. Alk. CPK LDH SOOT/
BUN
10D|—
-t
t
Qlu.
SOOj—
-r
t-
450-
40O —
70r- 350|-
60— 300J-
50J- 250^-
40— 200 —
ISO —
• UN
Data
0—
Qlu.
CM
APPENDIX C.BIood Graph Results
-------
tainly, the clinical neurologic examination and the nerve
conduction velocity studies, both of sensory and motor
nerves, are sensitive means of defining an early change
and are of such a nature that they are reproducible
enough to be able to follow these changes in a sequential
and serial manner.
The implications of this study are: That chronic
arsenic exposure in an industrial setting affects the pe-
ripheral nervous system and the neurologic parameters
involving both clinical and quantitative measurements
are sensitive means of screening the population for
changes related to arsenical neuropathies.
Summary
A pilot study in a copper smelter was undertaken to
determine whether nerve conduction velocity can be
utilized as a biological indicator of the subtle health
effects of long-term exposure to airborne arsenic trioxide
among a sample of residents of Pierce and King Coun-
ties, Washington. The study involved comparing active
working men heavily exposed to arsenic in the work place
with workers not so exposed.
Since in clinical practice arsenic is recognized as
causing a peripheral neuropathy, this study was carried
out to determine whether sustained industrial exposure
to inorganic arsenic was associated with changes in neu-
rologic response. This has not previously been looked for
in workers with sufficient unimpaired ability to perform
their required assignments.
The study team designed a double-blind elec-
trophysiologic and clinical study of the experience of four
groups of workers—two with varying levels of exposure
to arsenic and two sets of controls. Demographic and
other characteristics, including smoking, were obtained
through use of a pre-tested questionnaire. The parame-
ters studied included clinical examination (history and
neurologic examination) and electrophysiologic tests of
the sensory and motor nerve conduction in the upper and
lower limbs. Levels of arsenic and eight other metals
were measured in the blood, hair, urine and nail speci-
mens collected. The study team screened the workers'
blood for carboxyhemoglobin levels, diabetes mellitus
and anemia. The neurological status of the men was
categorized both on the clinical judgement of the exam-
ining neurologist and the objective measurements from
the electrophysiocologic examination. Care was taken in
the analysis not to confuse age effects with other effects.
To reduce the dimensionality of the neurological
measurements, a classification of clinical neuropathy,
sub-clinical neuropathy, and normal was developed.
Men classified as cases of subclinical neuropathy had to
exhibit either one or more reduced velocities, or two or
more nerves with reduced amplitudes. The classification
clinical neuropathy was based on the neurologist's diag-
nosis. All other men were classified as normal.
With this classification into clinical neuropathy, sub-
clinical neuropathy, and normal categories, a statistically
significant association was demonstrated with the sum of
the logarithms of arsenic levels in urine, hair and nails.
This association remained after adjustment for smoking,
"excuses" and average age differences in the categories
of the new classification.
Accordingly, there is unequivocal evidence that a
statistically significant relationship exists between arsenic
exposure and reduced neurological function in this in-
dustrial population.
REFERENCES
1. U.S. Department of Health, Education, and Welfare. National
Institute for Occupational Safety and Health. Criteria for a Recom-
mended Standard .... Occupational Exposure to Inorganic Arse-
nic. New Criteria 1975. 127 pp. HEW Publ. No. (NIOSH) 75-149.
Washington, D.C.: U.S. Government Printing Office (1975).
2. Frost, D. V.: Arsenicals in biology—retrospect and prospect. Fed.
Proc. 26:194-208, (1967).
3. International Agency for Research on Cancer. IARC Monographs
on the Evaluation of the Carcinogenic Risk of Chemicals to Man.
Vol. 2. Some Inorganic and Organometallic Compounds. Lyon:
World Health Organization, IARC, (1973). 181 pp.
4. National Academy of Sciences. National Academy of Engineering.
Committee on Medical and Biologic Effects of Environmental
Pollutants. Subcommittee on Arsenic. Arsenic. Washington, D.C.
(1977). 332 pp.
5. Landau, E.: NAS Report on Arsenic: A Critique, Specialty Confer-
ence on: Toxic Substances in the Air Environment, New England
Section, Air Pollution Control Association, November 1977, (1977).
pp. 65-77.
6. Baetjer, A. M., et al.: Cancer and occupational exposure to arsenic,
p. 393. In abstracts. 18th International Congress on Occupational
Health, Brighton, England, 14-19 September (1975).
7. Reynolds, E. S.: An account of the epidemic outbreak of arsenical
poisoning occurring in beer-drinkers in the north of England and
Midland Counties in 1900. Lancet 1:166-170 (1901).
8. Mizuta, N., et al.: An outbreak of acute arsenic poisoning caused
by arsenic contaminated soy-sauce (shoyu): A clinical report of 220
cases. Bull Yamaguchi Med. Sch. 4(2, 3): 131-150 (1956).
9. Tseng, W. P., et al.: Prevalence of skin cancer in an endemic area
of chronic arsenicism in Taiwan. J. Nat. Cancer Instit. 40:453-463
(1968).
10. Chuang, C.: Interactions of Man and Environment in Taiwan, pp.
52-53, August (1974).
11. Bergoglio, R. M.: Mortalidad por cancer en zonas de aguas arse-
nicales de la Provincia de Cordoba, Republica Argentina. Prensa
Med. Argent. 57:994-998 (1964).
12. Borgono, J. M., and R. Greiber. Epidemiological Study of arse-
nicism in the city of Antofagasta, pp. 13-24. In D. D. Hemphill,
Ed. Trace Substances in Environmental Health-V. Proceedings of
Univ. of Missouri's 5th Annual Conference on Trace Substances in
Environmental Health. Held June 29-July 1, 1971. Columbia:
University of Missouri (1972).
13. deVilliers, A. J. and P. M. Baker: An investigation of the health
status of inhabitants of Yellowknife, Northwest Territories. Occu-
pational Health Division, Environmental Health Directorate, De-
partment of National Health and Welfare, Ottawa, Unpublished
report, undated.
14. Rozenshtein, I. S.: Sanitary toxicological assessment of low concen-
trations of arsenic trioxide in the atmosphere, Hyg. Sanit,
35:16-21 (1970).
15. Canadian Public Health Association Task Force on Arsenic. In-
terim Report. Submitted to the Department on National Health
and Welfare, May (1977).
16. Pinto, S. S.: Mortality experience of arsenic exposed workers. Un-
published data.
17. Milham, S., and T. Strong.: Human arsenic exposure in relation to
a copper smelter. Environ. Res. 7:176-182 (1974).
18. Landrigan, P. J. et al.: Increasing lead absorption with anemia and
slowed nerve conduction in children near a lead smelter. J. Pediat-
rics, 59:904-910, (1976).
19. Gilliatt, R. W. and Sears, T. A. Sensory nerve action potentials in
25
-------
patients with peripheral nerve lesions. J. Neurol. Neurosurg. Psy-
chiat. 21:119-128(1958).
20. Thomas, P. K., Sears, T. A., and Gilliatt, R. W. The range of
conduction velocity in normal motor nerve fibers to the small
muscles of the hand and foot. J. Neurol. Neurosurg. Psychiat.
22:175-181 (1959).
21. Cilliatt. R. W. am'. Thomas, P. K. Changes in nerve conduction
with ulnar lesions at the elbow. J. Neurol. Neurosurg. Psychiat.
23:312-320 (1960).
22. Burke, D., Skuse, N. F., and Lethlean, A. K. Sensory conduction of
the sural serve in polyneuropathy. J. Neurol. Neurosurg. Psychiat.
••77:647-652(1974).
23. Thomas. J. E. and Lambert. E. H. Ulnar nerve conduction velocity
and H-reflex in infants and children. J. Appl. Physiol. 75:1-9
(1960).
24. Harrison, W. W.. et al.: The determination of trace elements in
human hair by atomic abs. spectroscopy. Clinical Chim. ACTA.
2.3:83-90(1969).
25. Harrison, W. W., et al.: The determination of trace elements in
human fingernails by atomic ahs. spectroscopy. Clinical Chim.
ACTA. .37:63-73 (1971).
26. Uthe, J. F. et al.. Mercury Determination in fish samples by wet
digestion and flameless atomic abs. spectrophotometry. J. of Fish-
eries Research Board of Canada. 27:805-811 (1970).
27. Orheim, R. M., et al.: Lead and Arsenic levels of dairv cattle in
proximity to a copper smelter. Environmental Letters.
7(3):229-236 (1974).
28. Malissa, H. and E. Schoffmann: Uber die verwendung von sub-
stituieren dithiocarbomaten in der mikroanalyse. Ill, Mikrochi-
mica Acta, pp. 187-202 (1955).
29. Mulford, C. E. and Perkin-Elmer Corp.: Solvent extraction tech-
niques for atomic abs. spectroscopy. Atomic Abs. Newsletter
(Perkin-Elmer). 5:88-90 (1966).
30. Orheim, R. M. and H. H. Bovee: Atomic absorption determination
of nanogram quantities of arsenic in biological media; Analytical
Chemistry, 46:921 (1974).
31. Guillamin, J. C.: Determination of trace metals in power plant
effluents. Atomic Abs. Newsletter. (.3:135 (1974).
32. Analytical methods for atomic abs. spectroscopy using the HGA
Graphite Furnace; Perkin-Elmer: Puhl. 990-9972. Sept. (1973)
33. Lindstedt: A rapid method for the determination of mercury in
urine. Analyst, .95:264 (1970).
34. Application for Laboratory Accreditation. Prepared for American
Industrial Hygiene Association: Spring (1976).
35. Petren, K»: Etudes cliniques sur I'etologie et les symptomes de
rempoissonnement arsenical du a 1'habitation on des objets de
1'emploi domestiques. (Clinical studies on the etiology and symp-
toms of arsenical poisoning due to home conditions or domestic
objects). Acta Med. Scandin. 53:217-230 (1923).
APPENDIX A. Questionnaire
Application No. 6-289-A, "Association of Nerve Conduction Times and Exposure to Arsenic in ASARCO Employees".
1 Participant number
2. Last Name
3. First Name
01
4. Middle Initial.
5. Street Address.
6. City
7. State.
8. Zip_
9. Name of your Physician.
10. Address of your Physician.
11. Birthdate
12. Date test administered.
13. Have you lived at the above address for more than 1 year?
Yes.
No.
14. If no, what was your previous address?
street city
15. How long did you live at the previous address?..
16. How long ago did you take your urine sample?
17. Have you eaten shellfish in the past 36 hours?
state
zip
Yes.
No.
18. Have you had any recent injuries or operations?
Yes.
No.
26
-------
19. If yes, describe.
20. Have you ever taken any of the following types of medicines?
a. Anti-convulsants Yes.
(anti-seizure medicines)
(Dilantin, Phenurone, Tridione)
b. Anti-tuberculous drugs (INH, PAS,
pyrazinamide, ethionamide) Yes.
c. Pills for diabetes (Dlabinese, Yes
Orinase, Cymelor, Tolinase) 0
d. Blood pressure medicines or water pills Yes
(Aldomet, Diuril, Hydrodiuril) 0
e. Male hormones Yes
(methyltestosterone, oxymetholone) 0
f. Thyroid drugs (Tapazole, Yes
propylthiouracil) 0
g. Antidepressants or major
tranquilizers (not including
barbituates, valium, librium,
etc.)
(Thorazine, Sparine, Mellaril,
Stelazine, Compazine, Niamid, Yes
Nardil, Marplan, Parnate; 0
Tofranil, Elavil)
21. Have you ever taken other medications (not covered in the preceding list)
for longer than 2 weeks at a time.
No.
No
No
No
No
No
1
1
1
1
No.
Don't Know.
Don't Know.
Don't Know.
Don't Know.
Don't Know.
Don't Know.
Don't Know.
Yes.
No.
22. If yes, list below
Name of
drug
year
prescribed
length of time
drug taken
associated
medical problems
23. Have you ever taken medicines containing arsenic?
Yes.
No.
Don't Know.
24. Have you ever had a substantial weight loss (more than 10 Ibs) in
the last three months?
Yes.
No.
Don't Know.
25. If yes, what is the reason?
Special diet.
Don't Know.
26. Do you have any history of liver disease?
Yes.
No.
Don't Know.
27. Do you drink alcoholic beverages now?
Yes.
No.
28. If yes, have you consumed any alcohol in the past 48 hours?
Yes.
No.
27
-------
29. If yes, what and how much?
30. Do you currently smoke cigarettes?
Yes.
No.
31. If yes, how many per day?
less than Vz pack.
1/2 to 1 pack
1 to 2 packs
over 2 packs
32. Do you get your drinking water from a well?
Yes.
No-
Don't Know.
33. Are you personally applying insecticides in your garden?
Yes.
No.
Don't Know.
APPENDIX B. Clinical Work Sheet-
Name
Date: _
Other Factors:
Trauma
Metabolic Disturbances
Nutritional Disburbances
Alcohol Consumption
Medication:
Yes No Duration
Age
D.O.B:.
In:.
No: .
Sex
Duration in present position:.
Neurological Signs:
Numbness
Painful/Burning
Pinprick
Vibration Sense
Muscle Weakness
RUE LUE LUE ELE
Reflexes Bl TR Sup. K A P
Left
Right
Room Temp:.
Skin Temp:.
NERVE: ULNAR R.
APPENDIX C. Nerve Conduction Velocity Testing
In:
No:
DISTAL LATENCY
M.C.V.Ab. Elb-Wr. (_
28
-------
SENSORY
ACTION
POTENTIAL
NERVE: COMMON
MOTOR
NERVE: SURAL
SENSORY
ACTION
POTENTIAL
Amplitude (neg-phase)
Stimulation
Stimulation
Decrement
Amplitude
Form
Latency to Peak
Latency to Onset
Velocity to Onset
PERONEAL R.
DISTAL LATENCY
M.C.V.
MAP.
Amplitude (neg-phase)
Stimulation
Stimulation
Decrement
AMPLITUDE
Form
Latency to Peak
1 atency tn Onset
Velocity to Onset
73
APPENDIX D. Coefficient of Each Metal
Coefficient of variation among blind and replicated
BLOOD
Blind 65.8%
Pb
Replicate 11.0%
Blind )
Hg ) None
Replicate )
Blind )
As ) None
Replicate )
Blind 39.2%
Zn
Replicate None
Blind 11.9%
Cu
Replicate 14.8%
Blind 18.1%
Ni
Replicate 17.4%
Blind 69.2%
Cd
Replicate 21.1%
-Blind ) None
Sb )
Replicate )
Blind 4.0%
Se
Replicate None
\ oi 1 1)
( }
1 r.m\
I r.m\
( rm)
( cm) •
( )
{ rm)
{ rm)
samples for each metal: specimen set of data
URINE NAILS HAIR
68.4% 42.3% 76.2%
8.6% 5.9% 5.5%
13.8% 68.1% 53.8%
5.4% None None
2.3% 40.7% 40.8%
None 8.6% 7.0%
10.7% 15.5% 25.8%
None 3.1% 5.0%
53.6% 34.4% 35.5%
6.1% 6.3% 7.0%
None 48.7% 78.1%
16.7% 15.9% 15.0%
75.6% 62.3% 60.5%
12.2% 11.1% 8.6%
0 34.0% 69.2%
None 9.4% 6.6%
1.6% 49.8% 62.8%
32.2% 12.3% 18.9%
29
-------
APPENDIX E. Cations in Biological Specimens
Normal levels and mean levels of cations found In biological specimens
Normal
Pb
Found
Normal
As
Found
Normal
Cd
Found
Normal
Zn
Found
Normal
Cu
Found
Normal
Hg
Found
Normal
Ni
Found
Normal
Sb
Found
APPENDIX F. Assay of Biological
BLOOD
ppb
(100-350)
144+126
(10-640)
1.5+8.9
(.002-.007)
(900-9000)
8480 + 2550
(1230)
968+307
(75-90)
:
—
Specimens
URINE
PPb
(10-35)
13.9 + 13.4
(15-60)
38.2 ±73.3
(1.4-30)
2.09 + 2.97
(630)
627 ±385
(10-50)
80 + 284
(2-20)
2.06 ± 3.48
:
—
HAIR
ppm
(0.5-360)
123 + 251
(.045-1.7)
60.0+122.3
(0.20-1.0)
1.07 ±1.92
(150-200)
215.7 ±136.4
(2-20)
5*8.8 ±116.4
(3-40)
.0
13.1 ±48.6
(7.9-9.7)
5.5 ± 11.4
NAILS
ppm
5-6
7.9 ±11.7
(.080-5.5)
29.3 ±63.3
0.1 ±10
0.32 + 0.80
(180-220)
100.4 ±40.4
(45-60)
17.2 + 33.7
1.0
0.5
9.4±51.1
2.0
0.78 ±1.79
Sample and digestate size, normal values, analytical method and detection limits pertaining to biological specimens assayed during
the Tacoma Study.
2-4 ml
10ml
50-100 mg
50-100 mg
Analytical Volume
normal levels (ppb)
ARSENIC
detection limit (ppb)
MERCURY
NICKEL
COPPER
ANTIMONY
SELENIUM
CADMIUM
50ml
BLOOD
100-500
Arsine Generator
3-5
75-90
Cold Vapor
0.1-0.3
1.0-4.2
APDC/Furnace
2.3
1230
C,Hj Flame
10-20
APDC/Furnace
1-2
200
APDC/Furnace
.002-.007
APDC/Furnace
.01-.03
10ml
URINE
15-60
Arsine Generator
5-10
2-20
Cold Vapor
0.1-0.3
11
APDC/Furnace
10-15
10-50
C,H, Flame
10-20
APDC/Furnace
2-5
10-150
APDC/Furnace
1.4-30
APDC/Furnace
.01 -.03
10ml
HAIR
45-1700
Furnace
2-3
3,000-40,000
Cold Vapor
0.1-0.3
1000-4000
Furnace
2.5-5
2,000-20.000
C,H, Flame
20-50
7900-9700
Furnace
0.5-1
300-6000
Furnace
10-20
200-1000
Furnace
.03-05
20ml
NAILS
80-5600
Furnace
1-2.5
Cold Vapor
0.1-0.3
Furnace
2.5
45,000-60,000
C,H, Flame
30-50
Furnace
0.5-1
1,140
Furnace
20-50
Furnace
.02-.05
Normal Levels, Misc. Media
Soil
Hair
Feces
Urine
Air
Blood
Plants
Soil
Urine
Air
Air
Exposed Urine
Milk
Soil
Plants
500-14,000
18,000-30,000
lO^g/day
0.5 ^g/day
20 ng/ms
110
5,000-20,000
30,000
20-60 /jg/day
2 no/m9
3 ng/ma
600
17-30
60
500
30
-------
900-9000
ZINC CjHj Flame
30-50
100-350
LEAD APDC/Furnace
or
Furnace
0.5-1
630
C2H2 Flame
50-100
10-35
APDC/Furnace
1-3
150,000-200,000
CjH, Flame
20-30
500-360,000
Furnace
0.5-1
180,000-220,000
CjH2 Flame
20-30
Furnace
30-50
Serum
Liver
Bone
Air
Prostate
Retina
Intake
Soil
Plants
Air
Gas
Intake
800-1600
60,000
60,000
0.lM9/m3
860,000
500,000
1 2 mg/day
12,000
100,000
4/zg/m3
3.8g/imp. gal
300 Mg/day
APPENDIX G. CLINICAL RESULTS
Jse#
iS
!
'<3
O.
IB
T)
C
TO
C
'c
CD
T>
1
s
I
Q.
Q.
=J
Results
1
1
3
1
0
0
0
4
1
0
0
1 (diabetic)
1 (alcohol)
2
2
0
0
0
1
0
0
0
1 (frostbite)
0
0
4
0
0
1
0
0
0
0 (alcohol)
0 (alcohol)
0
1
0
0
0 (diabetic)
0
0
0
0
Case#
Q.
"o
I
Q.
13
_C
D>
C
CO
°
•g
Q.
Q.
OT
Results
0
1 (alcohol)
0
0
0
0
0
0
0
0
3 (alcohol)
4
0
1
0
0
4
1
0
3
1
1
0
0
0
0
4
0
0
1
0
1
1
1 (diabetic)
3
1
0
4
0
1
4
0 -
1 (alcohol)
0
1 (alcohol)
Case#
i
Q.
"o
a
o>
c
c
CD
•o
1
Q.
a
OT
Results
0
1
1
4
4
0
0
0
0
0
0
0
0
0
0
0
0
3
1
0
0
0
0
0
0
0 (alcohol)
1
4
0
0
1 (alcohol)
1
1 (alcohol)
2
0 (spondylosis)
CODES
0 = NO NEUROPATHY
1 = SENSORY NEUROPATHY
2 = MOTOR NEUROPATHY
3 = SENSORY-MOTOR
NEUROPATHY
4 = RETEST
31
-------
APPENDIX H. Electrical Values
MOTOR
ULNAR
MCV
Vel W-E
55 m/s
50 m/s
62 m/s
49 m/s
50 m/s
46 m/s
46 m/s
55 m/s
63.5 m/s
58 m/s
58 m/s
60 m/s
48.2 m/s
58.6 m/s
52 m/s
56 m/s
58 m/s
55 m/s
61 m/s
52.8 m/s
58 m/s
55 m/s
50 m/s
56 m/s
64 m/s
54 m/s
60 m/s
67.6 m/s
61 m/s
51 m/s
49 m/s
54 m/s
54 m/s
55 m/s
64 m/s
54 m/s
55 m/s
52 m/s
50.5 m/s
57 m/s
59.5 m/s
51 m/s
53 m/s
57 m/s
57 m/s
56 m/s
56 m/s
53 m/s
56 m/s
49 m/s
56 m/s
59 m/s
58 m/s
53 m/s
57 m/s
56 m/s
58 m/s
52 m/s
58 m/s
55 m/s
48 m/s
58 m/s
MCV
Vel E-A
60 m/s
66 m/s
62 m/s
55 m/s
72 m/s
59 m/s
63 m/s
65 m/s
59 m/s
57 m/s
64.2 m/s
63 m/s
72 m/s
62.5 m/s
61 m/s
71 m/s
69.2 m/s
60 m/s
67 m/s
70 m/s
54 m/s
66 m/s
79 m/s
75 m/s
56 m/s
64 m/s
64 m/s
70 m/s
64 m/s
56 m/s
55 m/s
77 m/s
66 m/s
59 m/s
53 m/s
51 m/s
48 m/s
54 m/s
64 m/s
70. 5 m/s
55 m/s
40 m/s
51 m/s
57 m/s
55 m/s
59 m/s
78 m/s
63 m/s
71 m/s
65 m/s
55 m/s
60 m/s
50 m/s
50 m/s
72 m/s
56. 5 m/s
75 m/s
59. 5 m/s
77 m/s
23.7 m/s
23 m/s
64 m/s
MAP
Dura
13.2
17. 5 m/s
12.7
16.7
14
11.5
11.2
12 m/s
12 m/s
10. 7 m/s
11.4
16
8.7
10.9
13.9
11. 2 m/s
11. 2 m/s
14.3
9
12.7
10.7 m/s
13 m/s
14.8 m/s
10.4
12.8
12
12.2 m/s
14.2
11.2
14
13
13 m/s
12.9
12.1
13.3
14.3 m/s
11.3
13
13
9 m/s
12. 9 m/s
14
13. 5 m/s
12 m/s
10.9
10 m/s
14.1 m/s
13.2
14.6
16.7
13.3
14.4
13
11.5
14.2
9.5
11.5
12.5
11.7
11. 8 m/s
13 m/s
14.7 m/s
W
10.5
10
5
11.5
8.5
11
4
7.5
6.5
10
8.5
12.5
4
9
12.5
12
10
10.5
13
9.5
4
8
14
8
8.5
8
13
10
11.3
11.5
14.5
8.5
11.5
9>
8*
15.5
9.5
13.5
11
3.2
12
7.5
9.5
3
9.5
5.2
12
12.5
2.2
11.5
16.5
7
8.5
10.7
9
12
5.2
2.4
8
8.5
12.5
9.2
AMPL.
E
9.5
9.3
3.4
11
8.5
9
4
7.7
6.7
9.5
7.0
12.5
3
8.0
11.3
10.5
9.7
9
12.5
8.5
4.5
8.7
14.6
6.5
8.7
6.5
12
10
10.5
11.5
13.5
7.5
10
9
8.5
15.5
8.5
13
10.5
2.8
12
8
9.0
3.2
8.6
4.4
11.5
10
1.8
9
15.5
7
8.5
7.5
9
11
6
2.4
4.5
8.5
12
11
A
9.2
8.3
4
10.5
8.5
8.5
4
7.7
5
9
6.5
11.5
4
7.5
10.5
9.2
10
8.5
12
8.5
4.5
8.3
14.6
6.5
6.5
6.5
12
10
9.5
11.5
13.3
7.5
10
9
7.5
14.5
8.5
13
10.5
2.0
6.5
8.3
9.0
3.2
6.8
4.4
11
10
1.8
9
14.5
7
7.5
8
7.5
11
4.5
2.2
3.5
7.5
11.5
10
SENSORY
VEL. AMPL.
30.5 m/s
43 m/s
55 m/s
50 m/s
54 m/s
46 m/s
41 m/s
46 m/s
70 m/s
63 m/s
44 m/s
55 m/s
57.5 m/s
52 m/s
46 m/s
42 m/s
53 m/s
52 m/s
47.5 m/s
50 m/s
50 m/s
56 m/s
48 m/s
50 m/s
49 m/s
47 m/s
50 m/s
55 m/s
50 m/s
49 m/s
46 m/s
45 m/s
61 m/s
57 m/s
55 m/s
43 m/s
52.6 m/s
42 m/s
absent
43.5 m/s
49.8 m/s
48 m/s
40 m/s
40.5 m/s
52.2 m/s
55 m/s
45 m/s
52 m/s
56 m/s
46 m/s
48 m/s
60 m/s
42 m/s
55 m/s
48 m/s
63 m/s
45 m/s
47.5 m/s
40 m/s
5 1.8 m/s
40 m/s
50 m/s
8 u
4.5 u
11 u
13 u
12 u
6 u
6 u
7 u
9 u
14 u
9 u
10 u
6 u
3 u
*18 u
9 u
4 u
5 u
24 u
11 u
10 u
7 u
10 u
8 u
6 u
13 u
17 u
11 u
6 u
16 u
6.5 u
4 u
12 u
6 u
7 u
14 u
5 u
8 u
absent
3 u
5 u
7 u
4 u
3.5 u
6 u
4 u
11 u
8 u
9 u
16 u
13 u
13 u
5 u
6 u
5 u
5 u
10 u
10 u
4 u
3 u
4 -u
10 u
MOTOR
PER-
ONEAL
VEL.
46.8 m/s
44 m/s
50 m/s
42 m/s
39 m/s
43 m/s
45 m/s
52 m/s
48 m/s
46 m/s
35.7 m/s
50 m/s
60 m/s
40.7 m/s
50 m/s
48 m/s
52 m/s
47 m/s
49 m/s
37.6 m/s
40 m/s
46 m/s
55 m/s
54 m/s
55 m/s
38 m/s
53 m/s
48.6 m/s
45 m/s
53 m/s
47 m/s
46 m/s
43 m/s
45 m/s
43 m/s
45 m/s
54 m/s
47 m/s
41. 7 m/s
40.5 m/s
42.5 m/s
44 m/s
39 m/s
34.5 m/s
45 m/s
52 m/s
39 m/s
47 m/s
47 m/s
47 m/s
45 m/s
47 m/s
48 m/s
45 m/s
36.9 m/s
55.5 m/s
51 m/s
48 m/s
46 m/s
44. 6 m/s
38.3 m/s
46 m/s
AMPL.
Ankle Knee
1.5
2.5
3
4
7.4
7.7
6.8
.8
1.8
9.5
3.7
1.6
10
1.5
5.2
2.1
7.2
1.6
5.6
4
6.6
2.5
2.5
6.0
9.2
3.4
7
4
2.5
1.9
3.05
6.4
4.4
8
10.7
10
10
7.8
3.4
4.8
4.7
4.4
3.2
2
2
5.6
7.0
7
3.7
2.7
1.2
5
4.6
4
2.8
8.5
7.7
4.8
1.7
3.8
1.2
2.5
2
2.9
3
3.2
6.6
6.6
6.2
1
1.2
7.5
3.7
2
9
2.8
4.2
1.6
6.5
1.6
5.6
3
6.2
2.5
1.7
5.0
9.5
2.6
7
4
2.5
1.7
3.5
6.4
3.4
8
9
9.5
10
6.6
3.0
4.0
4.7
5.1
3.0
3.4
2
5.2
6.0
8
3
2.1
1.1
5
6.0
3.8
1.0
7.5
6.7
4.4
1.2
3.6
0.5
2.6
SENSORY
SURAL
VEL.
41 m/s
40 m/s
48 m/s
36 m/s
41 m/s
44 m/s
41 m/s
42 m/s
40 m/s
36.7 m/s
40 m/s
40 m/s
40 m/s
46.6 m/s
39 m/s
39 m/s
41 m/s
42 m/s
48 m/s
41.1 m/s
38.8 m/s
38 m/s
44 m/s
44 m/s
42 m/s
41 m/s
43 m/s
50 m/s
50 m/s
48 m/s
40 m/s
42 m/s
45 m/s
43 m/s
44 m/s
44 m/s
43 m/s
41 m/s
33 m/s
35 m/s
42 m/s
45 m/s
31 TTI/S
33.5 m/s
40 m/s
42.5 m/s
45 m/s
40 m/s
45 m/s
44 m/s
42 m/s
48 m/s
36 m/s
47 m/s
37.8 m/s
45 m/s
45 m/s
43 m/s
39 m/s
41 -m/s
36 m/s
42 m/s
AMP.
15 u
13 u
18 u
21.5u
24 u
8 u
10 u
6 u
9 u
21 u
3 u
16 u
6 u
6 u
20 u
13.5u
12.5u
8 u
14 u
4 u
5 u
13.5u
11 u
36 u
6 u
11 u
12 u
11 u
14 u
15 u
10 u
13 u
12 u
19 u
16 u
7 u
7 u
6.5 u
11 u
11 u
13 u
7 u
3 u
9 u
5 u
22 u
14 u
14 u
9 u
12 u
7.5 u
19 u
5 u
20 u
11 u
7 u
18 u
18 u
1 u
8 u
7 u
15 u
32
-------
APPENDIX H. Electrical Values
MOTOR
ULNAR
MCV
Vel W-E
52 m/s
47 m/s
51 m/s
49 m/s
57 m/s
72.5 m/s
55 m/s
57 m/s
61 m/s
54 m/s
53 m/s
50 m/s
53 m/s
56 m/s
48.2 m/s
50 m/s
58 m/s
58 m/s
58 m/s
57 m/s
56 m/s
49 m/s
57 TT1/S
54 m/s
56 m/s
55 m/s
59.5 m/s
52 m/s
51. 8 m/s
61 m/s
53 m/s
54 m/s
54 m/s
58.6 m/s
58 m/s
52 m/s
58 m/s
58 m/s
47.5 m/s
47 m/s
56.7 m/s
49 m/s
48 m/s
56 m/s
54 m/s
53 m/s
64 m/s
56 m/s
61 m/s
58 m/s
52 m/s
48 m/s
55.8 m/s
60 m/s
53 m/s
53 m/s
50 m/s
54 m/s
54.5 m/s
54 m/s
57 m/s
52 m/s
MCV
Vel E-A
59 m/s
75 m/s
60 m/s
58 m/s
52 m/s
61. 3 m/s
63.5 m/s
73 m/s
67 m/s
57 m/s
64 m/s
57 m/s
50 m/s
57 m/s
71. 5 m/s
61 m/s
57.5 m/s
63 m/s
63 m/s
49 m/s
61 m/s
50 m/s
66 m/s
62 m/s
70 m/s
69 m/s
73 m/s
55. T m/s
58 m/s
65 m/s
61 m/s
41 m/s
70 m/s
85 m/s
63 m/s
49 m/s
70 m/s
48 m/s
66 m/s
57.5 m/s
48 m/s
57 m/s
55 m/s
70 m/s
64 m/s
67 m/s
47 m/s
60.5 m/s
67 m/s
78 m/s
70 m/s
53 m/s
62 m/s
61 m/s
66 m/s
62 m/s
53 m/s
60 m/s
71. 6 m/s
53 m/s
53 m/s
MAP
Dura
13. 6 m/s
11. 5 m/s
15.7
13.6
13.2
13.3
12 m/s
13.8
12.8
11
9.1
11.5
14
12
11.4
11.9
12
12.3
17
13.2
14.7
12.5
13.2
14.3 m/s
9.7 m/s
13. 7 m/s
14.3 m/s
12.2
10.8
21 m/s
10.3
11.3
16.7 m/s
12.1
13.1
9.9
14 m/s
13.4 m/s
15.3
10.9
9.5
12.4 m/s
15. 5 m/s
13.2 m/s
10. 2 m/s
13 m/s
14. 8 m/s
14. 2 m/s
13.9
12 m/s
15.3
10 m/s
9.6 m/s
17.4 m/s
13.1
9
10.3 m/s •
11
12. 9 m/s
14 m/s
15. 7 m/s
10. 8 m/s
W
9
12.5
8
12
6.5
6.5
12
9
13.5
8
10
12
8.5
10
5
12
10
11.5
9.8
7.5
12
8
11.5
8.5
12.5
10
15
10
11.5
12.5
9.5
7.5
11.7
15
6.5
10
17
13.5
6
6.5
14
6.5
8.5
8.5
7
16
8.5
9
15.5
7.6
13
13.5
10
8
15
10
13
10
13
8.5
10.5
9
AMPL.
E
9
12.4
8
12
6.5
7
12
9
12.5
7.5
10
8
10
10
6
11
9.5
11
10
7
11.5
7.5
12
8
12
10
13.8
11
13.5
12
9
5
11.3
15
6.5
10
15
12
5
6.5
14
5
8.2
8
6
16
8.5
8.5
15.5
6.6
10
12
10
7.5
15
10.5
12
7
7.6
8.5
11.5
8
A
8.5
12.4
8
10
6.5
7
12
9
12.3
7.5
9
9
10.5
10
5.5
10.5
9.5
10.7
9.6
6.5
6
10.7
8
10
10
13
10.5
13
12.5
9.5
5
11.5
14
6.0
10
13
12
5
6.5
13.5
5
8.6
8
6
15
8.0
8.7
15.5
6
12
12
9.5
7.5
14
10
12
6.5
6.8
7.5
11
7
SENSORY
VEL. AMPL
48 m/s
42 m/s
46 m/s
50 m/s
45 m/s
62.5 m/s
63.7 m/s
55 m/s
46 m/s
50 m/s
55 m/s
53 m/s
46 m/s
52 m/s
57.5 m/s
52 m/s
45.8 m/s
57 m/s
44 m/s
54 m/s
41 m/s
46 m/s
47 m/s
50 m/s
52 m/s
50 m/s
51 m/s
50 m/s
47.8 m/s
50 m/s
55 m/s
52 m/s
57 m/s
57.8 m/s
55 m/s
55.5 m/s
42 m/s
52 m/s
50 m/s
52 m/s
60.5 m/s
52 m/s
45 m/s
46 m/s
52 m/s
41 m/s
50 m/s
52 m/s
48 m/s
78 m/s
48 m/s
52 m/s
57.8 m/s
48 m/s
45 m/s
48 m/s
54 m/s
54 m/s
56 m/s
47.9 m/s
53 m/s
46 rn/s
10 u
5 u
9 u
9 u
10 u
4 u
12 u
6 u
6.5 u
7 u
14 u
13 u
6 u
15 u
6 u
10 u
9 u
7.5 u
6 u
14 u
13 u
5 u
4 u
14 u
7 u
4 u
24 u
8 u
6 u
14 u
6 u
8 u
5 u
10 u
5 u
8 u
3 u
8 u
5 u
6 u
9 u
4 u
7.5 u
5 u
9 u
3 u
10 u
2 u
8 u
13 u
8.5 u
10 u
12 u
9 u
18 u
7 u
7 u
4 u
8 u
11 u
5.5 u
11 u
MOTOR
PER-
ONEAL
VEL.
47 m/s
41 m/s
44 m/s
43 m/s
48 m/s
47 m/s
44.5 m/s
46 m/s
44 m/s
41 m/s
50 m/s
44.5 m/s
48 m/s
50.5 m/s
43.5 m/s
38 m/s
39.3 m/s
49 m/s
41 m/s
46 m/s
48 m/s
43 m/s
55 m/s
47 m/s
47 m/s
44 m/s
47 m/s
50 m/s
46. 7 m/s
42 m/s
47 m/s
44 m/s
66 m/s
45.9 m/s
43 m/s
49 m/s
37 m/s
44.6 m/s
45.5 m/s
38 m/s
45.9 m/s
44
47 m/s
42 m/s
46 m/s
45 m/s
47 m/s
43 m/s
43 m/s
48 m/s
42 m/s
50 m/s
45.8 m/s
46 m/s
50 m/s
43 m/s
40 m/s
44.5 m/s
53.2 m/s
45.8 m/s
42 m/s
52 m/s
AMPL.
Ankle Knee
5
1.9
2.3
4.8
9
6
11.5
1.5
3.2
1.1
2.8
9
5.4
2.5
1
3.8
7.5
14
7.8
8
3.4
1.7
6.9
3.8
12.5
3
9.2
10.5
9.5
1.3
2.5
6
2
8.5
5.7
3.8
0.8
2.0
3.6
5.6
7
7.5
4.6
2.1
7.0
2.5
3.8
8
8.8
4.5
1.7
12.5
5.4
7.5
10 me
5.8
6.
4.8
6
7.4
3.2
3.1
5
1.4
1.9
4.6
5
10.7
1.3
2.8
1.1
2.5
7.5
4.9
2.5
1
3.2
7
13.5
7.3
6
2.9
1.6
5.4
3.8
12.5
2
8.5
9.5
8.0
1.6
2.4
4.3
2
7.5
5.7
3.6
0.75
2.6
3.6
4.0
7
7.5
5.3
1.2
6.5
4.2
2.8
8.5
7.8
4.7
1.6
11.5
5.4
6.5
10 me
5.8
5.6
4.8
5.4
6.4
2.9
2.5
SENSORY
SURAL
VEL.
38 m/s
38 m/s
41 m/s
45 m/s
40 m/s
40 m/s
40.7 m/s
42 m/s
39 m/s
42 m/s
42.5 m/s
41. 5 m/s
42 m/s
36.8 m/s
41 m/s
40 m/s
40 m/s
40 m/s
44 m/s
42 m/s
54 m/s
38 m/s
47 m/s
43 m/s
37 m/s
43 m/s
44 m/s
48 m/s
40 m/s
45 m/s
50 m/s
38 m/s
41 m/s
42.4 m/s
35 m/s
50 m/s
38 m/s
43 m/s
43 m/s
43 m/s
48.2 m/s
35 m/s
42 m/s
42 m/s
40 m/s
34 m/s
42 m/s
47 m/s
42 m/s
56 m/s
44 m/s
45 m/s
43.7 m/s
42 m/s
42 m/s
39 m/s
43 m/s
43 m/s
41 m/s
42 m/s
44 m/s
40 m/s
AMP.
15.5u
6 u
13.5 u
10 u
19 u
19 u
15 u
10 u
12 u
6 u
8 u
6 u
8.5 u
5 u
34 u
24 u
9 u
12.5u
6.5 u
8 u
11 u
9 u
15.5u
12 u
12 u
8 u
11.5u
6.5 u
3 u
10 u
17 u
8 u
10 u
6 u
12 u
15 u
3 u
10 u
14 u
4 u
7 u
8 u
9 u
13 u
2 u
6 u
18 u
16 u
13 u
9 u
10 u
12 u»
6 u
13 u
16 u
10 u
16 u
9 u
4 u
*T U
19 u
12 u
3.5 u
33
-------
APPENDIX I. Normal Values
ULNAR
SENSORY
VELOCITY
S.A.P. AMPLITUDE
MOTOR
VELOCITY
Upper Arm
Fore Arm
M.A.P. AMPLITUDE
COMMON PERONEAL
MOTOR
VELOCITY
Leg
MAP. AMPLITUDE
SURAL
SENSORY VELOCITY
S.A.P. AMPLITUDE
APPENDIX Ja. ASARCO Employee Record
(50-70) M/S
( 8-28 uV)
(52.8-74.0) M/S
(47 -65.4) M/S
( 5.6-20.8) mV
(41-59) M/S
(2.2-14.8) MV
(40-54.7) M/S
( 6-42 uV)
1967—142
EMPLOYEE'S WORK RECORD
Key
Seniority or
Date Hired
Occupation
Foreman
Rate
Day Went
to Work
Left
Key
Remarks
34
-------
MILITARY AND DRAFT STATUS RECORD
Branch of Service Army
Years in Service
Rank Sp/4
Special Awards
Draft Status
Local Board
Serial N umber ER 56380384
VACATION AND LEAVE OF ABSENCE RECORD
From
To
Reason
From To
Reason
From To Reason
ASARCO HEALTH PLAN:
Key E = Employed
DR = Drafted
O = Laid Off
Q = Quit
T = Transferred
R = Rehired
D = Discharged
N = Dropped
APPENDIX Jb. ASARCO Employee Record
APPENDIX Jb-Tacoma Plant
URINARY ANALYSES—METALS
Last Name
First Name
Initial
Soc. Sec. No.
Plant No.
Date
ELEMENT—Microgram/Liter
As Pb Hg Sb Se Cd
Department
35
-------
------- |