Vi använder kakor för att förbättra webbplatsen

Den här webbplatsen använder kakor (cookies) för att underlätta ditt besök och för att göra webbplatsen bättre. Läs mer om kakor.


The explosion in the artificial fertiliser factory in France 2001 – KAMEDO-report 86

The Kamedo-reports are published by the Swedish Disaster Medicine Study Organisation (Kamedo), at the National Board of Health and Welfare. Observers study the medical, psychological, organisational and social aspects of disasters. The results, with a focus on experiences gained, are presented in the reports.

This report is entirely in Swedish. Only summary in English.


The event

On the morning of 21 September 2001 there was an extremely powerful explosion in the city of Toulouse in France. It occurred in a factory that makes artificial fertiliser. It later transpired that a chemical reaction had taken place in a hangar where sensitive industrial waste is stored – mostly residual products containing ammonium nitrate.
None of the city's inhabitants could have escaped noticing the explosion. It was caused by an earthquake measuring 3.4 on the Richter scale and a powerful pressure wave, bringing serious damage. A cloud of dust and smoke formed, and at first it was not known whether or not this cloud was poisonous. Later it was possible to confirm that the explosion had left be-hind a crater nearly 5 metres deep and 50 metres in diameter.
The disaster caused the loss of 30 lives. The Toulouse Prefecture esti-mated that about 3,500 people were injured as a result of the explosion, 50 of them seriously. During the first day 862 patients were taken into hospital. Afterwards about 40,000 damage claims were submitted to the insurance companies.

At the site of the disaster

The alarm system at the factory never went off, and major traffic problems quickly arose, since many cars were damaged. Despite this the first rescue team was on site a mere thirteen minutes after the explosion. They encoun-tered a stream of dusty, injured persons fleeing the industrial area on foot. Many industrial buildings were razed to the ground, and nearby residential buildings were in need of immediate evacuation.
People now began to realise the scope of the damage. Rescue work began, even though no risk assessment had been performed. Only after half an hour did the results of measurements show that the cloud had a low toxic content.
A good 20 minutes after the explosion a major disaster alarm was trig-gered in Toulouse. That was the starting signal for a huge rescue effort. Within twelve hours 1,046 firemen were on site, despatched from 13 differ-ent fire prevention districts. It quickly became clear that the number of res-cue staff exceeded requirements, but this made it possible for them to re-lieve each other.
After a few hours 60 doctors were present at the site of the disaster, and most of them carried out their duties at the assembly point a few kilometres away. During the course of the day nearly 300 injured persons were taken care of there.

At the hospitals

During the course of the day the two biggest hospitals – the univer-sity linked hospitals Rangueil and Purpan – received over 1,500 injured persons. At both hospitals the disaster plans were undergoing revision.
It was not at first clear whether it was possible for Rangueil Hospital to be used. It is the hospital closest to the site of the disaster, and it was damaged by the explosion. The hospital was temporarily evacuated, but following a brief inspection it was possible for work to resume. Most of the repairs were started that very day.
Soon after the return to the hospital the disaster alarm went off and the injured began streaming in. The hospital received 435 injured persons during the course of the day, of whom over a quarter needed to be admitted for medical care. In addition to this, medical care was received by 50 people injured at the hospital when it was damaged as a result of the explosion.
At Purpan Hospital they improvised and dealt with the massive onrush of injured persons in the main ambulance entrance. They conducted initial tri-age there, and then allocated the injured to various injury sectors in the hos-pital. During the day they received 1,048 injured persons, a quarter of whom were admitted.
Three quarters of the injured who were received at Rangueil Hospital and Purpan Hospital were able to leave the hospital the same day. Of those re-maining for medical care, 25 had suffered injuries, some of them serious. Four people were flown to other hospitals. The injured were also looked after in 24 other medical units, several of them private, and by private GPs.
Few injuries were confirmed which could be said to be directly attribut-able to pressure wave injuries, e.g. injuries to the ears and lungs.

Observers' experiences

• Ambulances manned with doctors on constant stand-by duty are an effi-cient way of quickly getting a high level of medical expertise at a disas-ter site. It quickly proved possible to find staffing for management of the medical work, but this was exacerbated by the fact that there was an initial lack of material resources and transportation.
• With accidents in special risk zones where toxic clouds can arise, the injured may become contaminated by toxic chemicals. There should then be the opportunity to quickly assess the need for decontamination in order to prevent the medical staff being exposed to dangerous sub-stances.
• When there are large numbers of injured persons you can help the hospi-tals deal with the initial onrush by setting up one or more collection points. These places should be identified previously and should be in-cluded in the disaster plan.
• Disaster management requires a good communication network. The pub-lic telephone network often gets overloaded. It is then important to have access to other established means of communication such as direct phone lines between hospitals that do not use the usual network, as well as radio connections.
• When a disaster occurs in modern society, the need for patient transpor-tation is quickly assessed. Most of the injured will be transported in pri-vate vehicles or ambulances. Helicopter transportation is seldom used, and air transportation is not usually available until many hours after the event.
• In the event of large numbers of injuries, all hospital resources in a town or region must be utilised. This should be taken into consideration when drawing up disaster plans – something which does not always happen in Sweden.
• The field of emergency medicine is now an established speciality in many Western countries, including France. This means that emergency doctors play a crucial role in reception in hospitals in the event of mass injury – and this is not the case in Sweden.
• The premises used for reception of a large number of injured persons should be of an adequate size. There must also be a readiness to open up extra areas. With the lack of space currently prevalent in Swedish emer-gency hospitals it is not possible to send a large number of patients home.
• Registration, identification and localisation of the injured is made easier if a sufficient number of disaster case records are available. In future, computer technology will facilitate other possibilities, e.g. barcode marking for every patient.
• The hospitals' disaster stocks should also include large signs that can be set up to show the way to the various sectors mentioned in the disaster plan, e.g. the sectors for the injured, relatives and the press.
• Looking after worried relatives is often difficult, and it calls for a lot of work. It is thus important to quickly gain access to social workers, coun-sellors, psychologists or psychiatrists.
• It is important that the deceased be correctly looked after. Bearing in mind forensic medicine considerations, precise guidelines for this should be included in all types of disaster plans.
• Regardless of their strength, explosions can vary considerably with re-gard to the damage they cause. The nature of the surroundings at the detonation site is of crucial importance. The commonest injuries will be those caused by flying objects or people being thrown around. Despite this, with similar events you should always suspect pressure wave inju-ries, principally to the ears and lungs.
• Pressure-wave injuries, principally to the lungs (blast injuries), can now be confirmed or excluded using computer tomography, measurement of blood gases and several hours of observation. A day's observation used to be recommended.

Read the full Summary


Susannah Sigurdsson
+46 (0)75 247 30 00