Txs95$p312

Persistent respiratory effects in survivors of the Bhopal disaster
Updated information and services can be found at: References
1 online articles that cite this article can be accessed at: Email alerting
Receive free email alerts when new articles cite this article - sign up in the box at the top right corner of the article To order reprints of this article go to: Thorax 1998;53(Suppl 2):S43–S46
Persistent respiratory effects in
survivors of the Bhopal disaster
William S Beckett
Departments of Environmental Medicine and Medicine, University of Rochester School of Medicine and Dentistry, Rochester, NY 14642, USA Respiratory morbidity 10 years after the Union Carbide gas leak at Bhopal: a cross
sectional survey
P Cullinan, S Acquilla, N Ramana Dhara, on behalf of the International Medical Commission on
Bhopal
Objective. To examine the role of exposure to the 1984 Bhopal gas leak in the development of persistent
obstructive airways disease. Design. Cross sectional survey. Setting. Bhopal, India. Subjects. Random
sample of 454 adults stratified by distance of residence from the Union Carbide plant. Main outcome
measures. Self reported respiratory symptoms; indices of lung function measured by simple spirometry
and adjusted for age, sex, and height according to Indian derived regression equations. Results.
Respiratory symptoms were significantly more common and lung function (percentage predicted forced
expiratory volume in one second (FEV1), forced vital capacity (FVC), forced expiratory flow between
25% and 75% of vital capacity (FEF25–75), and FEV1/FEV ratio) was reduced among those reporting
exposure to the gas leak. The frequency of symptoms fell as exposure decreased (as estimated by
distance lived from the plant), and lung function measurements displayed similar trends. These findings
were not wholly accounted for by confounding by smoking or literacy, a measure of socioeconomic
status. Lung function measurements were consistently lower in those reporting symptoms. Conclusion.
Our results suggest that persistent small airways obstruction among survivors of the 1984 disaster may
be attributed to gas exposure.
(BMJ 1997;314:338–43)
The effects of inhalation of a highly irritant gas on the Other factors which could contribute to respiratory subsequent morbidity of long term survivors is the symptoms (smoking, age, sex, and socioeconomic sta- question addressed in this population based survey of tus) were accounted for in the analysis by separating out symptoms and lung function carried out among the groups stratified by these factors to determine whether general population living in Bhopal, Madhya Pradesh.1 confounding—for example, higher prevalence of smok- Over a period of nine days a randomly selected sample ing among those living closer to the plant—could explain of 454 adults who in 1994 were living along the path the observed association between proximity to the plant believed to have been followed by a cloud of methyl on the night of the methyl isocyanate release and the isocyanate vapour on 3 December 1984 were in- presence of symptoms and pulmonary defects 10 years terviewed. All gave information about their own ex- posure and their current respiratory symptoms. A Figure 1 shows symptoms in relation to the distance randomly selected 20% of these were invited to have subjects lived from the plant in 1984. Symptoms specific spirometric tests before and after salbutamol and results for airway pathophysiology (phlegm, strongly correlated were obtained on 74, 82% of those selected.
in previous anatomicopathological studies with mucus Respiratory symptoms inquired about were breath- gland hypertrophy)2 and wheeze (the symptom most lessness when walking on the level or climbing hills; strongly correlated with airway hyperreactivity in pre- cough for more than three months; phlegm for more vious studies),3 had the lowest prevalence in both the than three months; and wheeze in the past year. Symp- exposed and control (residence >10 km from the plant) toms and lung function were analysed in relation to populations. It is not clear from this paper whether distance from the source of the gas release, based on subjects were asked about “phlegm on most days for the assumption that the inhaled concentration of vapour more than three months each year, for two or more diminished by dilution with ambient air as the cloud consecutive years”, the question often used to define moved southward from the pesticide manufacturing chronic bronchitis in epidemiology. Yet although dys- pnoea was reported in 50% of those living at a great ticularly when the mean effect observed is relatively small, one cannot be sure whether the same association would be seen if the 19% of individuals randomly selected but never tested had also been included in the study.
Stratification of these subjects by smoking history weakened but did not abolish the association betweenestimated exposure and lower mid expiratory flow rate.
The number of subjects in the lung function study(74) did not permit multivariable analysis which allows independent examination of the size of the effect of different factors (such as proximity to the plant andsmoking). Only two of the 74 subjects tested, both ofwhom had lived <2 km from the plant, had an increasein forced expiratory volume in one second (FEV1) of200 ml and >15% after salbutamol.
The investigators do not address the possible effect on respiratory symptoms of current air pollutant ex- posures in relation to distance from the Union Carbide Trends in reported respiratory symptoms in 454 plant. Very large cities may have gradients of ambient survivors 10 years after the Bhopal incident, still living in pollution exposure, symptoms, and lung function re- their original residences, according to distance from the lated to variability in exposure between neigh- Union Carbide plant. Reproduced with permission fromCullinan et al.1 bourhoods.4 Whether such an effect might be presentin Bhopal at the present time is not discussed.
Thus, in summary, the investigators observed a cor- distance from the plant, a consistent gradient in symp- relation between a crude estimate of past exposure to toms was seen for each of the symptoms.
irritant gas and current symptoms and mid-expiratory Part of the gradient seen in fig 1 could be associated flow rate, but not with FEV1, in survivors of the Bhopal with a trend for lower income (lower socioeconomic status is usually strongly associated with respiratorysymptoms) in those who had lived closer to the plant.
However, the opposite trend (higher symptoms with Significance of the study
greater distance from the plant) would be expected After a disaster in which thousands died mainly from based on the higher reported “ever” smoking rate among respiratory failure in the first hours and days after those living at a greater distance from the plant. It exposure, what is the point of demonstrating a residual should be noted that the overall prevalence of ever effect on the lungs of survivors? In addition to the having smoked cigarettes in the interviewed population provision of medical care and compensation to victims, was 14%, consistent with the low current overall smok- ing prevalence in India compared with the UK or USA.
The surprising answer is that relatively little is known Less susceptible to potential reporting bias are the about the long term outcome for adults who survive a lung function measurements (fig 2) expressed as per- single major “gassing” episode such as this. The broad centage predicted for a reference non-smoking adult implications of Bhopal for the safety of communities Indian population reported in 1987. While a relationship living near major industries were immediately evident.
between proximity to the plant and lower percentage Such communities become more frequent as many of predicted lung function might be suspected from ex- the world’s large industrial cities continue to expand.
amining the figure, a statistically significant result was The number of citizens affected was unprecedented but seen only for mid expiratory flow rate (FEF25–75). Par- similar exposures, usually to much less irritant clouds,are a frequent occurrence in industry and have been anoccasional occurrence in community episodes world-wide. Clinical experience with small numbers of patients ffected in such episodes would suggest that most sur- vivors who become symptomatic within hours after exposure are nonetheless free of residual effects 10 yearslater. Yet follow up of a number of severely affectedgroups has also shown that long term effects are possible.
As is the case with all epidemiological studies, this study can only demonstrate associations—in this case the association between distance from the origin of thetoxic cloud and the prevalence of symptoms and lowerFEF25–75. Yet it adds an important dimension to followup of Bhopal survivors by its random selection of sub- jects and use of an objective, if only approximate,surrogate index of exposure. The importance of the random selection of subjects made in this study cannot be overemphasised. After a major toxic exposure, avoid- ing bias in the selection of study subjects can onlybe overcome by this approach. The group studied is Relation between respiratory function and distance lived from the Union Carbide plant in 74 survivors representative of survivors still living in the community 10 years after the Bhopal incident. Percentage predicted and omits those who died from any cause during the 10 lung function is based on published norms in Northern years after exposure and those who left the community.
Indian adults. Reproduced with permission from Cullinan etal.1 Demonstration of cause and effect associations are Persistent respiratory effects in survivors of the Bhopal disaster beyond the scope of any epidemiological studies. Most studies of respiratory response to environmental ex-posures are weakened by imprecise estimates of ex- posure of the respiratory system. This quantitativeuncertainty in exposure results in quantitative un- certainty about the magnitude of the effect.
The lack of a larger and more comprehensive epi- demiological follow up of Bhopal survivors has been noted,5 yet follow up studies conducted by a numberof groups have produced a coherent, though incomplete, picture of the natural history of methyl isocyanate inhala-tion toxicity. This latest study, conducted on behalf of the International Medical Commission on Bhopal, addsimportant new perspectives to that picture.
Within hours of the event, groups of affected patients were assessed and then followed up by several differentinvestigators. The degree to which information from Pulmonary function results in relation to severity of exposure in 60 patients with persisting respiratory such small cohorts is applicable to larger groups depends symptoms 1–7 years after the Bhopal incident. One way on how representative the cohorts are of the larger ANOVA: FEV1/FVC% declined as severity of exposure groups. From the many follow up cohorts I have selected increased (p=0.05). NS=non-smoker; SM=smoker; ANOVA: analysis of variance. Dark stippled bars=FVC % predicted; two. Patients who presented with respiratory symptoms open bars=FEV1 % predicted; striped bars=FEV1/FVC%. were followed up with bronchoalveolar lavage (BAL) Reproduced with permission from Vijayan and Sankaran.6 and pulmonary function testing,6 and with sequentialrespiratory and immunological testing at 3, 6, 12, 18,and 24 months.7 The selection of subjects in these Significance of lung function findings
studies was on initial clinical illness, so these groups These two cohort studies selected subjects based on can be considered to have been less exposed than those clinical severity and, not surprisingly, found more who died within the first weeks after the event, but more marked physiological impairment over shorter periods exposed than the random population sample chosen by of follow up. The picture which emerges in those with Cullinan et al.1 Even so, within these groups there may acute respiratory illness is of predominant airway injury be overlap in disease severity, particularly as Cullinan with a less striking parenchymal effect, perhaps the et al chose a random sample living far away from the residue of an acute interstitial injury. Although not gas release, assumed to be healthy controls.
followed for 10 years, these survivors show incomplete Kamat’s follow up cohort consisted of patients re- resolution of an initial acute lung injury. The very mild porting exposure and immediate illness who had detailed obstructive findings in the randomly selected general clinical, physiological and psychological follow up for population at 10 years suggests a bronchiolar lesion in which they travelled 675 km to Mumbai or were Mum- some of the exposed population, even in those without bai residents visiting Bhopal in 1984. The investigators an immediate respiratory illness. Average values for lung noted an initial improvement in symptoms over the first function in an epidemiological study population can be 12 months with a subsequent worsening over the next misleading. A few severely abnormal individuals can 12 months. Two years after exposure, dyspnoea and produce the same average as a uniform but mild defect cough were more prevalent than phlegm, as in Cullinan’s spread over the majority. Also, depending on sample size, population based study. Lung function improved over quite severe defects in a small severely affected group who the first year and at two years declined slightly, with a were either more heavily exposed or more susceptible to restrictive pattern. It is interesting to note that the mid- toxic injury may not be detected as a statistically significant expiratory flow rate alone appeared to decline pro- effect of the exposure. It is worth noting that previous gressively over the period of observation. As is often the follow up studies of young cigarette smokers with ab- case in cohort studies, dropout of subjects can introduce normally low FEF25–75 values alone (normal FEV1) showed bias with effects difficult to estimate.
no increased risk for disability or mortality compared with Vijayan and Sankaran6 excluded patients with a his- those with normal FEF25–75 values.
tory of respiratory disease and included patients from a Because of a possible association of pre-existing re- clinic for disaster victims and symptomatic exposed duced lung function with susceptibility to toxic injury subjects from the community. Their 60 subjects were and the association of lung disease from any cause with studied once from one to seven years after exposure.
more symptoms and lower lung function, studies based The severity of exposure was based on initial illness on subjects identified by disease status rather than severity (including death of a household member). Using exposure status are susceptible to bias toward finding predicted values from North Indian subjects (but differ- more severe pathophysiology. It is in this respect that ent from those used by Cullinan et al) and a group of the study by Cullinan et al is most illuminating, sug- control subjects, they noted an association between gesting that the effects of brief exposure to methyl initial disease severity (as a surrogate for exposure) and isocyanate (and, by extrapolation, other highly irritant impaired lung function. Obstructive as well as restrictive gases) may persist for years after exposure.
patterns were seen in the more severely affected. More The detection of a mild and persistent airway lesion severe abnormalities on spirometric tests were seen in is not surprising in view of animal studies initiated in those estimated to have had higher exposures, including the wake of Bhopal. It is, however, an unusual finding those without a history of smoking (fig 3). Total cell as community based follow up studies of less severe gas counts (predominantly macrophages) were significantly releases, such as the release of 50 tons of chlorine from higher in the BAL fluid of non-smokers with severe a railroad tank car in 1978,8 have not shown persisting exposure. Interstitial markings on the radiograph were defects as have studies of some smaller scale industrial noted in both this group and Kamat’s, possibly also a releases affecting smaller groups of more intensely ex- posed workers.9 It is significant that, among the long L E A R N I N G P O I N T S
∗ A small study designed to provide a representative survey of all survivors of the Bhopal
gas release conducted 10 years after the event found a higher prevalence of respiratory
symptoms and lower mean mid expiratory flow rate associated with residence closer to
the site of release.
∗ The study did not find lower FEV1 (associated with risk for premature mortality) or evidence
of airway hyperresponsiveness as measured by bronchodilator response to salbutamol.
∗ In relation to other follow up studies from the Bhopal incident, including those with acute
respiratory illness, the toxic gas injury may be associated with a long standing macrophage
alveolitis and symptoms in some of the most severely affected survivors.
∗ While the average defect in mid expiratory flow rate is modest, the persisting effect of a
single gas exposure is noteworthy since such long standing effects from community wide
events have not previously been documented.
∗ The extent of the Bhopal disaster was much more severe than most common, smaller scale
industrial gas exposure events. Nonetheless, the findings, particularly if confirmed by
further studies, suggest that major industrial accidents can produce both acute respiratory
mortality and prolonged respiratory morbidity.
term effects of a single massive industrial accident, An unanswered question
residual lung damage may be found among the general The prevalence of airway symptoms 10 years after ex- population living downwind of the released gas. It shows posure is higher than would be expected from the how severe the exposure may need to be to produce a prevalence of mild airflow obstruction which, when lasting effect on the respiratory system. It would be present in young cigarette smokers, is usually asympto- interesting to know whether defects in lung function matic. The disparity could be explained by the presence among non-smokers without previous lung disease were of non-obstructive chronic bronchitis affecting larger found in individuals who were not acutely symptomatic airways, or persistent hyperreactive airways from irritant injury (known variously as irritant asthma, asthma withoutlatency, or reactive airways dysfunction syndrome). Inan editorial accompanying the study of acutely affected Relationship to other studies of large scale
patients by Vijayan and Sankaran, Nemery suggested the gassings
value of testing for non-specific bronchial hyperreactivity Studies of the effects of inhalation of irritant gases began in follow up studies such as this.13 Such tests would permit during the first World War when chlorine, sulphur- the detection of chronic airway hyperreactivity as a possible mustard, and phosgene were among gases used as chem- result of exposure to methyl isocyanate.
ical weapons. The pathologist Milton Winternitz per- This study was supported in part by NIEHS Grant P30ES01247 and by formed detailed studies of the lungs of experimental New York State Contract C-010126 to the Department of Environmental animals quantitatively exposed to these gases10 and drew Medicine, University of Rochester School of Medicine and Dentistry.
attention to the presence of chronic lesions which per- 1 Cullinan P, Acquilla S, Ramana Dhara, on behalf of the International sisted long after the acute injury had healed. However, Medical commission on Bhopal. Respiratory morbidity 10 years after although at that time Hutchinson’s spirometer could the Union Carbide gas leak at Bhopal: a cross sectional survey. BMJ
1997;314:338–43.
have been used for epidemiological studies, appreciation 2 Reid L, DeHaller R. Adult chronic bronchitis. Morphology, histo- of the need for measurement of respiratory physiology chemistry and vascularization of the bronchial mucous glands. Me-
dicina Thoracalis 1965;22:549–67.
in relation to environmental exposure was not to come 3 Chan Yeung M, Enarson DA, Vedal S, et al. Asthma, asthma like for many years. It was estimated that 100 000 soldiers symptoms, chronic bronchitis, and the degree of bronchial hyper-
responsiveness in epidemiologic surveys. Am Rev Respir Dis 1987;136:
died from gas inhalation and subsequent infectious complications in World War I. Assessments of the res- 4 Xu X, Dockery D, Wang L. Effect of air pollution on adult pulmonary function. Arch Environ Health 1991;46:198–206.
piratory sequelae of war gas exposure in survivors were 5 Dhara VR, Kriebel D. The Bhopal gas disaster: it’s not too late for that “cases of recurrent bronchitis, bronchiectasis and sound epidemiology. Arch Environ Health 1993;48:436–7.
6 Vijayan VK, Sankaran K. Relationship between lung inflammation, emphysema . . . appeared to have a clear association changes in lung function and severity of exposure in victims of the with severe pulmonary effects at the time of gas exposure Bhopal tragedy. Eur Respir J 1996;9:1977–82.
7 Kamat SR, Patella MH, Pradhan PV, et al. Sequential respiratory, . . .” but that “the most significant observation was that psychologic and immunologic studies in relation to methyl isocyanate the major proportion (of gas casualties evacuated to exposure over two years with model development. Environ Health
Perspect 1992;97:241–53.
field medical units) exhibited complete recovery.”11 The 8 Jones RN, Hughes JM, Glindmeyer H, et al. Lung function after chlorine idea that exposure to a single agent over a short period exposure. Am Rev Respir Dis 1986;134:1190–5.
9 Bherer L, Cushman R, Courteau J-P, et al. Survey of construction of time might produce clinically distinct outcomes—for workers repeatedly exposed to chlorine over a three to six week period example, bronchiolitis obliterans versus reactive airways in a pulpmill: II. Occup Environ Med 1994;51:225–8.
10 Winternitz, MC. Collected studies of the pathology of war gas poisoning. disease—based on variations in exposure or individual New Haven: Yale University Press, 1920.
susceptibility seems worthy of additional study. Sulphur- 11 Pennington AH. War gases and chronic lung disease. Med J Aust 1954; 1:510–6.
mustard gas victims from the Iran-Iraq War had a 12 Emad A, Rezain GR. The diversity of the effects of sulfur mustard gas spectrum of pathologies including asthma, chronic bron- inhalation on respiratory system 10 years after a single, heavy exposure.
Chest 1997;112:734–8.
chitis, bronchiectasis, airway narrowing due to scarring, 13 Nemery B. Late consequences of accidental exposure to inhaled irritants: RADS and the Bhopal disaster. Eur Respir J 1996;9:1973–6.

Source: http://bhopal.net/document_library/medical_documents/medical_documents/1998%20-%20Bhopal.pdf