Share this article on:

Biomass Stoves and Lens Opacity and Cataract in Nepalese Women

Pokhrel, Amod K.*; Bates, Michael N.*; Shrestha, Sachet P.; Bailey, Ian L.; DiMartino, Robert B.; Smith, Kirk R.§

doi: 10.1097/OPX.0b013e3182820d60
Original Articles

Purpose Cataract is the most prevalent cause of blindness in Nepal. Several epidemiologic studies have associated cataracts with use of biomass cookstoves. These studies, however, have had limitations, including potential control selection bias and limited adjustment for possible confounding. This study, in Pokhara City, in an area of Nepal where biomass cookstoves are widely used without direct venting of the smoke to the outdoors, focuses on preclinical measures of opacity while avoiding selection bias and taking into account comprehensive data on potential confounding factors.

Methods Using a cross-sectional study design, severity of lenticular damage, judged on the LOCS (Lens Opacities Classification System) III scales, was investigated in women (n = 143), aged 20 to 65 years, without previously diagnosed cataract. Linear and logistic regression analyses were used to examine the relationships with stove type and length of use. Clinically significant cataract, used in the logistic regression models, was defined as a LOCS III score ≥2.

Results Using gas cookstoves as the reference group, logistic regression analysis for nuclear cataract showed evidence of relationships with stove type: for biomass stoves, the odds ratio was 2.58 (95% confidence interval, 1.22 to 5.46); and for kerosene stoves, the odds ratio was 5.18 (95% confidence interval, 0.88 to 30.38). Similar results were found for nuclear color (LOCS III score ≥2), but no association was found with cortical cataracts. Supporting a relationship between biomass stoves and nuclear cataract was a trend with years of exposure to biomass cookstoves (p = 0.01). Linear regression analyses did not show clear evidence of an association between lenticular damage and stove types. Biomass fuel used for heating was not associated with any form of opacity.

Conclusions This study provides support for associations of biomass and kerosene cookstoves with nuclear opacity and change in nuclear color. The novel associations with kerosene cookstove use deserve further investigation.

Supplemental Digital Content is available in the text.





School of Public Health, University of California, Berkeley, Berkeley, California (AKP, MNB, KRS); N.D. Joshi Department of Ophthalmology, Manipal Teaching Hospital, Manipal College of Medical Sciences, Pokhara, Nepal (SPS); and School of Optometry, Minor Hall, University of California, Berkeley, Berkeley, California (ILB, RBD).

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (

Kirk R. Smith School of Public Health 747 University Hall University of California, Berkeley Berkeley, California 94720 e-mail:

Cataract, from opacification of the crystalline lens of the eye, accounts for more than 50% of reversible blindness worldwide.1 Globally, the prevalence of cataract is higher in females than in males, and in developing countries, cataracts occur at younger ages.2–4 Nuclear opacity is the most prevalent cataract type in South Asia.5,6 In Nepal, despite networks of eye hospitals throughout the country, cataract-related blindness is still prevalent and only slowly decreasing. For example, in 1981, the Nepal Blindness Survey documented a 3.8% prevalence of bilateral blindness (visual acuity <3/60), mainly because of cataract, in people older than 45 years.7 In 1998, the corresponding prevalence was 3.0%.8

Epidemiologic studies have indicated that environmental and genetic factors play an etiologic role in cataract formation.9,10 Nearly two-fifths of the world’s population uses biomass (wood, crop residues, and cow dung) as cooking fuel and sometimes space heating. Several epidemiologic studies have associated use of biomass-burning stoves with cataract.11–17 This is plausible because tobacco smoking has been implicated as a risk factor for cataract,18–23 and tobacco smoke and biomass fuel smoke have some common constituents.24,25 The studies of biomass, however, have had limitations, including possible selection bias in controls. Until now, the focus of epidemiologic studies investigating the use of biomass cookstoves has been the clinical diagnosis of cataract, usually by slit lamp examination. To our knowledge, however, no study has yet examined associations with the severity of lens changes or preclinical cataract. The main aim of this study was to investigate whether cataract at the preclinical stage is associated with the use of biomass fuel for cooking and heating in households, based on slit lamp examination, lens photography, and LOCS (Lens Opacities Classification System) III grading. Our study design avoids selection bias and allows testing for an exposure-response relationship based on reported histories of stove use.

Back to Top | Article Outline


Human subjects’ approvals were obtained from the institutional review boards at the University of California, Berkeley, and the Nepal Health Research Council.

We conducted a cross-sectional study from July to November 2006 among women aged 20 to 65 years visiting the Outpatient Department at Manipal Teaching Hospital, Manipal College of Medical Sciences, Pokhara, Nepal. The study was limited to women because, in Nepal, women do most of the cooking. The first five patients present at the Outpatient Department between 9:00 and 10:00 each morning were invited to participate. Excluded were women with congenital cataract, women having had previous diagnosis of cataract (because they will usually have had lens surgery), women with macular degeneration (which may have changed their cooking behavior), or women who had had penetrating eye trauma. Pregnant women were excluded for safety reasons because we dilated pupils for ocular photography. Women undergoing chemotherapy or with diabetes were also excluded because these are risk factors for cataract.4,26–28

Back to Top | Article Outline

Ocular Examination

After oral consent, participants’ visual acuities were measured at 3 m with an illuminated Bailey-Lovie visual acuity chart.29 Visual acuity was measured with habitual correction, with pinhole correction, and with best-corrected visual acuity. Each eye was measured separately, followed by measurement of intraocular pressure and examination of the adnexa and crystalline lens. We used the following criteria to categorize refractive error: (1) emmetropia, spherical equivalent between −0.50 and +0.50 diopter spheres (DS); (2) hypermetropia, spherical equivalent more than 0.50 DS; (3) myopia, spherical equivalent less than −0.50 DS; and (4) astigmatism, cylindrical error more than 0.50 D cylinder (DC) in any axis.

Participants’ pupils were then dilated to 7 mm or more with a mixture of 1% tropicamide and 10% phenylephrine. The crystalline lenses were then reexamined with a slit lamp; first with diffuse illumination, then with a slit beam, and finally with retroillumination. The slit lamp had a built-in digital camera (Topcon Model SL-D2), and lens photographs were taken when the pupil diameter reached 7 mm or more.

Back to Top | Article Outline

Lens Photography and Cataract Grading

Cross-sectional slit lamp photographs were used to document nuclear opalescence (opacity). The vertical beam height was extended to the maximum, and the breadth was fixed at the minimum. Retroillumination photographs were obtained to document cortical and posterior subcapsular cataract severity. The slit beam was fixed at 5 mm high and 1 to 2 mm wide. Each participant had a total of six lenticular photographs, consisting of one cross-sectional and two retroilluminated photographs per eye. All participants had photographs taken.

The digital photographs were sent to the School of Optometry at the University of California, Berkeley, where two experienced graders (I.B. and R.D.) scored each photograph by comparison with the LOCS III reference photographs.30 Compared with the simple categorization of cataract as present or absent, the LOCS III system can be used to quantify lenticular changes on four scales (nuclear opacity, nuclear color, and cortical and posterior subcapsular cataract), with fine increments. For both nuclear opacity and nuclear color, there are six photographs as exemplars of severity grades 1 through 6. Graders assigned decimalized grades from 0.0 to 6.0 in increments of 0.1. For both the cortical and posterior subcapsular opacities, there are five photographs as exemplars of severity grades 1 through 5. Possible scores are 0.0 to 5.0 in increments of 0.1.

Consistency of the corresponding scores between the two graders was evaluated graphically using Bland-Altman plots.31 This method plots differences between two scores against the average of the two scores. In the event that there were marked differences between graders that could not be readily eliminated by regrading, a consensus approach was used. In this process, batches of 10 photographs were presented simultaneously to the two graders who independently assigned a severity grade. After the independent gradings, the graders revealed their assigned scores to each other, discussed any differences, and came to an agreement on a consensus grade.

For the purposes of the logistic regression analyses, participants were characterized as having evidence of clinically significant cataract if they had a consensus score for nuclear opacity or, for the other color/cataract types, an average score between the graders of 2.0 or more in either eye. This criterion has been used in other studies.32,33

Back to Top | Article Outline


Questions were asked about each participant’s main present and previous cooking stoves: gas, kerosene, and biomass. Participants were asked about changes in the fuel type they used since they started actively helping or cooking in the kitchen, before or after marriage, and their duration of use of each stove type.

Questions also covered kitchen location and presence of kitchen windows, as well as presence in the home of other sources of indoor emissions, such as heating fuel, household lighting, and burning mosquito coils or incense. Detailed histories of tobacco smoking and alcohol consumption, exposure to environmental tobacco smoke, sunlight exposure (time worked outside each day and number of years of work), and protection from sunlight were collected. Other questions focused on socioeconomic status (annual family income, education, area of residency) and dietary practices.

Back to Top | Article Outline

Statistical Analysis

Information on kitchen location and presence of opening windows in the kitchen was combined to create a composite dichotomous ventilation variable.14 Kitchens were categorized as “fully/partially ventilated” when they were open-air kitchens or kitchens indoor or in a separate structure with an opening window. Kitchens were categorized as “unventilated” when they were separate kitchens outside without windows or partitioned kitchens inside without windows or nonpartitioned kitchens inside the house.

Pack-years of smoking was calculated as the average number of cigarettes or bidis (South Asian cigarettes with tobacco wrapped in a tendu leaf) smoked per day multiplied by the duration in years of smoking, divided by 20, assuming that a pack contains 20 tobacco products.34

The initial analysis was descriptive and examined the mean scores for participants categorized according to basic demographic variables and a number of known risk factors for cataract, including UV exposure,35–44 tobacco smoking,18–23 lower socioeconomic status,11,45–48and alcohol consumption49–51 after averaging scores across eyes and across graders. Average consensus scores for the two eyes were used for nuclear opacity. Analysis of variance tests were used to examine differences between the categories. Also examined descriptively, according to levels of the same variables, were the proportions of individuals designated as having cataract. Chi-square tests were used to examine differences between categories.

For each participant and type of opacity/cataract, there were two (consensus) scores for nuclear opacity and four scores (two graders and two eyes) for nuclear color, and cortical and posterior subcapsular cataract types. All scores were used in the analysis. We selected variables for the final regression models in the following way. First, we selected the variables that are directly related to exposure to participants (heating fuel and ventilation in the kitchen). We then identified remaining potential confounders of the association with stove type by examining (1) the association between stove type and potential confounders in the group without cataract (LOCS III score, <2) and (2) the association between stove type and potential confounders in the group using only the reference stove type (liquefied petroleum gas [LPG]). Any variables that had a statistical association of p ≤ 0.2 in χ 2 test in both (1) and (2) were considered for inclusion as additional covariates in the model. However, household lighting type was not included because virtually everyone used electricity or solar lamps; only two participants reported they used kerosene wick lamps.

Because opacity scores obtained from the two eyes of a single individual are likely to be positively correlated, their inclusion in the same statistical analysis violates the assumption of statistical independence and may result in overly reduced variances.52 This was addressed by using a robust-variance option in the multivariate risk models.53 This option adjusts for within-cluster correlation and gives robust SEs with unbiased 95% confidence intervals for the mean scores.54

Linear and logistic multivariate regression models were used to estimate associations of exposures with opacity. First, opacity/color/cataract scores on continuous scales were analyzed by multivariate linear regression with the robust option. Second, the scores were dichotomized into those with and without evidence of clinically significant cataract, applying a cut point score of 2.0. Odds ratios (ORs) for cataract were then calculated for nuclear cataract, nuclear color, and cortical cataract with logistic regression also using the robust-variance option. Because very few participants had posterior subcapsular cataract, we did not calculate the ORs for this cataract type.

Exposure-response patterns of clinically significant cataract by duration of use of biomass cookstoves were examined with logistic regression analysis. We used 20-year use duration categories, similar to those we used in a previous study.14

Back to Top | Article Outline


A total of 145 women aged 20 to 65 years were invited to participate. Of these, one woman who was pregnant and one with congenital cataracts were excluded. Full data were available for 143 participants. Most of the participants (70%) were emmetropic, with 15% having hypermetropia and 14% myopia. The prevalence of astigmatism was 13%. The mean and the SD of intraocular pressure were 14.0 and 2.9 mm Hg, respectively.

Visual inspection of Bland-Altman plots for nuclear opacity scores of the two graders showed a linear trend, with negative differences for low mean scores and positive differences for higher values, indicating a systematic difference between the two scorers. Compared with the other grader, one grader tended to give slightly lower scores at the low end of the range and higher scores at the high end. Consequently, the nuclear opacity photographs were regraded after some retraining of the scorers. The rescoring reduced, but did not totally remove, the trend evident in the Bland-Altman plots. Therefore, for the analysis of nuclear opacity, we used consensus scores. For nuclear color and cortical and posterior subcapsular cataract types, there was no evidence from the Bland-Altman plots of a systematic difference between scorers and 93% to 95% of scores fell within 2 SD of the mean opacity scores. For the analyses of these types, we used average scores across both eyes of the two graders.

Table 1 presents demographic and exposure data and mean scores for the three types of opacification/cataract and change in nuclear color. The mean age of participants was 45 years (SD, 12 years). Ninety-five percent were Hindu, 50% were illiterate (could not read or write Nepali), and 86% were urban residents. Forty percent currently used biomass stoves without chimneys or hoods, and none reported that they had ever used biomass stoves with chimneys or hoods. Fifty-seven percent reported using a gas cookstove (LPG or biogas), and only five participants (3.5%) used kerosene cookstoves. Ninety-nine percent of participants had electricity supply to their houses, and 79% reported well-ventilated kitchens. Forty-five (32%) reported that they had smoked tobacco; but of those, only 17 (38%) were current smokers. Each of the opacity/cataract types and nuclear color was statistically associated with several of the demographic and exposure variables.



The mean score for nuclear opacity was 2.11 (SD, 0.73); for nuclear color, 1.70 (SD, 0.76); cortical cataract, 0.95 (SD, 0.85); and posterior subcapsular cataract, 0.22 (SD, 0.46).

Table 2 presents the distribution of nuclear cataract (LOCS III score, ≥2.0) according to the set of demographic variables used in Table 1. It shows the prevalence of cataract increasing with age group, a higher prevalence in ever-smokers and with greater pack-years of smoking, and a higher prevalence in women who are illiterate. Although there were only five women who used kerosene for cooking, four (80%) of them had cataract compared with 55% in all the other participants.



Table 3 presents the results of multivariate logistic regression analyses separately by color/cataract type. Corresponding linear regression analyses by opacity/color/cataract type are in the Appendix, Table A1, available online at



Using gas as the reference category, nuclear cataract showed the strongest evidence of relationships with stove types: for biomass stoves, the OR was 2.58 (95% confidence interval [CI], 1.22 to 5.46; p = 0.01); for kerosene stoves, the OR was 5.18 (95% CI, 0.88 to 30.38; p = 0.07). Similar, if slightly weaker, relationships were observed for nuclear color (LOCS III score ≥2): for biomass stoves, the OR was 1.98 (95% CI, 0.82 to 4.74; p = 0.13); for kerosene stoves, 5.48 (95% CI, 1.23 to 24.37; p = 0.03). Cortical cataract showed no evidence of association with either biomass or kerosene stoves.

The linear regression analysis examined associations of variables in Table 1 with scores on a continuous scale. The logistic regression analysis examined associations between variables in Table 1, and scores were classified as clinically significant cataract (LOCS III score ≥2.0). The reference category for stoves was gas stoves.

The linear regression analysis produced positive coefficients for both biomass and kerosene stoves in relation to nuclear opacity and nuclear color scores, but CIs always included the null value, zero (Appendix, Table A1, available online at Negative coefficients were obtained for cortical opacity scores for both biomass and kerosene stoves. For posterior subcapsular opacity scores, a negative coefficient was found for biomass stoves but a positive coefficient for kerosene stoves. Both CIs included the null. For each opacity type, the coefficient for kerosene stoves was larger than that for biomass stoves (in the case of cortical opacity, in a negative direction).

As would be expected, increasing age (as a categorical variable) was positively associated with opacity categories in all models.

The linear regression models explained 35% of variance for nuclear opacity, 55% for nuclear color, 22% for cortical cataract, and 16% for posterior subcapsular cataract.

Use of biomass fuel used for heating showed no evidence of an association with any of the forms of opacity in linear regression analyses or cataract in the logistic regression analyses. Table 4 presents exposure-response relationships from logistic regression for each cataract type (LOCS III, ≥2.0) based on 20-year bands of biomass cookstove use. A corresponding analysis was not carried out for kerosene stoves because of the small number of participants who had used them. For increasing duration of use of biomass fuel, there was a monotonic increasing trend in the ORs for nuclear opacity (p = 0.01) and some evidence for nuclear color (p = 0.07). For the cortical opacity, the longest exposure category had the highest ORs, but overall trends were less apparent (p = 0.33).



Back to Top | Article Outline


The results of this study provide evidence that exposure to air pollutants from cooking with biomass or kerosene may increase the risk of nuclear cataract and change in nuclear color (LOCS III, ≥2.0) in this population of Nepalese women. We found little evidence of associations between biomass for cooking or heating, or kerosene use for cooking, with changes in cortical cataract.

Previously, seven epidemiologic studies have consistently suggested an association between biomass cookstove use and cataract or blindness.11–17 We are not aware of any such studies that found no evidence of an association. Potentially, however, selection bias, information bias, or confounding could have influenced the results of these studies. Selection bias in recruitment of participants was of concern in our previous study of cataract and cookstove use in Nepal. In that study, controls were selected from patients with refractive error.14 There is evidence of an association between refractive error and higher education, a surrogate for socioeconomic status, which is likely to be a correlate of stove type.55–57 Our controls seemed to be of higher socioeconomic status than the experimental population. Thus, it was possible that controls represented a different population, with different exposure patterns, including lower exposure to biomass fuel smoke, than the population represented by the cases. In the present study, we did not select controls, and all participants were drawn from the same patient population. The possibility of selection bias from recruiting only cooperative participants was avoided by a systematic recruitment process for all participants. There were no refusals to participate. Therefore, we believe that potential for selection bias is low in this study.

Information bias may take the form of outcome or exposure misclassification. In this study, LOCS III scores were confirmed by two scorers independently evaluating the photographs according to an agreed protocol. Because we found some observer inconsistency for scoring nuclear opacity, we addressed this problem by obtaining consensus scores.

Grading the severity of lenticular changes from a single slit lamp photograph is a complex task because identical severity scores may be given to cataractous changes quite different in their appearance. For the slit lamp photographs used to evaluate nuclear opacity, there was always a gradation of the quality of focus along the anteroposterior axis, and this could have influenced the severity scores. When there were large discrepancies between the two graders’ scores, they were usually caused by differences in the balance of the importance given to broadly distributed diffuse opacification versus well-defined regions of opacity. The consensus scoring procedure minimized the effect of these discrepancies. Scheimpflug camera systems can bring the whole cross section of the lens into focus at the same time, and this would eliminate the variation in the quality of focus across the photograph, making scoring easier. While Scheimpflug systems are expensive, they might be considered for use in future studies.58

One method of evaluating the validity of opacity scoring is to examine results for well-established strong predictors of opacity or cataract. Two of these for which we collected data are age and cigarette smoking. In all our statistical models, age was the most consistent predictor variable, and in the logistic regression model for nuclear cataract, known to be caused by smoking, the highest smoking category (>9 pack-years) had the highest OR (adjusted OR, 2.55; 95% CI, 0.81 to 8.00). These results provide confidence in the scoring procedures that were used. Any outcome misclassification is likely to be nondifferential by stove type and, therefore, to bias results toward the null.

Exposure data were obtained by questionnaire, and the cataract graders were not aware of participants’ exposure status. However, theoretically at least, some potential for information bias may have arisen from the manner in which the participants’ photographs were labeled. For the original gradings, the photographs were labeled with study ID and age. The age information potentially could have influenced the graders to give higher LOCS III scores for older participants, and, ideally, participants’ ages should have been masked. For the regrading of nuclear opacity, the age information was deleted. For all the photograph sets, the photographs for the right eye were immediately followed by the photographs for the left eye from the same person. Knowing the LOCS III scores assigned to the right eye might have influenced the scores assigned to the left eye. Any such bias will, to at least some extent, have been corrected in the multivariate risk models by use of the robust option.

Participants in this study were not aware of their own lens opacity status, so their responses could not have been influenced by this knowledge. Diabetes is an important risk factor for cataract. We asked participants about previously diagnosed diabetes but did not conduct objective clinical tests to rule out this condition. Although no participants reported this diagnosis, the possibility remains that some may have had undiagnosed diabetes. Because any such misclassification is likely to be unrelated to fuel use status, this should have no affect on results or, at worst, tend to bias results toward the null.

There may be some degree of nondifferential exposure misclassification, which is likely to affect some variables more than others. The questions asked, however, were about common exposures encountered by participants on a daily basis that would generally change infrequently. In particular, there is no reason to believe that the participants would have misidentified the types of cookstoves or fuels they used. In that regard, we verified the high level of accuracy of reporting of two key exposure variables—fuel type and ventilation in the kitchen—in the homes of 28 participants in another study conducted at the same time and in the same area and for which the exposure questions were the same.34 Thus, exposure misclassification is likely to be minimal, at least for fuel type. One possible limitation, however, is that we asked about only the main cooking fuel used. This might have led to some misclassification of exposure status because people may also use a secondary fuel. This would tend to bias results toward the null. Similarly, we did not collect information on whether participants, when they were children, accompanied their mothers while they cooked. Not taking this into account would have caused us to underestimate exposures. Because such misclassification is likely to have been nondifferential by present stove/fuel type, it would also tend to bias estimates toward the null.

The third main area of potential bias is confounding. In one large case-control study, which suggested that indoor cooking fuel smoke was associated with cataract, authors reported that it was likely that the association found between biomass fuel use and cataract formation was confounded by factors related to socioeconomic status.11 The present study collected a comprehensive range of data on potential confounding factors, particularly those associated with socioeconomic status. In the bivariate analysis (Table 1), the mean opacity scores did not vary by family income, generally regarded as a good indicator of socioeconomic status. The mean opacity score, however, varied by level of education and land ownership. We adjusted for level of education in all models, but this did not eliminate the associations found with nuclear opacity and nuclear color. We cannot rule out, however, the possibility of an unknown confounding factor causing the associations found.

Except for the study of Mohan et al.,11 other studies on indoor air pollution and cataracts did not examine cataracts by type. In that study, use of biomass fuels was found to be associated with increased risk of nuclear and cortical cataracts (as well as mixed cataracts—not directly evaluated in our study) but not with posterior subcapsular cataract. This is, to some extent, consistent with our finding that the strongest evidence was for an association with nuclear opacity.

Providing support for a relationship between use of biomass cookstoves and nuclear cataract, we found an exposure-response trend (p for trend 0.01) for increasing years of use (Table 4).

A causal relationship between exposure to smoke from biomass fuel combustion and cataract is biologically plausible. There is evidence that smoke from tobacco and biomass fuel combustion can induce oxidative stress by depleting antioxidants, such as plasma ascorbate, carotenoids, superoxide dismutase, and glutathione.24,59–61 These antioxidants protect against cataract formation.11 In addition, studies have indicated that tobacco and biomass fuel smoke condensate enhance the formation of superoxide radicals, which decrease antioxidants and cause lens discoloration and cataract.24,62–65 Both biomass and kerosene fuel combustions emit naphthalene,66–69 which has been shown to cause cataract in laboratory animal studies.70–72 In the previously mentioned indoor air pollution study conducted in the 28 households in the study area, naphthalene concentrations were found to be higher from biomass and kerosene stoves than from gas stoves (results not shown here).73 It is unclear whether direct entry of the combustion products into the eyes or inhalation of smoke and/or heat from the stove leads to the pathogenic process of cataract formation. Some studies have indicated that exposure to high ambient temperatures damages lenses.74–76

In conclusion, this study provides support for previous evidence that use of biomass cookstoves is associated with an increased risk of cataract, specifically nuclear opacification. It also provides a suggestion that kerosene use for cooking may be a risk factor for nuclear cataract and nuclear color. Although we had only five participants who reported using kerosene stoves (mean duration of use, 7.5 years), four (80%) of them had nuclear cataract compared with 55% of the other participants (Table 2). Kerosene fuel is widely used for cooking in low- and middle-income countries but, to the best of our knowledge, has not previously been investigated as a cataract risk factor. This study provides justification for such an investigation to take place. Kerosene is also widely used for lighting in the developing world and this also deserves investigation as a possible cataract risk factor. Bias, including potential confounding, does not provide an obvious explanation for the association with nuclear cataract, which is biologically plausible. Chance, however, cannot be excluded, and a larger population-based study, possibly with similar design, is needed to confirm these findings. Irrespective of the evidence for an association between biomass cookstoves and nuclear opacity, traditional biomass cookstoves produce substantial indoor and outdoor air pollution, including naphthalene and fine particulate matter. Therefore, replacement with cleaner alternatives, such as LPG, electricity, or advanced combustion biomass cookstoves, is justified. If the association with nuclear lens opacity is confirmed in larger studies, consideration might also be given to targeting cooks in biomass-using (and possibly kerosene-using) households for early screening of lens opacity as part of cataract prevention programs.

Kirk R. Smith

School of Public Health

747 University Hall

University of California Berkeley

Berkeley, CA 94720


Back to Top | Article Outline


The authors thank the management and staff of the Nepal Health Research Council, the Manipal Teaching Hospital, Manipal College of Medical Sciences, Pokhara, the former Dean of Manipal Teaching Hospital, Satish Kumar Dham, MD, and the Director of the Regional Tuberculosis Center, Pokhara, Sharat C Verma, MD, for their invaluable support during this study. Our special thanks go to the Topcon Medical Systems, Inc., for loaning the slit lamp with an integrated digital camera to conduct this study. This research was funded in part by the Fogarty International Training and Research in Environmental and Occupational Health Program (award 3D43-TW000815).

The authors declare that they have no competing financial interests.

Received June 30, 2011; accepted November 26, 2012.

Back to Top | Article Outline


An Appendix, Table A1 is available online at

Back to Top | Article Outline


1. World Health Organization (WHO). State of World’s Sight: VISION 2020: The Right to Sight: 1999–2005. Geneva, Switzerland: World Health Organization; 2005.
2. Nirmalan PK, Robin AL, Katz J, Tielsch JM, Thulasiraj RD, Krishnadas R, Ramakrishnan R. Risk factors for age related cataract in a rural population of southern India: the Aravind Comprehensive Eye Study. Br J Ophthalmol 2004; 88: 989–94.
3. Allen D, Vasavada A. Cataract and surgery for cataract. BMJ 2006; 333: 128–32.
4. Chatterjee A, Milton RC, Thyle S. Prevalence and aetiology of cataract in Punjab. Br J Ophthalmol 1982; 66: 35–42.
5. Nirmalan PK, Krishnadas R, Ramakrishnan R, Thulasiraj RD, Katz J, Tielsch JM, Robin AL. Lens opacities in a rural population of southern India: the Aravind Comprehensive Eye Study. Invest Ophthalmol Vis Sci 2003; 44: 4639–43.
6. Murthy GV, Gupta SK, Maraini G, Camparini M, Price GM, Dherani M, John N, Chakravarthy U, Fletcher AE. Prevalence of lens opacities in North India: the INDEYE feasibility study. Invest Ophthalmol Vis Sci 2007; 48: 88–95.
7. Brilliant LB, Pokhrel RP, Grasset NC, Lepkowski JM, Kolstad A, Hawks W, Pararajasegaram R, Brilliant GE, Gilbert S, Shrestha SR, et al.. Epidemiology of blindness in Nepal. Bull World Health Organ 1985; 63: 375–86.
8. Pokharel GP, Regmi G, Shrestha SK, Negrel AD, Ellwein LB. Prevalence of blindness and cataract surgery in Nepal. Br J Ophthalmol 1998; 82: 600–5.
9. Hockwin O. The causes and prevention of cataract blindness. Endeavour 1985; 9: 132–8.
10. Barrett JR. Focusing on vision through an environmental lens. Environ Health Perspect 2005; 113: A822–7.
11. Mohan M, Sperduto RD, Angra SK, Milton RC, Mathur RL, Underwood BA, Jaffery N, Pandya CB, Chhabra VK, Vajpayee RB, et al.. India-US case-control study of age-related cataracts. India-US Case-Control Study Group. Arch Ophthalmol 1989; 107: 670–6.
12. Mishra VK, Retherford RD, Smith KR. Biomass cooking fuels and prevalence of blindness in India. J Environ Med 1999; 1: 189–99. Available at: Accessed December 5, 2012.
13. Zodpey SP, Ughade SN. Exposure to cheaper cooking fuels and risk of age- related cataract in women. Indian J Occupat Environ Med 1999; 3: 159–61. Available at: Accessed December 5, 2012.
14. Pokhrel AK, Smith KR, Khalakdina A, Deuja A, Bates MN. Case-control study of indoor cooking smoke exposure and cataract in Nepal and India. Int J Epidemiol 2005; 34: 702–8.
15. Sreenivas V, Prabhakar AK, Badrinath SS, Fernandez T, Roy IS, Sharma T, Shah B. A rural population based case-control study of senile cataract in India. J Epidemiol 1999; 9: 327–36.
16. Haq I, Khan Z, Khalique N, Amir A, Jilani FA, Zaidi M. Prevalence of common ocular morbidities in adult population of Aligarh. Indian J Community Med 2009; 34: 195–201.
17. Saha A, Kulkarni PK, Shah A, Patel M, Saiyed HN. Ocular morbidity and fuel use: an experience from India. Occup Environ Med 2005; 62: 66–9.
18. McCarty CA, Nanjan MB, Taylor HR. Attributable risk estimates for cataract to prioritize medical and public health action. Invest Ophthalmol Vis Sci 2000; 41: 3720–5.
19. Christen WG, Manson JE, Seddon JM, Glynn RJ, Buring JE, Rosner B, Hennekens CH. A prospective study of cigarette smoking and risk of cataract in men. JAMA 1992; 268: 989–93.
20. Hankinson SE, Willett WC, Colditz GA, Seddon JM, Rosner B, Speizer FE, Stampfer MJ. A prospective study of cigarette smoking and risk of cataract surgery in women. JAMA 1992; 268: 994–8.
21. Klein BE, Klein R, Linton KL, Franke T. Cigarette smoking and lens opacities: the Beaver Dam Eye Study. Am J Prev Med 1993; 9: 27–30.
22. Solberg Y, Rosner M, Belkin M. The association between cigarette smoking and ocular diseases. Surv Ophthalmol 1998; 42: 535–47.
23. Cheng AC, Pang CP, Leung AT, Chua JK, Fan DS, Lam DS. The association between cigarette smoking and ocular diseases. Hong Kong Med J 2000; 6: 195–202.
24. Shalini VK, Luthra M, Srinivas L, Rao SH, Basti S, Reddy M, Balasubramanian D. Oxidative damage to the eye lens caused by cigarette smoke and fuel smoke condensates. Indian J Biochem Biophys 1994; 31: 261–6.
25. Smith KR. Biofuels, Air Pollution and Health: A Global Review. New York, NY: Plenum Press; 1987.
26. Klein BE, Klein R, Wang Q, Moss SE. Older-onset diabetes and lens opacities. The Beaver Dam Eye Study. Ophthalmic Epidemiol 1995; 2: 49–55.
27. McCarty CA, Mukesh BN, Fu CL, Taylor HR. The epidemiology of cataract in Australia. Am J Ophthalmol 1999; 128: 446–65.
28. Rowe NG, Mitchell PG, Cumming RG, Wans JJ. Diabetes, fasting blood glucose and age-related cataract: the Blue Mountains Eye Study. Ophthalmic Epidemiol 2000; 7: 103–14.
29. Bailey IL, Lovie JE. New design principles for visual acuity letter charts. Am J Optom Physiol Opt 1976; 53: 740–5.
30. Chylack LT Jr., Wolfe JK, Singer DM, Leske MC, Bullimore MA, Bailey IL, Friend J, McCarthy D, Wu SY. The Lens Opacities Classification System III. The Longitudinal Study of Cataract Study Group. Arch Ophthalmol 1993; 111: 831–6.
31. Bland JM, Altman DG. Comparing methods of measurement: why plotting difference against standard method is misleading. Lancet 1995; 346: 1085–7.
32. Leske MC, Connell AMS, Wu S-Y, Hyman L, Schachat Athe Barbados Eye Study Group. Prevalence of lens opacities in the Barbados Eye Study. Arch Ophthalmol 1997; 115: 105–11.
33. Tsai SY, Hsu WM, Cheng CY, Liu JH, Chou P. Epidemiologic study of age-related cataracts among an elderly Chinese population in Shih-Pai, Taiwan. Ophthalmology 2003; 110: 1089–95.
34. Pokhrel AK, Bates MN, Verma SC, Joshi HS, Sreeramareddy CT, Smith KR. Tuberculosis and indoor biomass and kerosene use in Nepal: a case-control study. Environ Health Perspect 2010; 118: 558–64.
35. Neale RE, Purdie JL, Hirst LW, Green AC. Sun exposure as a risk factor for nuclear cataract. Epidemiology 2003; 14: 707–12.
36. Young RW. Sunlight and age-related eye disease. J Natl Med Assoc 1992; 84: 353–8.
37. West SK, Duncan DD, Munoz B, Rubin GS, Fried LP, Bandeen-Roche K, Schein OD. Sunlight exposure and risk of lens opacities in a population-based study: the Salisbury Eye Evaluation project. JAMA 1998; 280: 714–8.
38. West SK, Munoz B, Schein OD, Duncan DD, Rubin GS. Racial differences in lens opacities: the Salisbury Eye Evaluation (SEE) project. Am J Epidemiol 1998; 148: 1033–9.
39. Brilliant LB, Grasset NC, Pokhrel RP, Kolstad A, Lepkowski JM, Brilliant GE, Hawks WN, Pararajasegaram R. Associations among cataract prevalence, sunlight hours, and altitude in the Himalayas. Am J Epidemiol 1983; 118: 250–64.
40. Hiller R, Giacometti L, Yuen K. Sunlight and cataract: an epidemiologic investigation. Am J Epidemiol 1977; 105: 450–9.
41. Hiller R, Sperduto RD, Ederer F. Epidemiologic associations with nuclear, cortical, and posterior subcapsular cataracts. Am J Epidemiol 1986; 124: 916–25.
42. Taylor HR, West SK, Rosenthal FS, Munoz B, Newland HS, Abbey H, Emmett EA. Effect of ultraviolet radiation on cataract formation. N Engl J Med 1988; 319: 1429–33.
43. Bochow TW, West SK, Azar A, Munoz B, Sommer A, Taylor HR. Ultraviolet light exposure and risk of posterior subcapsular cataracts. Arch Ophthalmol 1989; 107: 369–72.
44. Cruickshanks KJ, Klein BE, Klein R. Ultraviolet light exposure and lens opacities: the Beaver Dam Eye Study. Am J Public Health 1992; 82: 1658–62.
45. Italian-American Cataract Study Group. Risk factors for age-related cortical, nuclear, and posterior subcapsular cataracts. Am J Epidemiol 1991; 133: 541–53.
46. Leske MC, Chylack LT Jr., Wu SY. The Lens Opacities Case-Control Study. Risk factors for cataract. Arch Ophthalmol 1991; 109: 244–51.
47. Foster PJ, Wong TY, Machin D, Johnson GJ, Seah SK. Risk factors for nuclear, cortical and posterior subcapsular cataracts in the Chinese population of Singapore: the Tanjong Pagar Survey. Br J Ophthalmol 2003; 87: 1112–20.
48. Klein BE, Klein R, Lee KE, Meuer SM. Socioeconomic and lifestyle factors and the 10-year incidence of age-related cataracts. Am J Ophthalmol 2003; 136: 506–12.
49. Ritter LL, Klein BE, Klein R, Mares-Perlman JA. Alcohol use and lens opacities in the Beaver Dam Eye Study. Arch Ophthalmol 1993; 111: 113–7.
50. Munoz B, Tajchman U, Bochow T, West S. Alcohol use and risk of posterior subcapsular opacities. Arch Ophthalmol 1993; 111: 110–2.
51. Manson JE, Christen WG, Seddon JM, Glynn RJ, Hennekens CH. A prospective study of alcohol consumption and risk of cataract. Am J Prev Med 1994; 10: 156–61.
52. Ray WA, O’Day DM. Statistical analysis of multi-eye data in ophthalmic research. Invest Ophthalmol Vis Sci 1985; 26: 1186–8.
53. Hamilton LC. Statistics With Stata:Updated for Version 9 [computer program]. Belmont, CA: Thomson Brooks/Cole; 2006.
54. Rogers WH. Regression standard errors in clustered samples. Stata Tech Bull 1993; 13: 19–23.
55. Xu L, Li J, Cui T, Hu A, Fan G, Zhang R, Yang H, Sun B, Jonas JB. Refractive error in urban and rural adult Chinese in Beijing. Ophthalmology 2005; 112: 1676–83.
56. Hashemi H, Hatef E, Fotouhi A, Mohammad K. Astigmatism and its determinants in the Tehran population: the Tehran eye study. Ophthalmic Epidemiol 2005; 12: 373–81.
57. Bourne RR, Dineen BP, Ali SM, Noorul Huq DM, Johnson GJ. Prevalence of refractive error in Bangladeshi adults: results of the National Blindness and Low Vision Survey of Bangladesh. Ophthalmology 2004; 111: 1150–60.
58. Rosales P, Marcos S. Pentacam Scheimpflug quantitative imaging of the crystalline lens and intraocular lens. J Refract Surg 2009; 25: 421–8.
59. Padhy PK, Padhi BK. Effects of biomass combustion smoke on hematological and antioxidant profile among children (8–13 years) in India. Inhal Toxicol 2009; 21: 705–11.
60. Wegener A, Kaegler M, Stinn W. Frequency and nature of spontaneous age-related eye lesions observed in a 2-year inhalation toxicity study in rats. Ophthalmic Res 2002; 34: 281–7.
61. Garner MH, Garner WH, Spector A. H2O2 modification of Na, K-ATPase. Alterations in external Na+ and K+ stimulation of K+ influx. Invest Ophthalmol Vis Sci 1986; 27: 103–7.
62. Ramakrishnan S, Sulochana KN, Selvaraj T, Abdul Rahim A, Lakshmi M, Arunagiri K. Smoking of beedies and cataract: cadmium and vitamin C in the lens and blood. Br J Ophthalmol 1995; 79: 202–6.
63. Mussalo-Rauhamaa H, Leppanen A, Salmela SS, Pyysalo H. Cigarettes as a source of some trace and heavy metals and pesticides in man. Arch Environ Health 1986; 41: 49–55.
64. Avunduk AM, Yardimci S, Avunduk MC, Kurnaz L, Kockar MC. Determinations of some trace and heavy metals in rat lenses after tobacco smoke exposure and their relationships to lens injury. Exp Eye Res 1997; 65: 417–23.
65. Rao CM, Qin C, Robison WG Jr., Zigler JS Jr. Effect of smoke condensate on the physiological integrity and morphology of organ cultured rat lenses. Curr Eye Res 1995; 14: 295–301.
66. Cooke WM, Allen JM, Hall RE. Characterization of emissions from residential wood combustion sources. In: Cooper JA, Malek D, eds. Proceedings of the 1981 International Conference on Residential Solid Fuels, Environmental Impacts and Solutions, Portland, Oregon, June 1–4, 1981. Beaverton, OR: Oregon Graduate Center; 1981: 139–63.
67. Sannborn CR, Blanchet MA. Particulate emissions from residential wood combustion in Vermont. In: Cooper JA, Malek D, eds. Proceedings of the 1981 International Conference on Residential Solid Fuels, Environmental Impacts and Solutions, Portland, Oregon, June 1–4, 1981. Beaverton, OR: Oregon Graduate Center; 1981: 188–98.
68. Purvis CR, McCrillis RC, Kariher PH. Fine particulate matter (PM) and organic speciation of fireplace emissions. Environ Sci Technol 2000; 34: 1653–8.
69. Larson TV, Koenig JQ. Wood smoke: emissions and noncancer respiratory effects. Annu Rev Public Health 1994; 15: 133–56.
70. van Heyningen R, Pirie A. Naphthalene cataract in pigmented and albino rabbits. Exp Eye Res 1976; 22: 393–4.
71. Ghetti G, Mariani L. [Ocular changes caused by naphthalene: clinical and experimental studies]. Med Lav 1956; 47: 533–8.
72. van Heyningen R. Naphthalene cataract in rats and rabbits: a resume. Exp Eye Res 1979; 28: 435–9.
73. Pokhrel AK. Preclinical Measures of Eye Damage (Lens Opacity), Case-Control Study of Tuberculosis, and Indicators of Indoor Air Pollution from Biomass Smoke. PhD thesis. University of California, Berkeley, Berkeley, California; 2010.
74. Sasaki H, Jonasson F, Shui YB, Kojima M, Ono M, Katoh N, Cheng HM, Takahashi N, Sasaki K. High prevalence of nuclear cataract in the population of tropical and subtropical areas. Dev Ophthalmol 2002; 35: 60–9.
75. Sharon N, Bar-Yoseph PZ, Bormusov E, Dovrat A. Simulation of heat exposure and damage to the eye lens in a neighborhood bakery. Exp Eye Res 2008; 87: 49–55.
76. Kojima M, Okuno T, Miyakoshi M, Sasaki K, Takahashi N. Environmental temperature and cataract progression in experimental rat cataract models. Dev Ophthalmol 2002; 35: 125–34.

cookstove; heating stove; cross-sectional study; indoor air pollution; kerosene; LOCS III scale

Supplemental Digital Content

Back to Top | Article Outline
© 2013 American Academy of Optometry