In Dec 1984, a large-scale chemical disaster occurred in Bhopal, India. An explosion at the Union Carbide India pesticide plant released toxic gas in the form of methyl isocyanate (MIC) and its reaction products over the city. The estimated mortality of this accident is believed to have been between 2500 and 5000 people, with up to 200,000 injured (Mehta et al., 1990). The Bhopal event was the worst industrial disaster ever, and provides a classic case to study from an environmental health viewpoint, as it raises important issues, not only in terms of toxicology, but also in terms of occupational health and safety, air pollution, epidemiology, risk assessment, disaster management, and environmental protection. There has been much written about Bhopal, pointing out the lessons which we should learn, to prevent tragedies of this kind from occurring in the future.
Bhopal is the capital of the State of Madhya Pradesh, and lies 744 km south of New Delhi. Its population last decade was approximately one million people (Mehta et al., 1990). Union Carbide, an American company and one of the largest industrial companies in the world, established a chemical plant at Bhopal in the late 60’s, with the aim of supplying pesticides to protect Indian agricultural production. The main product was Sevin, a carbamate insecticide involving methyl isocyanate in its production. Initially, MIC shipped from the States was used in Sevin production, but in the late 70’s a plant was constructed locally for manufacturing methyl isocyanate at Bhopal. This plant was located on the outskirts of Bhopal. A densely populated shantytown had grown up near the plant, with an estimated 100,000 people living within a 1 km radius of the plant at the time of the tragedy (Jackson, 1993).
A colorless liquid with a low boiling point of 39°C, MIC, when in contact with water, causes an exothermic reaction resulting in the formation of carbon dioxide, methylamine gases and nitrogenous gases. Fire involving MIC results in hazardous decomposition products such as hydrogen cyanide, oxides of nitrogen, and carbon monoxide. Up until 1984 little data existed regarding MIC toxicology. Animal toxicology studies in 1964 had established the lungs as the primary site of injury from MIC. In 1978, the National Institute for Occupational Safety and Health published guidelines for MIC use. The permissible exposure limit of MIC was documented as 0.02ppm averaged over an 8hr work shift. It was stated in the guidelines that at 2ppm, no odor is detected, but subjects experience eye, nose and throat irritation and lacrimation. At 4ppm the symptoms of irritation are more marked, and at 21ppm, exposure is unbearable. Little was known, prior to the incident, about the toxicity of MIC reaction products, or about treatment for toxic effects due to MIC exposure (Mehta et al., 1990).
Generally stored in liquid form in holding tanks, there were 2 steel MIC holding tanks at the Bhopal plant, of 57,120 L capacity each. It is noteworthy that in plants in the USA and Europe, the storage tanks have smaller capacities for safety reasons - 17,500 L each, with allowable filling of 50% (Mehta et al., 1990). In the tanks at Bhopal, the amount of MIC stored was more than ten times the amount required for daily use. In retrospect, this is an example of poor safety management at the plant.
According to the scientific and legal investigations carried out by Union Carbide following the incident, evidence showed that an employee at the Bhopal plant had deliberately introduced water into a methyl isocyanate storage tank, with the result being the release of a cloud of poisonous gas (Jackson, 1993). The event occurred in the early hours of the morning of Dec 3rd 1984, at approximately 12:30 a.m. As the deadly cloud of gas drifted through residential areas surrounding the plant, residents woke coughing and choking, with stinging eyes. By 2 am most of the MIC had been dispersed over an area of 40km. The first deaths were reported to the police by 3am, and by morning, there had been 1000 reported deaths, some as far as 8km from the plant. 90,000 patients were seen in local hospitals and clinics within the first 24 hrs, and in total, about 200,000 people suffered acute effects of the MIC leak.
Atmospheric conditions at the time exacerbated the danger, as winds blew the gases in the direction of settlements near the plant, and temperature inversion reduced the dilution of the gas cloud. Twenty-seven tons of toxic gas was dispersed over the city, traveling slowly southward from the plant, exposing one quarter of the city’s inhabitants (Cullinan et al., 1994). The initial exposure to MIC was therefore by inhalation. The extent to which subsequent exposure from contaminated water and food supplies might have occurred is unknown. A study of thiocyanate levels in Bhopal Lake and tap waters 15 weeks after the incident found levels to be twice as high as those in Bombay. At the end of 9 months, these levels had returned to baseline (Mehta et al, 1990).
The acute toxicity of inhaled MIC or its reaction products was devastating - most of the fatalities occurred during the first week. In the few days after the accident, treatment was limited to symptom management, as it was still uncertain whether the effects observed were due to MIC, phosgene, HCN, or other MIC reaction products. With regard to its potential for long-term toxicity, it has been shown that MIC exposure has resulted in chronic respiratory illness among Bhopal residents (Cullinan et al., 1994), but few long-term studies have been carried out since the disaster occurred, and the extent of ongoing health problems is poorly documented.
The immediate health effects observed in those exposed were respiratory problems, eye irritation, and CNS, GIT and circulatory symptoms. The most common and serious of these was respiratory involvement.
The acute effects of heavy exposure were consistent with inhalation of a highly irritant and corrosive aerosol. Symptoms included breathlessness, choking, cough, chest pain and hemoptysis. Large numbers of people died rapidly of bronchial necrosis and pulmonary edema. Acute toxicity studies in animals later revealed the effects of MIC in causing inflammatory reaction, destruction of alveolar architecture, damage to bronchial epithelial lining, and pulmonary edema, suggesting that MIC predominantly damaged lung tissue due to its corrosive action.
Several months later, the pulmonary outcomes of survivors, as evidenced on X rays, appeared to be obliterative bronchiolitis and interstitial fibrosis (Cullinan et al., 1994). Thirty nine percent of 783 patients showed ventilatory impairment by lung spirometry tests (Mehta et al, 1990). There were no published studies regarding long-term pulmonary effects until a randomized retrospective cohort study was undertaken by Cullinan et al. (1994). In this study, lung function testing of 454 Bhopal residents exposed to the gas in 1984 showed that persistent small airways obstruction was present, 10 yrs after exposure. There appeared to be a direct gradient in effects according to an estimate of the intensity of gas exposure at the time (estimated by the patient’s distance from the plant in kilometers, at the time of the incident.)
Acute ophthalmic effects were common, including severe watering of the eyes, photophobia, profuse lid edema, and corneal ulceration. Blindness or irreversible eye damage was not seen. Prior to Bhopal, animal studies had shown that low molecular weight methyl isocyanates had proven to be skin and eye irritants, and when applied to rodents’ eyes, had caused severe necrosis. Animal studies conducted after the incident concluded that MIC causes acute eye injury, but probably does not cause permanent serious eye damage (Mehta et al, 1990).
Maternal-fetal, gynecological, and genetic effects have been illustrated through retrospective cohort studies. In an epidemiological survey 9 months after the accident, it was seen that 43% of 865 pregnancies amongst exposed women had suffered fetal loss, as compared to 6-10% among the general Bhopal population. The spontaneous abortion rate was highest among those exposed during their 1st trimester. Mehta et al. (1990) reported a study by Shilotri et al. in which there was a higher incidence of abnormal uterine bleeding and abnormal Pap smears amongst exposed women 15 weeks after the exposure. Thre also reported on Goswami, who carried out chromosomal studies on the exposed population using lymphocyte culture methods, and found that as late as 12 months after the exposure, 71% showed evidence of chromosomal aberrations as compared with 21% in a control population residing 20 to 50 km from the plant.
One area not adequately studied is the risk of fetal toxic effects as a result of in-utero exposure. As Mehta (1990) comments, “epidemiological studies on infants born several months following the disaster will be needed to uncover such effects due to MIC exposure”. Likewise, there has been a lack of attention to the possibility of long-term toxic effects in the exposed pediatric population, which is disturbing since so many children were affected by the explosion. Over 1000 children presented with breathlessness, cough, watery eyes, diarrhea and vomiting. Some had convulsions, hemiparesis and coma, and 119 deaths were reported in the first 12 days.
Few immunological toxicity studies of MIC have been reported. A study of humoral and cell mediated immunity in exposed subjects two months after exposure found that cell-mediated immunity was suppressed, and that MIC-specific antibodies persisted for several months. Several studies of Bhopal survivors suggest long-term immunological effects, including the potential of MIC to produce hypersensitivity reactions (Anderson, 1989).
No significant hematological or biochemical abnormalities were noted in Bhopal residents. However, if MIC had gained entry into the bloodstream, then systemic toxic effects would be expected, and would be reflected by abnormal hematological and biochemical values.
The serious psychological effects that disasters can have on the mental health of victims are well recognized, the most common of which is Post-traumatic Stress Disorder. According to McFarlane (1986), the incidence of PTSD following a natural disaster can be 30 - 59%, and as high as 80% following a man-made disaster. It would be expected that the exposed population of Bhopal would have a high incidence of psychiatric symptoms, but again this seems to be a neglected area that has not received much attention.
The tragic consequences of Bhopal raise issues of both a practical and philosophical nature. This discussion will focus on the toxicological issues illustrated in this case and will also look at the bigger picture; which is the effect of industrialization on the developing world. Industry is not always good - it can kill, as evidenced by Bhopal, and in poor countries, industrial risk is high. Every day in developing countries there are a wide range of industrial accidents, and people are poisoned over time by environmental pollution. As pointed out by the World Health Organization “in most developing countries there are no effective legal or institutional structures to deal with pollution in the workplace or surrounding areas” (Garner, 1997). Large multinational companies that set up industrial operations in the third world must take responsibility for their operations and actions. The onus must be on them to ensure that safety regulations are met with regard to the health of their workers, and the surrounding environment. It is their responsibility to operate under a minimum set of standards even if local laws do not impose these. Economic development in poor countries shouldn’t occur at the expense of health.
Ensuring that disasters like this do not continue to occur in the third world will require much attention to safety in the industrial workplace. Moeller’s (1992) basic principles of occupational health and safety can be applied well to Bhopal. The primary step is to reduce the risk associated with use and production of toxic chemicals, ideally through elimination or substitution of the toxic chemicals being used for less toxic ones. A review of processes used should aim to contain hazardous material within sealed equipment as far as possible, and to minimize exposures, and also minimize unwanted by-products. Isolation or enclosure of highly toxic materials is paramount. Adequate security is also essential. For such an act of sabotage to occur as it did at Bhopal, there can not have been an adequate security system in place.. Adequate local exhaust ventilation or air cleaning facilities within a plant are vital, as are personal protective equipment for workers, and ensuring safe work practices and hygiene.
An important issue also raised by Bhopal is the location of industrial plants. Legislation is required to ensure that dangerous sites do not exist in close proximity to heavily populated areas. All industrial plants should be built with the potentiality of disaster in mind, thereby minimizing the risk to the population if a spill, fire or leak occurs. In the case of Bhopal, it appears that the dense population distribution around the plant occurred as a result of people needing to live close to work. Local governments may need to intervene with residential restrictions to prevent similar scenarios from occurring around each new industrial site. It must also be the responsibility of multinational corporations to provide communities and governments with public information about the character and extent of releases of toxic materials from industrial processes.
Effective disaster management for any environmental accident such as the explosion at Bhopal needs to involve the following steps:
Although at Bhopal, the provision of care for the ill and injured was seen as an urgent priority, there did not occur the establishment of a surveillance and epidemiological study system to address the aftermath of the acute exposure. Initially, this was prevented by the chaos and trauma that followed the disaster; later inadequate resources compromised study design. Ideally, there should be the provision of facilities needed to obtain immediate epidemiological data at the onset of a major toxic disaster, monitor the health of the surviving victims, provide what treatment is possible, study the causes of toxic effects, develop a cure, and plan for future prevention (Mehta et al., 1990).
Both the exposure assessment and toxicity assessment following the Bhopal incident were incomplete. As mentioned earlier, one of the primary problems was the uncertainty about whether the observed effects were due to MIC itself, or due to a reaction product of MIC. No specific tests were available to confirm the cause of injury. Hydrogen Cyanide was implicated in the acute injury, and can be detected by measuring increased urinary thiocyanate levels in survivors, but unfortunately, high urinary levels of this compound can be explained also by other factors such as water contamination and dietary habits. No timely attempt was made to recreate the incident in animal experiments to examine the effects, but later animal studies suggested that cyanide was not involved in the fate of Bhopal survivors. Acutely, MIC is several times more toxic than HCN, but the danger of a mistaken emphasis on HCN is that cyanide, unlike MIC, is unlikely to produce long-term effects (Anderson, 1989).
Measuring personal exposure levels to MIC was difficult, as there had not been identified any biomarker of exposure, such as a blood chemistry level. Establishing a dose-response relationship between the environmental exposure of MIC and the health events was not possible, because wide-scale measurements of MIC or its breakdown products could not be obtained in time (Koplan et al., 1990). Most approximations were based on imprecise measures of exposure - usually the distance an individual lived from the plant. Several investigators used this simplified dose-response approach to compare health events in communities located 2, 4 and 8 km from the plant. Exposure would have been affected by other factors of course, such as the degree of ventilation of the house, and the shielding of the patient from the MIC vapors (Koplan et al., 1990).
In Koplan’s (1990) commentary on Bhopal, it is noted that the information available at that time (6 years after the event), was not significantly different from that available in the first weeks after the gas leak. This illustrates again, the need for long-term studies to be instituted after such events, to determine their long-term health effects. The price of not doing this is the cost of not recognizing, and not treating, significant ongoing health and psychosocial problems in the exposed population. The simple, but effective respiratory study (Cullinan et al., 1997) was a case in point of this.
In summary, the Bhopal disaster raised difficult and unaddressed issues such as how to prevent potentially dangerous chemical plants from being located in heavily populated areas, how to ensure that as much as possible is known about the risks and effects of toxins being used or produced, how to ensure the safe operation and maintenance of industrial facilities, and how to develop effective disaster plans for the protection of both workers, and nearby residents.
Anderson N (1989) Long-term effects of methyl isocyanate. The Lancet, June 3, p.1259.
Cullinan P, et al. 1997) Respiratory morbidity 10 years after the Union Carbide gas leak at Bhopal: a Cross Sectional Survey, BMJ 314:338-43.
Garner P (1997) Commentary: Industry can damage your health. BMJ 311:342.
Jackson B (1993) Disaster at Bhopal. Report by retired Vice President, Health, Safety and Environmental Programs, Union Carbide Corporation.
Koplan J, et al. (1990) Public health lessons from the Bhopal chemical disaster. JAMA 264:2795-96.
McFarlane A (1986) Implications for disaster planners and emergency services. MJA 145:561-63.
Mehta P, et al. (1990) Bhopal tragedy’s health effects, a review of methyl isocyanate toxicity. JAMA 264:2781-2787.
Moeller D (1992) Environmental Health. Harvard University Press, Boston