On Oct 13th 2016 Spain’s CIAIAC released their final report concluding the probable causes of the accident were:
The investigation into this event revealed circumstantial evidence suggesting that several crewmembers were affected by contaminated cabin air that was being supplied by the aircraft’s air conditioning system.
After several tests and analyses have been carried out till this day, the investigation yielded no conclusive findings as to the possible source of the contamination or the hypothetical toxic compound involved.
The CIAIAC reported that there had been numeours reports of presumed contaminated cabin air events around the globe, for example, Boeing confirmed a higher than expected rate of fume events on 757-200s operated by RB211-535C engines, almost all of which were operated by British Airways, which were found confirmed as result of engine oil leaks. As result of these investigations there was a change of engine overhaul and engine oil servicing procedures.
The CIAIAC continued describing the investigations conducted by Germany’s BFU:
As a result of the increased number of reports involving “odors” in the cabin, the German accident investigation bureau conducted a study based on 845 accidents, serious incidents and incidents reported from 2006 to 2013.
In 663 of the cases a relationship with the cabin air was established. In the remaining 180 cases health impairments were described, which were not connected with cabin air quality (e.g. broken toe, burned hand).
In 10 of the events reported to the BFU, the affected individual stated suffering from longterm symptoms. All of these cases involved fumes in which an oily or “dirty sock” smell was reported.
On the other hand UK’s Transport Ministry found in their studies:
In light of the growing concern in this area, the Transport Department in the United Kingdom decided to commission Cranfield University’s Institute for Environment, Health, Risk and Futures to conduct a study.
As part of this study, cabin air quality measurements were taken over the course of 100 flights. No fume events occurred in any of these flights, nor were the conditions needed to activate the incident reporting criteria met.
However, the UK’s Committee on Toxicity of Chemicals in Food, Consumer Products and the Environment (COT) reached following conclusions:
The Committee on Toxicity of Chemicals in Food, Consumer Products and the Environment (COT) is an independent scientific committee that advises the Food Standards Agency, the Department of Health and other government departments and agencies in the United Kingdom on matters related to the toxicity of chemical products.
In 2007, this committed was commissioned to advise the Department for Transport on cabin air quality, a task it continues to carry out to this day.
At this committee’s last meeting, held in late 2013, four research projects on this subject were reviewed: Cranfield University 2008, 2009; Institute of Environment and Health 2011 a/b; Institute of Occupational Medicine 2012. The committee also considered papers
Based both on the information previously considered and on the new projects, the COT drew a total of thirteen conclusions:
– Contamination of cabin air by components and/or combustion products of engine oils, including triaryl phosphates, does occur, and peaks of higher exposure have been recorded during episodes that lasted for seconds
– Episodes of acute illness, sometimes severely incapacitating, have occurred in temporal relation to perceived episodes of such contamination.
– There are a number of air crew with long-term disabling illnesses, which they attribute to contamination of cabin air by engine oils or their combustion products.
– The acute illness which has occurred in relation to perceived episodes of contamination might reflect a toxic effect of one or more chemicals, but it could also have occurred through nocebo effects.
– While there is strong scientific evidence that nocebo effects can lead to (sometimes severely disabling) illness from environmental exposures that are perceived as hazardous, there is no simple and reliable way of establishing that nocebo responses are responsible for individual cases of illness.
– The patterns of illness that have been reported following fume events do not conform with that which would be expected from exposure to triaryl phosphates such as o-TCP.
– The COT considers that a toxic mechanism for the illness that has been reported in temporal relation to fume incidents is unlikely.
– Decisions to undertake further research will need to balance the likelihood that it will usefully inform further management of the problem against the costs of undertaking the work.
– One possibility would be to collect better information. This would require airlines to record and retain a limited set of information on all flights that they operate
– As an extension to the above study, a case-control approach could also be used to investigate associations of fume incidents with operational parameters.
– Another possible extension to a systematic study of fume incidents would be to collect and store samples of urine, and possibly blood, from crew members within 48 hours (the earlier the better) after such events. These could then be analyzed for biomarkers of potential toxic pollutants, as in the studies by Schindler at al. (2013) and Liyasova et al. (2011).
– Since 2007, there have been significant advances in the technology that is available for air-monitoring, and in theory it should now be possible to develop a compact system in which a particle counter would run continuously, and trigger other sampling instruments if and when a fume incident occurred. The samples collected could then be used to identify any chemicals that occurred at exceptionally high concentrations.
The CIAIAC analysed (quoted in its entirety):
During the aircraft’s approach to the Gran Canaria Airport there was a fume event in the cabin that was noticed by all the crewmembers, causing them physical discomfort.
This odor, however, did not seem to have much of an effect on the passengers, since none of them made any complaints while deboarding the aircraft or within the following days.
However, there are reasonable explanations for this: Physiological explanation: Flight attendants are in general physically more active than passengers, i.e. they tend to have a higher respiratory rate. Every breath may then become an inhaled dose of an alleged toxic substance in the air.
Psychological explanation: Passengers tend to attribute an individual malaise either to nausea due to aircraft movements or to an individual event one had encountered before, e.g. late night party. At the time of this accident most passengers would have assumed no connection between a smell in the cabin and an individual malaise.
The medical service at the airport did not treat any passengers, and there is no conclusive evidence that any passengers were affected after the flight.
After the passengers deplaned, the flight attendants were asked about their physical condition, with 5 of the 6 answering that they were still physically affected to some extent by the event.
During the tests that were conducted before starting the return flight to Germany, the odor was noticed again, leading to more severe physical symptoms in some of the flight attendants.
The simultaneous nature of the fume event and the appearance and worsening of the physical symptoms in the crew requires consideration. It could indicate the presence of a cause-effect relationship between them. It could also indicate a nocebo effect.
2.2. Analysis of the aircraft
The cabin odor event could not be reproduced during the operational tests of the air conditioning system carried out at the Gran Canaria airport. None of the people onboard the aircraft noticed any unusual odors, nor did they feel any physical symptoms.
The inspections of the aircraft carried out at the Gran Canaria airport both before and after the operational tests did not provide any information as to the possible source of the odor, nor was any evidence found of an oil or hydraulic fluid leak in the engines or APU.
The only finding of note was 5 liters of deicing liquid that was found in the APU tray.
This liquid could have come from the deicing treatment that was performed at the Hamburg airport at the start of the flight in which the event took place.
During the subsequent positioning flight to the Frankfurt airport, the fume event reappeared on three occasions. The first was during the taxi phase, when the occupants noticed a strange odor but without feeling any physical symptoms. The “Greywolf” recorded an increase in Total volatile organic compounds (TVOC) at the time of the odor.
The other two episodes took place during the flight. The occupants’ statements indicate that the odor was much stronger than the first one; furthermore, several occupants felt physical symptoms, such as a sore throat and tongue and numb finger tips. These symptoms disappeared shortly after the odor faded. Neither sensor, however, registered any changes.
In light of the differences between the first and the two subsequent episodes, it stands to reason that they may have had different triggers.
The first did not have any physical effects and was detected by the occupants only through its smell, and by one of the sensors, which recorded an increase in TVOC. This episode could have occurred when combustion fumes from another aircraft or some other type of outside air contamination entered the cabin.
In the other two episodes, however, the odor seems to have been much stronger, though the sensors did not detect any changes. As for the physical effects, they were temporary, disappearing completely once the smell faded. Some of the symptoms perceived, like the throat and tongue irritation, could be compatible with glycol contamination of the air. It is possible that deicing liquid could have remained in another part of the aircraft in addition to the APU tray, and that it moved during the flight until it entered the air conditioning system.
The glycol, however, could not have produced the numbing sensations described. Also, if deicing liquid had found its way into the cabin, it should have been detected by the “aerotracer” sensor, which it was not. This would rule out the deicing liquid as the source of either the odor or the physical symptoms.
The fact that the symptoms disappeared as soon as the odor faded could point to a psychosomatic cause.
In this regard, it is worth remembering that the aircraft’s occupants were aware of the event that had occurred in the previous flight, as well as of the physical symptoms described by the affected crewmembers. The appearance of a strong odor and irritation could have resulted in a stressful situation that induced the affected individuals to “feel” the remaining symptoms, that is, the numbing sensation.
A further contributing factor could be the heightened awareness of fume events in Germany.
These symptoms could have been real, however, though if they had, the mechanism that produced them must have been different from the one involved in the accident flight, given the large difference between the symptoms occurring in the two cases.
The operator ran a test after a deicing treatment that showed that the liquid used in the treatment could find its way into the cabin through the air conditioning system. Though this test proves that this type of contamination is possible, and consequently it can adversely affect the cabin air quality, it does not conclusively identify deicing liquid as the source of the fume event since this compound, in its original state as well as heated up to 200ºC, does not produce the symptoms that affected the crewmembers.
The remaining inspections and tests conducted at the operator’s facilities, aided by the manufacturer, did not produce any clues as to the possible cause of the odors.
Following these tests many of the air conditioning components were replaced as a precaution and the cabin was cleaned.
The aircraft was later returned to service.
This situation could lead to the conclusion that the source of the odor was in one of the air conditioning components that was replaced, or in a compound that was deposited on panels in the cabin combined with some other circumstance, like the smell of the adhesive used when the rug was replaced. It could also have been a one-time event.
In conclusion, the tests conducted on the aircraft did not offer clear evidence as to what might have been the source of the odor.
In addition, the tests carried out with the sensors, as well as the analysis of the cabin air samples, the APU tests and so on, did not detect the presence of any compound in a high enough concentration, based on currently accepted thresholds, to produce the physical symptoms described.
2.3. Medical information on the crew
One of the most notable circumstances is the large difference between the symptoms that affected the crewmembers, both in terms of the acute and long-term symptoms.
This could be due to multiple causes, such as, for example, different sensitivities to specific compounds, previous exposure, physical condition, psychological differences, etc. It could also be that the supposed contaminant was not mixed thoroughly with the air in the system.
As a result, there could have been large differences in the concentration of the contaminant at the system’s outlets, which could have caused it to affect some crewmembers more than others.
The symptoms reported by the crewmember in position 2R, especially after the second episode, are compatible with an exposure to organophosphates.
The second single-fiber EMG done on the crewmember in the 2R position detected a slight dysfunction in neuromuscular transmission.
The single-fiber EMG is a test that is carried out using a special needle that is capable of exploring an isolated muscle fiber (a normal EMG analyzes motor units consisting of several muscle fibers). The two main readings detect changes in the motor unit structure and neuromuscular transmission. In order for the muscle to contract, the nerve has to be stimulated. This releases a quantity of acetylcholine at the nerve-muscle interface, the neuromuscular junction. In normal conditions, there is enough acetylcholine to contract the muscle several times. What the single fiber, or jitter, study does is stimulate the nerve near the muscle repeatedly and measure how long it takes for the muscle to contract. In a normal muscle the contraction always occurs a certain period of time after the nerve is stimulated because there is enough acetylcholine to produce each contraction. If there is not enough acetylcholine, after a few stimulations the muscle contraction can be delayed or even not take place. This phenomenon, called jitter, leads to increased variability in muscle contraction and is evident in practically every case involving problems with the neuromuscular junction.
The battery of tests and medical reports available at this time for this crewmember seem to verify the existence of physical and cognitive symptoms compatible with poisoning by some kind of neurotoxin.
2.4. Poisoning. Cause-effect relationship
The diagnosis from the hospital in Berlin where crewmember 2R was treated indicates tricresyl phosphate (TCP) as the source of the poisoning, based solely on the fact that the poisoning symptoms of this compound are similar to those exhibited by the crewmember.
The same diagnosis, however, acknowledges that, it is difficult, if not impossible, to prove that TCP caused the poisoning.
The report by Dr. Abou-Donia corroborates the diagnosis of neuronal damage due to the presence of antibodies for nervous system proteins. However, complete details of Dr. Abou-Donia’s testing are unavailable. And as Dr. Abou-Donia himself states, this test is entirely non-specific, meaning that it would only serve to verify the effect (neuronal damage due to poisoning), but not to identify the cause (toxin).
The blood tests done on the first officer and crewmembers 2R and 2L on the day of the accident at the Clínica del Perpetuo Socorro included a determination of serum cholinesterase, with the results for all three being normal.
As noted in Section 220.127.116.11, organophosphoric compounds such as TCP inhibit the enzyme acetylcholinesterase in the nerve synapses and in erythrocytes. They also inhibit butyrylcholinesterase, also called plasma cholinesterase.
The fact that the plasma cholinesterase values determined during the analyses done on the day of the accident were normal would tend to rule out the involvement of organophosphoric compounds in the poisoning.
It should be noted, however, that plasma cholinesterase values are not definitive indicators of organophosphate involvement, which is why the analysis results should be taken with some reservation, as they cannot be used to fully rule out the presence of this type of compound.
In contrast, the erythrocyte cholinesterase values are regarded as representative. In this case the cholinesterase values are obtained from cells (erythrocytes) instead of plasma.
Determining these values is much more complex than for plasma cholinesterase, and the procedure is only done at certain specialized laboratories. As a result this value is not usually available.
The tests done in Gran Canaria did not determine the erythrocyte cholinesterase value.
There is no record of this test having been performed subsequently in Germany either.
None of the available data could be used to identify the toxic agent; it was thus impossible to verify or rule out tricresyl phosphate as the cause of the poisoning.
2.5. Events in which a cause-effect relationship was suspected
Most, if not all, of the cases in which a direct cause-effect relationship between the aircraft and the health effects on individuals was suspected involved events in which a onetime contamination or cabin air occurred due to a leak of one of the fluids used in the aircraft. These leaks were caused by material failures, such as broken seals, or by improper maintenance practices or other reasons.
Some of these cases involved aircraft of the same type as in the event analyzed in this report and led to the issuance of a safety recommendation. As a result, measures were adopted to improve maintenance practices so as to avoid having leaked engine oil contaminate the bleed air due to incorrect maintenance operations.
There is no record of any event having taken place to date in which physical effects in individuals were determined to have resulted from repeated exposures to normal cabin air, i.e. the air supplied by the air conditioning system from the engine or APU bleed with these components operating normally and without any oil, hydraulic fluid or other leaks.
2.6. Current situation
As noted in various sub-sections in point 1.18, there is currently a great controversy as to the potential contamination of aircraft cabin air and its effect on people’s health.
The information available shows an increase in the number of cases reported in recent years in which an alleged exposure to contaminants in cabin air resulted in physical and cognitive symptoms. This could be due to an actual increase in the number of cases or to increased reporting due to crews’ heightened awareness of this issue. None of the parties seems to call into question the toxicity of certain compounds present in engine oil, such as TCP. The differences mainly involve the cumulative effects of exposures to small concentrations of these compounds, to the effects of altitude and to the toxicity thresholds, which some argue are based on exposure to a single toxic agent, while cabin air can contain many of these agents.
While the use of engine oil without TCP would minimize the risk of contaminating the cabin air with organophosphates, it would not completely resolve this problem, since the possibility remains that other toxic agents can contaminate the air. The tests and research carried out to date do not offer any definitive conclusions on the effects that cabin air has on human health; moreover, several studies recommend additional research and there are several such initiatives ongoing.
There also have not been any epidemiological studies that offer a scientific perspective on the real impact that cabin air has on human health.
In light of the concern that various aviation stakeholders have in this area, as well as of the increased number of cases of physical symptoms allegedly caused by contaminated cabin air and the little real knowledge on the effects that cabin air quality has on health, it seems prudent to continue the studies and research needed to gain an adequate understanding of this subject.
The CIAIAC released following safety recommendation to the ICAO: “It is recommended that the International Civil Aviation Organization (ICAO) monitors research and/or studies conducted by organizations representing civil aviation, authorities, the industry and academic and research institutions to determine the real impact that exposure to contaminated cabin air has on human health and takes actions to improve safety, as necessary.”
By Simon Hradecky, created Saturday, Mar 23rd 2013 17:20Z, last updated Tuesday, Mar 18th 2014 14:07Z
|A Condor Boeing 757-300, registration D-ABOC performing flight DE-5944 from Hamburg (Germany) to Las Palmas,CI (Spain) with 242 passengers and 8 crew, was descending towards Las Palmas when an odour on board caused three flights attendants to feel unwell, the first officer donned his oxygen masks. The aircraft continued for a safe landing.
The airline confirmed the odour on board and reported that three flight attendants felt unwell, the first officer donned his oxygen masks on approach to Las Palmas’ Gran Canaria Airport. The other crew members and passengers did not report feeling unwell. After landing the crew activated the APU to determine the cause of the odour, following the activation the odour re-intensified causing two flight attendants to become temporarily unconscious. The flight attendants and first officer were taken to a hospital in Las Palmas, released, and are on the way home. The aircraft had undergone a C-Check in March 2013. Spanish and German Authorities are investigating.
The German BFU confirmed they were informed about the event as described, Spain’s CIAIAC is investigating the occurrence with the assistance of the BFU.
Passengers described a strong odour of oil fumes throughout the flight, they felt unwell with head aches and dizziness.
The return flight DE-5945 was postponed to the next day, a replacement Boeing 767-300 registration D-ABUC was dispatched to Las Palmas and reached Hamburg with a delay of 19 hours.
The occurrence aircraft resumed service on Mar 26th 2013.
On Apr 17th 2013 the Spanish CIAIAC reported that the aircraft had undergone de-icing before departure from Hamburg. Departure and cruise had been uneventful, during the approach at about 6000 feet the flight crew noticed a strong smell in the cockpit that seemed to originate from the air conditioning outlets. Immediately after the purser called the cockpit reporting that the strong smell was perceived in the cabin, too. About 2 minutes later the first officer indicated he felt unwell with dizziness, the captain recommended to use the oxygen mask, the first officer donned his oxygen mask and felt immediate improvement. The landing was continued without further incident, the first officer removed the oxygen mask during taxi. After the passengers had disembarked, preparations for the return flight began, company dispatch instructed to have the engines checked for bird ingestion and verify hydraulic and oil quantity levels, check the waste water lines and the air conditioning particle filters, no anomalies were identified in these tests. An engine run near the threshold of runway 03L was coordinates with tower, the aircraft was towed to the runway, only the APU was running at that time, on board was flight and cabin crew as well as a maintenance technician and a company operator. Near the threshold the crew connected the APU bleed with the left hand air conditioning system, which right away resulted in a strong smell, two cabin crew suffered from physical problems. The air conditioning system and APU bleed air was disconnected, all aircraft doors opened to ventilate the aircraft, oxygen was provided to the two flight attendants and paramedics called in who took the flight attendants to the hospital, where they stayed over night. 2 CIAIAC inspectors were dispatched to Las Palmas who together with company technicians examined the aircraft however without finding any anomaly. An aerotracer device found traces of glycol and Pattex (adhesive) in the cabin air. Another repeat of the tests performed by the crew near the runway threshold did not produce any smells, all tests remained negative. The only finding remained about 5 liters of glycol spilled in the APU compartment, remnants of the de-icing in Hamburg, which were removed before the aircraft returned to service.
On Mar 18th 2014 the CIAIAC released an interim statement stating: “In April 2013 the health of one of the flight attendants who had been onboard during the flight of 22 March 2013 worsened, requiring hospitalization. The symptoms presented were overall muscle fatigue, in particular proximal of the lower limbs, difficulty walking, sensory disorder, trouble concentrating and general fatigue. She was released from the hospital and continued treatment on an out-patient basis. The symptoms persisted and her health did not show improvement, even worsening at times to the point where she had to be hospitalized. As of the date of this interim report, she still has not been able to return to work. Although the tests performed on her have not been able to identify the cause of the symptoms afflicting her, the medical report from the hospital indicated poisoning caused by some type of neurotoxin.”
The CIAIAC reported that airline technicians in the presence of two CIAIAC inspectors exhaustively examined the aircraft at Las Palmas but did not find any anomaly. The tests were repeated with crew on board again with no findings, the crew did not smell anything and suffered no physical alteration, the aerotracers used during that test did not register anything abnormal. Another test with all combinations of possible configurations of air conditioning systems also did not detect any anomaly. The engines, air conditioning ducts, hydraulic lines, APU etc. were checked, the only noticeable finding was about 5 liters of glycol spilled into the APU compartment. The glycol was removed.
On Mar 26th 2013 the aircraft was ferried to Frankfurt, all test equipment and the technicians were on board that flight, the aerotracer was operational throughout the flight.
About 100 minutes into the flight the aircraft encountered light turbulence during which intense smell filled the cabin prompting the flight crew to don their oxygen masks. Nonetheless, the first officer as well as the purser felt their tongues going numb and their throat being irritated. The turbulence stopped after about 10 minutes and the smell dissipated. The crew removed the oxygen masks, the irritations and numbness ceased.
While descending towards Frankfurt the odour returned, the pilots again donned their oxygen masks. The purser felt her fingers going numb. The smell and symptoms ceased when the aircraft descended through 6000 feet.
The operator requested assistance by the aircraft manufacturer who deployed a specialist team to Frankfurt. Samples were taken during flights and analysed in laboratory without finding any anomaly.
The CIAIAC concluded the interim factual report: “A blood sample taken from the flight attendant was sent to a laboratory in the United States that specializes in neurotoxin poisoning, specifically in devising methods to identify the presence of damage to the nervous system that is usually caused by these substances. The analysis of the sample concluded that it exhibited characteristics consistent with damage to the nervous system.”
The CIAIAC stated the next steps of the investigation will be:
– Continue monitoring the physical condition of the two FAs who have not been able to return to work so as to determine the cause of their ailments.
– Investigation into the analytical methods used to identify toxins.
– Investigation to determine the source of the odor.
– Joint identification and review of similar cases with the German accident investigation authority (BFU).
To read the full report click here.