Health Effects of

Contaminants in
Aircraft Cabin Air
Summary Report v2.7

Professor Michael Bagshaw

MB BCh MRCS FFOM DAvMed DFFP FRAeS

https://www.airpilots.org/file/1277/air-contamination-health-effects-report-oct-13.pdf
http://www.asma.org/asma/media/asma/Travel-Publications/Health-Effects-of-
Contaminants-in-Aircraft-Cabin-Air-Report-v2-5-Aug13.pdf

April 2014
1
HEALTH EFFECTS OF CONTAMINANTS IN CABIN AIR (version 2.7)

BACKGROUND SUMMARY – April 2014

INTRODUCTION

The occupants of commercial aircraft are protected from hypoxia by the

aircraft cabin being pressurised to below 10,000 feet. The maximum certified
cabin altitude will not exceed 8,000 feet during normal operations, to provide a
safety margin for those who might be cardio-respiratory compromised. During
flight, air is derived from the compression stage of the jet engine or, in the
case of the B787, from electrically driven compressors. This bleed air is
conditioned and filtered, with an exchange of 10-15 times per hour with
outside air and 20-30 times per hour including outside and filtered recirculated
air. Pressurisation to sea level, although ideal, is not technologically and
economically feasible.

Concerns have been raised by organisations representing pilots and cabin

crew about the possible effects on aircrew health of oil/hydraulic fluid
smoke/fume contamination incidents in pressurised aircraft. Specific concerns
have been raised with respect to organophosphate compounds (OPs) in the
cabin air environment and the perceived effects on health of long term low-
level exposure (1, 2).

Some aircrew who report incidents experience a variety of symptoms, mainly

acutely irritant in nature. Less frequently, some aircrew report longer-term
symptoms. However, the epidemiological evidence is hampered by
inconsistency in reporting and the numbers are small.

The Global Cabin Air Quality Executive (GCAQE) was established in 2006 to
deal specifically with contaminated air issues and cabin air quality affecting air
crew (1, 2). It claims to represent more than 20 organisations worldwide,
although it appears to have no registered articles of association. The
organisation has a high media and political profile in the UK and other parts of
Europe.

The UK CAA Mandatory Occurrence Reporting (MOR) System in 2007 noted

116 fume event reports out of 1.3 million passenger and cargo flights, with
fume events estimated to occur on 0.05% of flights overall (1 in 2000). Of the
approximately 20,000 UK professional pilot population, the UK CAA Medical
Department in 2013 were aware of 28 individual cases who have reported
symptoms which they attributed to exposure to fumes. Of these 14 have
returned to flying or were never assessed as unfit. The remaining 14 remain
unfit (the CAA no longer differentiates temporary/long term unfit), although a
number would have passed normal retirement age. No new cases have been
documented since June 2012 and the CAA is not aware of any other on-going
cases (26).

2
The Australian Parliament conducted a Senate Investigation in 1999 into air
safety and cabin air quality. This followed concerns raised by crew members
working for Ansett Airlines who reported feeling unwell due to unpleasant
odours of engine oil inside BAe 146 aircraft. The Senate report concluded that
the BAe 146 had a record of unpleasant odours in the cabin as well as
occasional incidents of fumes from lubricating oil. Over a longer period, airline
employees had reported a variety of adverse health effects.
In response to the enquiry, BAe redesigned the original air circulation system
in the BAe 146. A number of health compensation claims were also filed
against Ansett but no damages were awarded at that time. However, in 2010
the Australian Dust Diseases Tribunal awarded damages to a former cabin
attendant for lung damage related to a single fume event. No other symptoms
of “aerotoxic syndrome” were at issue in that case, nor was the claim related
to exposure to neurotoxins.

In the USA similar problems were reported with early RB211-535C powered
Boeing 757 aircraft in which overfilling with engine oil could lead to
contamination of the environmental conditioning system (ECS).

Also, in the UK, incidents of smells in the cabin were reported on early B757s
operated by British Airways, and UK operators of the BAe 146 also
experienced oil fume incidents.

Although the evidence suggests that oil fume events of initial concern stem
from a design fault on two early series aircraft which has now been rectified,
occasional oil smells still occur (~1 in 2000 flights) and campaigners maintain
that these are leading to health problems for aircraft occupants. They are also
concerned that crew health is being affected by long term exposure to very
small amounts of contaminants which may be present in bleed air as a result
of leaking engine oil seals, in the absence of specific fume events (1, 2).

These concerns have led to a number of governmental, scientific and industry

reviews and investigations of the issue over the past decade.

Organophosphates and their use in Engine Oil

Jet engine oils contain synthetic hydrocarbons and additives, including an

organophosphate known as tricresyl phosphate (TCP) which acts as a high
pressure lubricant.

The term organophosphate encompasses a variety of chemical compounds

with a similar structure. Small differences in this structure alter the chemical
properties of the compound and thus any associated health effects.

TCP is a toxic mixture that can cause a wide array of transitory or permanent
neurological dysfunction when swallowed in sufficient quantity.

3
The neurotoxicity of TCP is due to its ortho isomers. The major toxic effect of
the ortho isomer, ToCP, is impairment of neuromuscular and peripheral nerve
synapse function; it is thought to have no toxic effect on centrally mediated
cognitive function. Other ortho isomers of TCP include MoCP (mono-ortho-
cresyl phosphate) and DoCP (di-ortho-cresyl phosphate) which have similar
toxicity.

The para and meta isomers are not toxic to humans. There have been no
independently peer-reviewed recorded cases of neurological harm in humans
following dermal or inhalation exposure to TCP, although cases have been
reported following ingestion (swallowing) of the ortho isomer.

An unpublished report commissioned by British Airways Health Services from

the Medical Toxicology Unit at Guy’s Hospital in 2001 stated that “the majority
of cases of tricresyl phosphate poisoning have been associated with the
swallowing of contaminated food or drink, not with occupational exposure.
The most frequent occupational exposures occur during manufacture,
packaging, shipping and storage, not at the point of product use, and reports
of occupational intoxication are rare”. The report authors researched all
documented exposures dating back to 1943 and they were all to high
concentrations greatly in excess of the amount present in jet oil.

The reported concentration of TCP used in most aircraft engine oils is less
than 3%, of which the ortho isomers constitute less than 0.2% of the total
TCP. This results in an overall concentration of ortho isomers of less than
0.006% of the total engine oil.

Consequently TCP mixtures used in engine oils are significantly less toxic
than pure ToCP (the tri-ortho-cresyl phosphate), for which an Indicative
Occupational Exposure Limit Value (IOELV) threshold limit value is set at 100
µg/m³ as an 8h time-weighted average (TWA), with an emergency 15min
short-term exposure limit of 300 µg/m³. This is equivalent to the North
American occupational exposure limit of 0.1 mg/m 3.

Peer reviewed studies have indicated total TCP concentrations on aircraft

during abnormal oil smell conditions significantly below this threshold limit.
Those studies able to distinguish between the 10 different TCP isomers have
confirmed that even during these abnormal conditions, no neurotoxic ortho-
isomers of TCP could be detected (3, 4, 5).

A Canadian study was published in 1998 by the Department of Health Care

and Epidemiology of the University of British Colombia (4). Following
complaints from crew of health effects thought to be related to oil odour on
BAe 146-200 aircraft, the components of cabin air, including TCP were
measured. This study was unable to detect any TCP during in-flight
measurements, and was unable to detect any health effects associated with
the oil odour. Another study of cabin air quality on Boeing aircraft by Harvard
University in the USA, also failed to detect any TCP during in-flight
measurements (4, 5).

4
British Airways commissioned a study by an independent specialist on indoor
air quality, BRE, the former Building Research Establishment, to investigate
this issue in 2001. The BRE study showed that the concentrations of all oil
compounds detected in cabin air on the B757 were each less than 100 parts
per billion (approx. 0.00125mg/m3), which is well below the toxicological
threshold for humans of 0.1mg/m3 over 8 hours or the emergency short term
limit of 0.3mg/m3 for 15mins.

In 2004 the UK government Aviation Health Working Group commissioned a

study into cabin air quality carried out by the independent BRE. The study
analysed a wide range air quality parameters during different phases of flight
aboard BAe 146s and older Boeing aircraft, including tests for oil vapours.
The project supplemented an earlier 2001-2003 EU-funded research project,
CabinAir, which monitored air quality on 50 European airline flights. Both
surveys concluded that no air pollutant exceeded recommended health limits;
hardly any trace of oil vapour was detected.

A similar finding was reached by another EU-funded project, Health Effects in

Aircraft Cabin Environment (HEACE), which examined all aspects of the
aircraft cabin working environment during 2001-2005. However, neither of the
EU studies specifically targeted the presence of TCP.

In 2005 the British Airline Pilots Association (BALPA) organised a two day
conference on contaminated air production at Imperial College London. The
conference called upon the government to take action on the grounds of
health and safety

In 2007, the United Kingdom Committee on Toxicity (COT) was asked by the
Department for Transport (DfT) to undertake an independent scientific review
of data submitted by the British Airline Pilots Association (BALPA) relating to
concerns of its members about the possible health effects from oil fume
contamination on commercial jet aircraft. The COT estimated that cabin air
quality events occur on roughly 0.05% of flights (~1 in 2000). It concluded that
whilst a causal association between cabin air contamination by oil mists and
ill-health in commercial air crew could not be identified, a number of incidents
with a temporal relationship between reports of oil/fume exposure and acute
ill-health effects indicated that such an association was plausible. The COT
recognised that further study of air quality events should therefore be
undertaken to determine the types and concentrations of substances present
in cabin air (6).

Accordingly, the DfT Aviation Health Working Group (AHWG) commissioned

Cranfield University to carry out cabin air monitoring for a range of potential
chemical contaminants.

The initial ground investigation in a BAe146 aircraft found low levels of tri-n-
butyl phosphate (TBP) and tricresyl phosphate (TCP) in air samples, together

5
with a range of other volatile or semi-volatile organic compounds (7). These
amounts were well below occupational exposure limits.

The subsequent investigation involved in-flight monitoring of the Boeing 757

cargo aircraft and the Boeing 757, Airbus A320/1, BAe 146 and Airbus A319
passenger aircraft (8). An in-flight fume event was observed during the study
on the Boeing 757. The data from a particle monitoring device showed that
during this event there was a very high number concentration of a very small
aerosol, although overall these represented a small mass concentration of oil.
Slightly elevated levels of TBP and TCP were again measured, but all were
significantly below the relevant Health & Safety Executive specified
Workplace Exposure Limits (8, 9)

To complement the Cranfield University work, the AHWG recognised that

additional information on potential contaminant residues on internal surfaces
could be informative of possible fume events and commissioned the Institute
of Occupational Medicine (IOM) to carry out a study. The results were
published in 2012.
A total of 86 sample sets were obtained from different aircraft types, ground
vehicles and offices. The residues were analysed by gas
chromatography/mass spectrometry for TCP, TBP, butyl diphenyl phosphate
(BDPP) and dibutyl phenyl phosphate (DBPP). The surface residues in the
passenger compartments were generally lower than in the cockpit. The mean
amounts of TBP, DBPP and BDPP detected in the aircraft were similar to
those in the control vehicles. For TCP the contamination in the control
vehicles and the office locations were similar, and slightly lower than found on
the aircraft. Estimates of air concentrations consistent with these surface
residues were in agreement with other published data (10).

In a similar study in 2009, the University of British Columbia had reported the
results of surface wipe samples taken in a Boeing 757 and BAe146 showing
the presence of TCP throughout the aircraft. However, this was inconclusive
and it was recognised that the results will be influenced by confounders such
as the use of cleaning materials, wear and tear of the surface sampled, and
proximity to air vents, etc. It was noted that TCP will be found in wipe samples
taken in buildings and other public places (23).

In February 2012, an invited international group of aviation, health and

toxicology experts participated in a workshop at Hunton Park in the UK, under
the auspices of BRE, to consider the issues (12).

The Hunton Park workshop reviewed evidence associated with cabin air fume
events. It was concluded that there are no published peer reviewed reports of
acute organophosphate poisoning with analytical confirmation of the diagnosis
after cabin air fume exposures. Similarly, there are no published peer
reviewed reports of organophosphate-induced delayed neuropathy after cabin
air fume exposure, with no evidence to support a causative association
between cabin air fume exposure and short or long term nerve damage.
However, the workshop noted lack of clarity and consistency in reporting
definitions and terminology which may lead to difficulties in establishing the

6
true incidence of events. It was also observed that there is a need for
standardisation in the methodology and calibration of the sampling and
analytical procedures carried out when making the relevant cabin air quality
measurements reported so far.
The workshop agreed that there is a need for consistent guidance on the
medical assessment of crew members following a cabin air fume event. It was
noted that there is similarity between the reported symptoms of some crew
members after fume events, particularly when emergency oxygen masks have
been used, and the classical symptoms of hyperventilation.

The Australian Government Civil Aviation Safety Authority independently

convened an Expert Panel on Aircraft Air Quality in 2012 which reached
similar conclusions (13, 14).

A study by Schindler et al was published in Archives of Toxicology in 2013

(15) A total of 332 urine samples of pilots and cabin crew members in
common passenger aircraft, who reported fume/odour during their last flight,
were analysed for three isomers of tricresyl phosphate metabolites as well as
dialkyl and diaryl phosphate metabolites of four flame retardants. None of the
samples contained o-TCP metabolites above the limit of detection (LOD 0.5
lg/l). Only one sample contained metabolites of m- and p-tricresyl phosphates
with levels near the LOD. Median metabolite levels of tributyl phosphate
(TBP), tris-(2-chloroethyl) phosphate (TCEP) and triphenyl phosphate (TPP)
(DBP 0.28 lg/l; BCEP 0.33 lg/l; DPP 1.1 lg/l) were found to be significantly
higher than in unexposed persons from the general population. Median tris-(2-
chloropropyl) phosphate (TCPP) metabolite levels were significantly not
higher in air crews than in controls. The authors concluded that health
complaints reported by air crews can hardly be addressed to o-TCP exposure
in cabin air. (Note that the German abbreviation o-TCP is synonymous with
ToCP as used elsewhere in this paper.)

The conclusions contrast with a recent descriptive study published by Abou-

Donia et al (16). The study reports the results of assays performed to detect
circulating autoantibodies in a panel of 7 proteins associated with the nervous
system (NS) in sera of 12 healthy controls and a group of 34 flight crew
members, including both pilots and attendants who experienced adverse
effects after exposure to air emissions sourced to the ventilation system in
their aircrafts (sic) and subsequently sought medical attention.
The authors state these results suggest the possible development of neuronal
injury and gliosis in flight crew members anecdotally exposed to cabin air
emissions containing organophosphates. The study concludes that increased
circulating serum autoantibodies resulting from neuronal damage may be
used as biomarkers for chemical-induced CNS injury.
The study is descriptive and the numbers are small. If the list of symptoms in
figure 2 of the study were to be presented to a sufficiently large, randomly
selected section of the population, it is likely there would be a background
pattern of positive responses for one or more of the complaints listed. With
respect to measuring IgG, it is known that IgM rises first and the subsequent
IgG response is open to interpretation. Furthermore, it seems doubtful

7
whether clinical and biochemical improvements are as closely and as directly
linked as the authors would suggest.
The clinical significance of many of the tests reported in this context is not
clear and is difficult to understand in terms of clinical toxicology.

HUMAN TOXICOLOGY

Virtually every chemical, including water, can produce an adverse effect on

the human body in sufficient amount. Toxic agents can be classified by the
potency or relative dose required to elicit a specific adverse effect, which
creates a spectrum of poisons with potencies differing by many orders of
magnitude.

Absorption is the process by which a toxic substance enters the body. In the
aviation environment inhalation is the most common pathway with vapours
(gaseous component), fumes (oxides of metals) and solid particles entering
the respiratory system. The depth of penetration is determined by water
solubility, particularly for gases. For fumes and dust particles, aerodynamic
size determines the depth of penetration. Particles may be trapped in the
nasopharyngeal region, the trachea or penetrate into the lung alveoli.
There are other routes of entry. The eyes and nasal mucosae readily absorb
water-soluble particles and respond to acids and bases. The skin, however, is
waterproof and lipid proof, and highly resistant to absorption of most
chemicals. Ingestion through the gastrointestinal tract provides opportunities
for chemicals requiring an acidic environment (the stomach) or alkaline
environment (oesophagus and duodenum) to be solubilised and absorbed.

The next phase of exposure is distribution. Chemicals are dispersed

throughout the body on the basis of pH-based solubility or solubility in fat, and
eventually reach the target organ as a result of specific binding sites in the
cells.

Ultimately a toxic substance will be eliminated via a number of excretion

systems. Many chemicals are excreted via the kidneys in urine, whereas
other are excreted via the liver in bile. Other chemicals such as solvents can
be excreted as vapour through the lungs, while other chemicals are deposited
in the hair, skin and nails.

The human body has its own defence mechanisms which protect against
harm from certain levels of hazardous substances. However, if these levels
are exceeded it is possible for health to be affected, either immediately (acute
effects) or sometime after the first exposure (chronic or delayed effects).

Individuals can vary in their response to toxic insult because of age, health
status, previous exposure or genetic differences. It has long been recognised
that some individuals are more susceptible to adverse effects when exposed
to certain chemicals; the genetic basis for differences in susceptibility is being
increasingly understood. However, a susceptibility to adverse effects still
requires a clinically significant level of the chemical to be absorbed by the
body in sufficient quantities and over sufficient time periods in order to

8
produce a toxic effect. Occupational exposure levels for chemicals are set to
take account of individual differences in susceptibilities and to provide a
significant margin of safety.
In addition, it can be difficult to disentangle the physical, psychological and
emotional components of well-being, and there is no doubt that different
people may respond in different ways on different occasions.

The human senses, particularly the sense of smell, are generally very
effective in detecting potentially hazardous substances at a level well below
that which causes harm (the major exception being carbon monoxide). For
most volatile organic compounds, the concentration level for detection by a
normal healthy human is around 1,000 times less than the concentration level
which is likely to harm health.
In the UK, the Health and Safety Executive (HSE) sets the exposure limits
(OELs) for hazardous substances at work and these are published by the
HSE in Document EH40 (www.hse.gov.uk). The European legal limits are
known as Indicative Occupational Exposure Limit Values (IOELVs), and are
broadly in line with the UK HSE occupational exposure standards.

Absorption and Distribution of Chemicals

Foreign or exogenous chemicals (xenobiotics) must be absorbed from the

surrounding environment and transported to their target site in the body for a
toxic effect to occur. The chemical has to cross many cell membranes which
form a lipoprotein barrier to the outside as well as maintaining the integrity of
the cell. Most xenobiotics are transported by simple methods and not complex
carrier-associated processes (there are exceptions such as paraquat
transport into lung cells) (25).
Lipid solubility is one of the major factors determining the extent and rate of
simple diffusion through a lipoprotein membrane. Lipophilic molecules diffuse
more readily than those which are hydrophilic, the rate of transport being
dependent on the partition coefficient (ie the ratio of solubility in
octanol/water). Non-ionised molecules are often more lipophilic, and ions
generally more hydrophilic. So the movement of electrolytes, such as organic
acids and bases, is related to the degree of ionic dissociation and the lipid
solubility of the non-ionised form of the compound.
The cell membrane controls the movement of chemicals in or out of the
cytoplasm; there are several methods of transport:
Simple diffusion. Does not require energy expenditure and is the principal
method of transport for most lipid soluble, non-ionised compounds. Fick’s
law states that the rate of gas diffusion through a tissue medium is
proportional to the tissue area and the difference between the gas partial
pressures on the two sides, and inversely proportional to the tissue
thickness.
Filtration allows water, ionic and hydrophilic molecules of appropriate size
to pass through small pores (~0.4nm diameter) in the cell membrane.
Facilitated diffusion is carrier-mediated, and transports chemicals with
specific common structures across the cell membrane.
Active transport allows the absorption of substances against a
concentration gradient.

9
 Phagocytosis and pinocytosis enables particulates and solutions to be
taken into the cell by the extrusion or invagination of an area of the
membrane.

Inhalation Kinetics

The constant diffusion of gases between the alveoli and the pulmonary
vessels leads to the composition of alveolar air differing from ambient
atmospheric air. At a body temperature of 37 degC water vapour exerts a
pressure of 47mmHg, remaining constant at all altitudes due to metabolism.
When gas partial pressures are calculated, water vapour pressure must be
subtracted from the total pressure.

The partial pressure of carbon dioxide in the alveolar air is about 40 mm Hg,
although this reduces with increasing altitude due to the effect of physiological
hyperventilation, and this similarly has to be taken into account when
calculating alveolar partial pressures.

It is important to note that it is the partial pressure (related to concentration) of

an individual gas which drives the exchange.

The lung tissue barrier (alveolar membrane) separating air and blood is only
0.5 - 1.0 thick and the 300 - 400 million alveoli provide a large surface area
for diffusion. In accordance with Fick’s law, the transfer of gases through the
alveolar membrane depends on the area and thickness of the membrane, and
the partial pressures of the gases in the blood and in the alveoli.
The media on either side of the alveolar membrane are being continuously
renewed; the air is changed 12 - 15 times per minute and the pulmonary
blood flows at 3.5 - 5 litres per minute at rest, at sea level. This leads to
efficient elimination of volatile chemicals.

Factors influencing the inhalation kinetics of a volatile compound include the

environmental air concentration, duration of exposure, rate of alveolar
ventilation, cardiac output, blood and tissue solubility and the degree of
metabolism of the chemical. Volatile compounds are usually inhaled as a gas
mixture with air and most are completely miscible in all proportions. The
concentration of gases and volatile compounds in a mixture is expressed in
terms of partial pressure, which is not equivalent to concentration. However,
the relative concentrations of dissolved materials can be expressed in terms
of partial pressures which add up to a total pressure of 100%. Solubility is
inversely related to the temperature and proportional to the pressure of the
chemical in the ambient gas. The partial pressures of constituent volatile
compounds vary with the absolute pressure but, at a fixed pressure, the
concentration of each gas or vapour varies directly with its partial pressure
and indirectly with the total pressures of the gas/vapour mixture.

Thus any inhaled gas will be part of the total gas mix in the alveolus, and its
absorption depends on the partial pressure exerted by that gas. Taking the
RB211 engine as an example, the maximum engine oil possible in the bleed
air is 0.4kg (20). Of this, 3% is TCP of which around 0.1% is ToCP. In the

10
unlikely worst case scenario of the total discharge of an engine’s lubricant into
the engine bleed system, 0.4kg of oil would pass into the cabin ventilation
system. This would give a peak cabin atmosphere ToCP level of 0.025
mg/m3, reducing rapidly due to normal cabin ventilation. This peak level would
thus be a quarter of the 8hr workplace limit of 0.1 mg/m 3, and less than a
tenth of the 15min emergency workplace limit of 0.3 mg/m 3.

Alveolar absorption depends on Dalton’s Law of partial pressures, as well as

Fick’s Law, and the partial pressure of bleed air contaminants would therefore
be a very small proportion of the total alveolar gas pressure, reducing rapidly.
Of the published levels of ToCP detected in cabin air, most are less than
0.005 mg/m3. Another way of expressing gas concentration is as parts per
billion (ppb), and for TCP 1 ppb is approximately 0.007 mg/m 3. [To assist
visualisation, in terms of time 1 ppb would be analogous to expressing 1
second in 32 years.]

It would be highly unlikely, if not impossible, for such small concentrations of

contaminant to cross the alveolar membrane so as to cause
organosphosphate poisoning through inhalation. It is important to note in this
regard that there are no published peer reviewed reports of acute
organophosphate poisoning with analytical confirmation of the diagnosis after
cabin air fume exposures.

EFFECTS OF ALTITUDE

Ascent to altitude is associated with a fall in air pressure paralleled by

decreases in density and temperature. Thus at 18,000 feet in the Standard
Atmosphere atmospheric pressure is half its value at sea level and the
ambient temperature is about -20 degC.

The behaviour of gases is explained by the Gas Laws.

Boyle’s Law can be expressed as V1/V2 = P1/P2 where V1 is the initial volume,
V2 is the final volume, P1 is the initial pressure, and P2 is the final pressure. At
a constant temperature, the density of a given mass of gas varies directly as
its pressure, so the expressions for pressure can be substituted by
expressions for density (D).
Charles’s Law shows that at a constant pressure the volume of a gas is
proportional to its absolute temperature.
Combining the two laws gives the General Gas Law which is P1V1/T1 =
P2V2/T2

When considering gas mixtures, Dalton’ Law states that the total pressure of a
gas mixture is the sum of the individual or partial pressures of all the gases in
the mixture. The partial pressure of each gas in the mixture is derived from
P1=F1xP, where P1 is the partial pressure of gas 1, F1 is the fractional
concentration of gas 1 in the mixture, and P is the total pressure of the gas
mixture.

11
Respiratory Physiology

The relationship between the oxygen saturation of haemoglobin and oxygen

tension is reflected in the shape of the oxyhaemoglobin dissociation curve
reproduced below:

Inspection shows a plateau indicating that the oxygen saturation does not fall
below 90% until the altitude exceeds about 10,000 feet (which equates to an
alveolar oxygen tension of approximately 55 mm Hg). As altitude rises above
10,000 feet the percentage saturation of haemoglobin falls precipitously
resulting in hypoxic, or hypobaric, hypoxia.

In moderate hypoxia, such as when breathing air at 25,000 feet, cardiac

output and heart rate are increased but overall peripheral resistance is
reduced, so that mean arterial blood pressure is unchanged. Cerebral blood
flow is increased, although the degree of increase is modified by the
magnitude of coexisting hypocapnia which results from increased respiratory
minute volume. Thus cerebral and cardiac perfusions are increased at the
expense of less vital organs.
Respiration increases under the hypoxic drive to help alleviate cerebral
hypoxia but is ineffective, and the symptoms and signs of hyperventilation
develop alongside those of hypoxia. Hyperventilation is a normal response to
a fall in alveolar oxygen partial pressure to below 55 – 60 mm Hg and may be
the dominant clinical feature. Symptoms both of hypoxia and hyperventilation

12
can include light-headedness, feelings of unreality and anxiety,
paraesthesiae, visual disturbances and palpitations (24).

Effect of Altitude on Absorption

Exposure to breathing air at cabin altitudes between 10,000 and 30,000 feet
never results solely in a reduction in the oxygen tension in the alveolar air
since there is always a simultaneous decrease in the alveolar carbon dioxide
concentration. The effect on alveolar ventilation rate is thus due to the
combined effects of a lowered alveolar oxygen concentration and hypocapnia.
There is a very considerable individual variation in the degree of increase in
alveolar ventilation caused by a given reduction in alveolar oxygen tension;
amongst individuals exposed to the same level of oxygen deprivation some
will achieve a lower alveolar carbon dioxide tension than others.
The tensions of oxygen and carbon dioxide in the alveolar air of resting
subjects breathing air at reduced barometric pressure reflect the changes in
the inspired oxygen tension and the changes in alveolar ventilation caused by
hypoxia.
It has been shown that there is no increase in pulmonary ventilation
(breathing rate) until the alveolar oxygen tension is reduced to about 65 mm
Hg (17). This occurs at an altitude of about 8,000 feet. It is only when the
barometric pressure is reduced further that pulmonary ventilation is increased.

However, none of these effects are significant at cabin altitudes below 10,000
feet. It can be seen from the oxyhaemoglobin dissociation curve that the
reduction in alveolar oxygen partial pressure is small, resulting in a
desaturation of less than 10%. The published air quality standards remain
valid at cabin altitudes up to 10,000 feet (24).

AEROTOXIC SYNDROME

A syndrome is defined as a set of symptoms which occur together, or the sum

of signs of any morbid state, or a symptom complex. It follows that there
should be a consistent set of common symptoms which together make up a
given condition.
Individuals reporting that they suffer from the so-called aerotoxic syndrome
describe a wide range of individual symptoms and signs, with insufficient
consistency to fulfil the requirements for the definition of a medical syndrome.
Many of the reported acute symptoms are largely the same as those reported
by participants in all phase 1 drug trials, being normal symptoms experienced
by most people on frequent occasions. It is recognised that 70% of the
population experience one or more of them on any given day.
The Aerospace Medical Association reviewed the scientific evidence and
concluded that there was insufficient consistency and objectivity to support
the establishment of a clearly defined syndrome (18). The US National
Academy of Sciences performed a similar review and reached the same
conclusion (19), as did the Australian Government CASA Expert Panel on
Aircraft Air Quality in 2012 (13, 14).

13
Thus the concept of the ‘Aerotoxic Syndrome’ is not recognised in the aviation
medicine community.

Of the UK professional pilot population of approximately 20,000, in 2013 the

CAA log of pilots who have reported symptoms which they attributed to
exposure to fumes has 31 cases, of which 3 are duplicates, leaving a total of
28 individuals. Of these 14 have returned to flying or were never assessed as
unfit. The remaining 14 remain unfit (the CAA no longer differentiates
temporary/long term unfit), although a number would have passed normal
retirement age. No new cases have been documented since June 2012 and
the CAA is not aware of any other on-going cases (26).

Following a review of the log document, the CAA Medical Department

determined that this information has no medical validity because:
the inclusion of an individual is based entirely on the individual’s
subjective perception of the cause of their condition;
there are no objective or diagnostic medical criteria for inclusion;
there is no consistency in the symptoms and signs reported by the
individuals or their treating doctors;
it is likely that there is incomplete reporting of cases where a pilot
experiences brief symptoms at the time of an event.

There are known to be individual genetic differences in sensitivity to smells

and chemical exposures, but there is no consistency in reported
symptomology. The concentration of oil products in the cabin air is very low
and greatly diluted in the free stream air.
Taking the RB211 engine as an example, the maximum engine oil possible in
the bleed air is 4kg. Of this, 3% is TCP of which around 0.1% is ToCP. In the
worst case scenario of the total discharge of an engine’s lubricant into the
engine bleed system, 0.4kg of oil would pass into the cabin ventilation system.
This would give a peak cabin atmosphere ToCP level of 0.025 mg/m3,
reducing rapidly as a result of normal cabin ventilation. The peak level would
be a quarter of the 8hr workplace limit of 0.1 mg/m3, and less than a tenth of
the 15min emergency workplace limit of 0.3 mg/m3 (20).
Alveolar absorption depends on Dalton’s Law of partial pressures, as well as
Fick’s Law (see above), and the partial pressure of bleed air contaminants is a
small proportion of the total alveolar gas pressure. Comparisons have been
made with the symptoms reported by farmers exposed to organophosphates
in sheep dip, but the concentrations and periods of exposure are many
magnitudes of difference as well as the route of chemical entry into the body
being different.

Symptoms reported by some crew members who have been exposed to

fumes in the cabin are similar to those reported by individuals complaining of
a range of conditions such as sick building syndrome, chronic fatigue
syndrome, Gulf War syndrome, Lyme disease and chronic stress. In all these
conditions, there is lack of consistency in reported symptoms and signs and
wide individual variability.

14
The UK Government Committee on Toxicity published a position paper in
2013 (27). In summary, the paper concluded:

a toxic mechanism for the reported long term illness is unlikely

further cabin air sampling would be technically feasible but prohibitively
expensive
collection and analysis of more comprehensive data in relation to fumes
events, evaluated against a suitable control sample, would be informative and
might help to target either further research or measures to reduce the number
of incidents
further studies involving collection and analysis of biological samples would
be feasible and useful.

Specifically, amongst the conclusions the paper stated:

“12 vii. More generally, the Committee considers that a toxic mechanism for
the illness that has been reported in temporal relation to fume incidents is
unlikely. Many different chemicals have been identified in the bleed air from
aircraft engines, but to cause serious acute toxicity, they would have to occur
at very much higher concentrations than have been found to date (although
lower concentrations of some might cause an odour or minor irritation of the
eyes or airways). Furthermore, the symptoms that have been reported
following fume incidents have been wide-ranging (including headache, hot
flushes, nausea, vomiting, chest pain, respiratory problems, dizziness and
light-headedness), whereas toxic effects of chemicals tend to be more
specific. However, uncertainties remain, and a toxic mechanism for symptoms
cannot confidently be ruled out.

13. Finally, it should be emphasised that illness can be disabling whether it

occurs through toxicity or through nocebo* effects, and therefore there is a
continuing imperative to minimise the risk of fume incidents that give rise to
symptoms.”

[* In medicine, a nocebo (Latin for "I shall harm") is a harmless substance that
creates harmful effects in a patient who takes it. The nocebo effect is the negative
reaction experienced by a patient who receives a nocebo. Conversely, a placebo is
an inert substance that creates either a positive response or no response in a
subject who takes it. The phenomenon in which a placebo creates a positive
response in the subject to which it is administered is called the placebo effect.
Both nocebo and placebo effects are entirely psychogenic. Rather than being
caused by a biologically active compound in the nocebo or placebo itself, these
reactions result from a subject's expectations about how the substance will affect
him or her. Though they originate exclusively from psychological sources, nocebo
effects can be either psychological or physiological.]

15
IRRITABILITY

Irritation is a state of over-excitation and undue sensitiveness of the nervous

system in response to a stimulus. For example, irritant receptors in the lungs
stimulate reflex constriction of the bronchioles in response to smoke and
smog. Similarly, sneezing, sniffing and coughing may be stimulated by irritant
receptors in the nose, larynx and trachea.
Some gases are known to be irritants and as a rule they are chemically
corrosive. They injure surface tissues and induce inflammation of the air
passages and the parenchymal region. Organophosphates are not classified
as irritant gases.
Individuals vary in their response to sensory stimuli, including smells. Genetic
differences are thought to cause some people to have enhanced sensitivity to
low levels of some volatile chemicals; they experience a range of irritant
symptoms affecting well-being.

HYPERVENTILATION

Hyperventilation is a normal response to emotional stress, particularly anxiety,

apprehension and fear. Even low levels of stress, which may not be perceived
as such by the individual, commonly give rise to hyperventilation. It is also a
natural response when resistance to breathing is encountered, such as when
using an emergency oxygen mask.
Obviously not every case of ‘aerotoxic syndrome’ is caused by
hyperventilation, but it offers a plausible explanation for some reported
events.

In the aviation environment it is generally recognised that hyperventilation is a

common condition. Studies have shown that a large proportion of aircrew
under training hyperventilate, as do experienced aircrew when confronted with
an unusual event or in-flight emergency. A 2009 study raised concerns about
the prevalence of unrecognised hyperventilation amongst airline pilots and the
potential risk to flight safety (21).
Symptoms can include light-headedness, headache, feelings of unreality and
anxiety, paraesthesiae, visual disturbances, palpitations, cognitive
impairment, loss of concentration and, in extreme cases, muscular tetany and
paralysis (22, 24).
The nervous system and blood vessels are very sensitive to the acidity of the
blood, which is reduced during hyperventilation. It causes a marked
constriction of blood vessels in the brain and skin and an increase in blood
flow through the muscles. The reduction in blood flow through the brain leads
to a reduction in the oxygen available to the cerebral tissues, giving effects
similar to hypoxia affecting mental performance. The shallow breathing
reduces gaseous exchange in the alveoli which may result in a fall in
peripheral blood oxygen concentration, adding to the hypoxic effects.

Whereas in general medicine, the hyperventilation syndrome may not always

be readily recognised as a clinical entity, falling as it does between

16
physiology, psychiatry, psychology and medicine, the condition of
hyperventilation is readily accepted in aviation medicine.
However, diagnosis can be difficult in the absence of a simple measurement.
The physiological diagnosis of hyperventilation is breathing in excess of
metabolic requirements, thus implying arterial hypocapnia (low CO2 tension)
and an abnormally high respiratory drive. However, in chronic cases
measurement of the alveolar partial pressure of carbon dioxide (PCO2) is
difficult and can be profoundly affected by the total physiological inputs to
respiration and the conscious state of the individual. There can be a tendency
to hyperventilate even though the resting PCO2 is normal.

There are a number of factors which may perpetuate hyperventilation. Apart

from renal compensation, there appear to be physiological mechanisms
resetting the PCO2 to a lower level independent of chemoreceptor setting.
Habit may be a perpetuating mechanism, as may be misattribution of
symptoms of hypocapnia (symptoms not dissimilar to those of carbon
monoxide toxicity).
The interaction of factors contributing to chronic hyperventilation remains
uncertain. One possible scenario is that an acute episode of hyperventilation,
such as might occur on perceived exposure to oil fumes, leads to symptoms
which are misdiagnosed or incorrectly diagnosed. The symptoms can be
alarming and as a consequence, the individual’s anxieties are increased and
further consultation sought, leading to perpetuation of the disorder. However,
providing satisfactory proof of such a model is difficult, and in the presence of
worrying physical symptoms without an obvious cause the individual and
his/her medical advisers may be reluctant to consider such a diagnosis.
With such a wide range of reported symptoms and signs amongst crew
members who are anxious that they have been exposed to oil fumes, it is
fatuous to suggest that all cases are the result of hyperventilation. However, it
should be considered in the differential diagnosis.

Hyperventilation Syndrome:

17
CONCLUSION

There has been an increase in reported incidents of in-flight smoke/fume

events since 1999, with a small number of crew members reporting adverse
health effects which they associate with the events.
The source of oil contamination of engine bleed air was identified in early
versions of the BAe 146 and the Boeing 757 and suitable modifications were
implemented. A range of chronic health effects continue to be reported by
some crew members.

The toxic effects of organophosphates are specific and are due to impairment
of neurotransmission in the peripheral nerves, giving rise to muscular
weakness and paralysis. In terms of medical toxicology, it is impossible to
explain the wide range of symptoms and signs reported by some crew
members as a unified result of TCP exposure.
Symptoms reported by some crew members who have been exposed to
fumes in the cabin, particularly when emergency oxygen masks are used, are
the same as those seen in acute or chronic hyperventilation. Obviously not
every case of ‘aerotoxic syndrome’ is caused by hyperventilation, but it offers
a plausible explanation for some reported events.
In some cases, the symptoms may be due to irritation associated with
enhanced chemical sensitivity to certain volatile organic compounds.

The reported symptoms are wide-ranging with insufficient consistency to

justify the establishment of a medical syndrome. It has been noted that many
of the acute symptoms are normal symptoms experienced by most people
frequently; some 70% of the population experience one or more of them on
any given day.

Individuals can vary in their response to potential toxic insult because of age,
health status, previous exposure or genetic differences.
In addition, it can be difficult to disentangle the physical, psychological and
emotional components of well-being, and there is no doubt that different
people will respond in different ways on different occasions.
It is not understood why most occupants of pressurised aircraft do not report
symptoms despite having the same exposure as those who do.

Finally, so far as scientific evidence has been able to establish to date, the
amounts of organophosphates to which aircraft crew members could be
exposed, even over multiple, long-term exposures, are insufficient to produce
neurotoxicity.

Investigations of aircraft cabin air world-wide have failed to detect levels of

TCP above well-established and validated occupational exposure limit values.
The partial pressure in the alveolar gas mixture of any TCP contamination of
the cabin air is so low that it is unlikely to cross the alveolar membrane.

Genetic or particular susceptibility to a particular adverse effect of certain

chemicals on the part of an individual does not alter the need for there to have

18
been a sufficient chemical exposure to cause the injury or damage. For the
reasons set out above, the possible exposure levels to ToCP on aircraft are
so low relative to what is required to create a toxic effect through inhalation
that a toxic injury is simply not medically feasible with current understanding.

Aviation medical professionals throughout the world continue to monitor the

scientific evidence and remain receptive to objective peer-reviewed evidence.

REFERENCES

1. GCAQE Press Release 19 April 2012. GCAQE: http://www.gcaqe.org

2. Inhalable toxic chemicals on board aircraft – ITCOBA Seminar, Cranfield,
Oct 2011. Journal of Biological Physics and Chemistry, Vol 11, No 4.
http://www.itcoba.net/
3. van Netten C (1998). Air quality and health effects associated with the
operation of the BAe146-200 aircraft. Applied Occupational and
Environmental Hygiene 13:733-739
4. Spengler JD, Burge H, Dumyahn T, Muilenberg M, Forester D (1997).
Environmental Survey on Aircraft and Ground-based Commercial
Transportation Vehicles. Harvard School of Public Health, Harvard
University, Cambridge, MA.
5. Dumyahn, TS, Spengler, JD, Burge, HA, Muilenburg, M (2000).
Comparison of the Environments of Transportation Vehicles: Results of
Two Surveys, Air Quality and Comfort in Airliner Cabins, ASTM STP
1393, N.L. Nagda, Ed, American Society for Testing and Materials, West
Conshohocken, PA 2000
6. Committee on Toxicity of Chemicals in Food Consumer Products and the
Environment: Statement on the Review of the Cabin Air Environment, Ill-
health in Aircraft Crews and the possible relationship to Smoke/fume
Events in Aircraft.
http://cot.food.gov.uk/cotstatements/cotstatementsyrs/cotstatements2007/
cotstatementbalpa0706
7. Muir H. (2008). Cabin air sampling study functionality test. Cranfield:
Cranfield University. https://dspace.lib.cranfield.ac.uk/handle/1826/2389
8. Cranfield University Institute of Environment and Health. (2011). Aircraft
Cabin Air Sampling Report (Part 1 of the Final Report)
https://dspace.lib.cranfield.ac.uk/handle/1826/5305
9. Cranfield University Institute of Environment and Health. (2011). Aircraft
Cabin Air Sampling Report (Part 2 of the Final Report)
https://dspace.lib.cranfield.ac.uk/handle/1826/5306
10. Lamb J, McGonagle C, Cowie H, Cherrie JW. Cabin Air – surface residue
study. IOM Research Report TM/11/06. March 2012
11. Spengler JD, Wilson DG. Air Quality in Aircraft. Proceedings of the
Institution of Mechanical Engineers, Part E: Journal of Process
Mechanical Engineering November 1, 2003 vol. 217 no. 4 323-335
12. BRE Cabin Air Quality Workshop. Aviat Space, Environ Med; 83(10):
1023. October 2012

19
13. Navathe P. What does EPAAQ add to our Knowledge? Proceedings of
IATA Aviation Health Conference, London. Oct 2012
14. www.casa.gov.au/cabin/epaaq/index.htm
15. Schindler BK, Weiss T, et al. Occupational exposure of air crews to
tricresyl phosphate isomers and organophosphate flame retardants after
fume events. Arch Toxicol (2013) 87:645–648
16. Abou-Donia MB, Abou-Donia MM, et al. Autoantibodies to nervous
system-specific proteins are elevated in sera of flight crew members:
Biomarkers for nervous sytem injury. Journal of Toxicology and
Environmental Health, Part A, 76:363–380, 2013
17. Gibson TM. Hyperventilation in aircrew: a review. Aviat Space Environ
Med 1979; 50: 725-33
18. Hudson MF. Cabin Air Quality in Commercial Aircraft: Is there any cause
for concern? A Review of the Current Evidence. Aviat Space Environ
Med; 82(3). March 2011:280-281
19. www.nap.edu/catalog.php?record_id=10238 pp180-181
20. The United Kingdom Parliament - Select Committee on Science and
Technology – Fifth Report (04-10-2006): 4.39
21. Karavidas KK, Lehrer PM. In-flight hyperventilation among airline pilots.
Aviat, Space, Environ Med 2009; 80 (5): 495-6
22. Brown EB. Physiological effects of hyperventilation. Physiological
Reviews 1953; 33: 445-71
23. van Netten C. Final Report on Aircraft Wipe Sample Analysis for
Tricresyl Phosphate Isomers (Feb 2009). University of British Columbia,
Vancouver BC. www.spph.ubc.ca
24. Ernsting J, Ward J, Rutherford OM. Cardiovascular and respiratory
physiology. In Rainford DJ, Gradwell DP (eds). Ernsting’s Aviation
Medicine 4ed (2006) ISBN-10 0 340 81319 9: 13-40
25. Blain PG. Absorption of chemicals and mechanisms of detoxification. In
Baxter PJ, Adams PH, Aw T-C, et al (eds). Hunter’s Diseases of
Occupations 9ed (2000) ISBN 0 340 67750 3: 67-80
26. Head of Aviation Health Unit, UK Civil Aviation Authority Medical
Department (September 2013). Personal communication.
27. http://cot.food.gov.uk/cotstatements/cotstatementsyrs/cotstatements2013/co
tpospacabair

20