Pupil size change in agricultural workers exposed to pesticides

ABSTRACT

Clinical relevance

Pupil size evaluation using clinical examination may be important for detecting and monitoring individuals at risk of neurotoxic effects from chemical exposure, as it may enable early intervention and the implementation of preventive measures.

Background

This work aimed to investigate the association between pesticide exposure and pupil size. Pupil size is regulated by muscarinic and nicotinic receptors, and it is well-established that common pesticide chemicals disrupt this regulation.

Methods

Twenty agricultural workers exposed to pesticides, and twenty participants not exposed, underwent visual screening, and pupil size evaluation under mesopic and photopic conditions. Additionally, signs of neurotoxicity and pesticide exposure in both groups were evaluated using the modified version of the neurotoxic symptoms questionnaire (Q16) and measuring cholinesterase (AChE) levels in blood, respectively.

Results

Agricultural workers exposed to pesticides had a score indicating medium–high level of neurotoxicity (49.85 (SD ± 8.94)) which was significantly higher (t (36) = 7.659, p ≤ 0.0001) than non-exposed participants who had low levels of neurotoxicity (27.25 SD ± 8.86). There was a significant difference in pupil size (mm) under mesopic (t (19) 4.42 p = 0.003) and scotopic (t (19) 4.63, p = 0.0002) conditions between the two groups. Additionally, there was a significant difference in AChE blood levels (t (19) 2.94 p = 0.008) between exposed and non-exposed participants, indicating that exposed workers had low levels of this enzyme (average exposed group 3381 U/L (SD ± 1306)) compared to the non-exposed group (average non-exposed group 4765 U/L (SD ± 1300)). A significant negative correlation between AChE levels, years of exposure, and pupil size was found. The latter finding importantly showed that smaller pupils are associated with the accumulation of acetylcholine or a decrease in the activity of the enzyme AChE.

Conclusion

Pupil size of agricultural workers exposed to pesticides can be abnormal and is associated with neurotoxicity as indicated by symptomatology and cholinesterase levels. Evaluation of pupil size may be useful for clinically detecting neurotoxicity.

KEYWORDS:

• Cholinesterase
• neurotoxicity
• organophosphates
• pesticides
• pupil size

Introduction

Pesticides are chemical substances used to prevent, destroy, or control plant and animal pests. They can be classified according to their chemical family (organochlorines, organophosphates, carbamates, pyrethroids, dipyridyl compounds, and inorganic salts).¹ Pesticides can cause toxicity and harm in humans if the level of exposure is sufficiently high.^2,3 It has been estimated that annually there are 385 million cases of unintentional, acute pesticide poisoning worldwide, including around 11,000 fatalities.⁴

Agricultural populations exposed to pesticides have been reported to suffer various health problems, mainly neurological disorders, and respiratory, reproductive, endocrinological, and dermal problems.^5–7 For example, organophosphorus compounds inhibit enzymes like acetylcholinesterase (AChE) and neuropathy target esterase (NTE)^8,9; this results in the accumulation of acetylcholine (ACh) and phosphatidylcholine, respectively which may lead to acute or intermediate poisoning due to the cholinergic effect and organophosphate-induced delayed polyneuropathy (OPIDP).^9,10

The main manifestations of acute intoxication are excessive overstimulation of muscarinic and nicotinic cholinergic receptors, which can lead to changes in consciousness, muscle weakness, and excessive secretory activity.^11,12 In addition, organophosphate and carbamate exposure is associated with neurodegenerative diseases.^13–15

Given the wide range of pesticide exposure on physiological and cognitive functions,^16,17 one possible sensitive diagnostic test to quantify the signs and symptoms of neurotoxicity, is the evaluation of vision and visual function, which previous studies have shown to be highly susceptible to neurotoxic damage.^18–20 A candidate measure is the pupil size,^21–23 which has experimentally shown reversible changes due to exposure to organophosphorus.^24–26 Importantly, changes in pupil size have occurred without other overt signs of toxicity and suggest that miosis may be an early sign of pesticide exposure.^22,27

Pupillary miosis is the function of the sphincter of the pupil, which is innervated by parasympathetic fibres. At the same time, mydriasis is the function of the dilator pupillae and is controlled by sympathetic fibres.^22,28 In this way, the pupil size reflects the function of the autonomic nervous system, which regulates the antagonistic actions of the iris sphincter and the dilator muscles of the eye and controls the amount of light entering the eye.²⁹

Synaptic transmission in the neural pathways regulating pupil size involves AChE, which, as mentioned, is inhibited by pesticide-related organophosphates.⁸ Organophosphates in the eye can cause significant AChE inhibition and muscarinic receptor downregulation which may lead to abnormal presentations of pupil size, particularly miosis.^27,30 Indeed, most visual and ocular signs and symptoms of acute organophosphate poisoning are reflected in different types of muscarinic manifestations (generally the parasympathetic system), such as pupillary contraction (miosis), stimulation of the different endings of the lacrimal gland and blurred vision (due to accommodation).³⁰ The miotic effect may be reduced after repeated exposure to organophosphates because of the development of tolerance to cholinergic agonists and desensitisation of muscarinic AChE receptors on the pupillary sphincter.³¹

Accordingly, miosis may be a potential marker for pesticide agents in the body.^22,27 Given the strong action of the pesticide on the neuronal system that supports miosis, the assessment of pupil size in individuals exposed to pesticides may provide a sensitive and efficient indication of neurotoxicity, as its measurement is non-invasive fast and easy to perform.

The present study aimed to establish whether exposure to occupational levels of pesticides in agricultural workers is associated with altered pupil sizes, particularly signs of miosis. In addition, pupil size with neurotoxic symptoms (using a modified version of the Neurotoxic Symptoms Q16 questionnaire) was associated with the level of blood AChE activity which is a commonly used indicator of pesticide exposure.³²

Recent research (conducted on farm workers) has shown that regular exposure to pesticides can lead to neurotoxicity resulting in deficits in visual function, particularly in contrast and colour perception, and elevated inflammatory markers (Substance p) in tears.³³ This finding suggests that measuring changes in vision may be a useful way of quantifying the impact on the brain. The present study seeks to add to this understanding and by investigating the relationship between neurotoxicity and pupil size in workers exposed to occupational levels of pesticides.

Methods

This research followed the principles of the Declaration of Helsinki for medical research in human beings. Each participant was informed of the benefits and risks of the study and a written informed consent was signed. The study was approved by the ethics committee of the Faculty of Health Sciences of the University of La Salle (Memorandum 029 17,042,017).

An observational cross-sectional analytical study was conducted to quantify the pupil size in agricultural workers exposed to pesticides, their relationship with neurotoxic symptoms, and blood cholinesterase levels. These results were compared to non-agricultural workers not exposed to pesticides. There were 38 participants in the present study. The non-exposed group comprised 20 subjects (8 women and 12 men) with a mean age of 27 years (SD ± 7.05), and the exposed group was 18 male subjects with a mean age of 31 (SD ± 5.68).

Agricultural workers in the exposed group were selected from potato farms that used pesticides such as organophosphates (OPs), carbamates (CAs), pyrethroids (PIs): Bifenthrin (2-methylbiphenyl-3-ylmethyl(1RS,3RS)-3-[(Z)-2-chloro-3,3,3-trifluoroprop-1-enyl]-2,2-dimethylcyclopropanecarboxylate), Profenofos (O-4-bromo-2-chlorophenyl O-ethyl S-propyl phosphorothioate), cypermethrin ((RS)-a-cyano-3-phenoxybenzyl (1RS,3RS;1RS,3SR)-3-(2,2-dichloro vinyl)-2,2- dimethyl cyclopropane carboxylate) and Carbosulfan (2,3-dihydro-2,2-dimethylbenzofuran-7-yl (dibutylaminothio) methylcarbamate).

Potato farms in Colombia use between 1.0 and 2.5 kg/ha of pesticides annually. Over 4 weeks, the type and frequency of use of pesticides by applicators and operators collaborating on farms who participated in the present study were: OPs: 22 times; PIs: 16 times; CAs: 22 times. The population under study was located geographically in the municipality of Belén, Boyacá, aged between 20 and 40 years, and worked as farmers for at least one year. The average number of years of exposure was 12.6 (SD ± 5.10) and ranged from 4 to 22 years.

Participants of the same age group who were not exposed to occupational levels of pesticides comprised the non-exposed group. The sample size used in the present study was estimated with a power of 80% and a type I error of 5%, based on two previous studies and assuming an effect size of 0.61 (Glass’s Delta). This effect size was derived from previously reported^18,34 means and standard deviations of the neurotoxic symptoms questionnaire (Q16) scores for participants exposed and not exposed to pesticides.

The exclusion criteria for both groups were those who reported an alteration or were previously diagnosed with alterations in the pupillary response and neurobehavioral deficiencies, systemic diseases, macular diseases, and opacities in the cornea and/or lens. Another exclusion criterion for the non-exposed group was living with someone who worked or was exposed to pesticides. The inclusion criteria for the non-exposed group were individuals who had never worked or were exposed for long periods to any pesticide or suffered from any disease related to changes in pupil size.

The neurotoxic symptoms questionnaire Q16 is commonly used to monitor the early effects of neurotoxic exposures in the working population and has been internationally recognised as the gold standard test for the detection of neurotoxic symptoms by the World Health Organization.³⁵ This questionnaire has been modified and validated,³⁶ to be used in different studies for screening of neurotoxicity due to exposure to organic solvents and their impacts on vision.^18,33,34

The questionnaire contained 16 short questions about the symptoms commonly described by workers exposed to solvents, such as: ‘I have a short memory’, Sometimes I have a painful tingling sensation in some part of my body”, or ‘I feel that I have less sensitivity or a complete loss of sensation in some parts of the arms or legs’. These questions are evaluated using a 5-level Likert scale ranging from totally disagree (1) to agree (5). Once the subject completed the questionnaire, the total sum of the scores is used to indicate the level of neurotoxicity, which can be categorised as Low (16–32), Medium-low (33–48), Medium high (49–64) and high (65–80).

Instrumentation and procedures

An optometric examination includes a battery of tests such as visual acuity assessment, anterior segment examination, direct ophthalmoscopy (for retinal fundus assessment), refraction test, and evaluation of the pupils. All tests were assessed binocularly, under standardised conditions, and by the same examiner. Participants were first visually screened, including a comprehensive anamnesis and distance and near visual acuity assessment using the Good-Lite ETDRS and for near vision and near visual chart, respectively. After examination, participants who displayed abnormalities in vision were excluded from the study. All participants filled out a baseline survey that included questions related to alcohol consumption, smoking, medications, metabolic diseases, or neurological diseases.

Biomicroscopy was assessed using the Ezer Slit lamp®, while direct ophthalmoscopy and retinoscopy were assessed using the Welch Allyn® diagnostic kit. Participants who required refractive correction were corrected at that time with a frame trial and lenses, this frame was used during the examination to take the different visual tests, and at the end, the final optical prescription was given to each participant.

After the visual screening, pupil size was assessed with the Autorefractor keratometer (Huvitz Wavefront Ref/Keratometer HRK −7000, version 5.02.00A made in Korea). This system comprises various features that provide measurements, including the pupillary diameter. To measure the pupil diameter, the subject was asked to observe the target without trying to make any visual effort to see the image but rather to be completely relaxed, and three measurements for each eye were obtained to provide an estimate of the average pupil size of each eye. This procedure was performed under mesopic and scotopic targets (0.80 lux for mesopic and 0.08 lux for scotopic) as these were the measurement conditions of the device. The distance between the corneal apex of the subject and the equipment was 40 mm.

The level of AChE in blood was measured by taking a venous blood sample from participants at the same visit. This sample was centrifugated (1200 rpm x 10 min), and the serum was separated. The serum was transported to a laboratory in an ice chamber at a temperature between 6.8°C and 7.4°C, and the kinetic method was used at 405 nm to determine the amount of cholinesterase in the serum. The Wiener lab® brand kit and procedure Fixed ΔT was used. Reagent A was reconstituted with 3 ml of reagent B and pre-incubated for 1 min at 25°C, 20 µl of the sample was added and mixed and immediately the absorbance at 405 nm was recorded, and immediately the stopwatch was triggered, it was reread at 30, 60 and 90 s. The average difference was determined every 30 according to ΔA/30) by subtracting the reading from the previous one and averaging the values. This was repeated with each of the samples. The reference values established by the Wiener lab® laboratory for the kit used and the temperature used are 3200–9000 U/L.

Statistical analysis

The Shapiro–Wilk test was used to verify that the data was normally distributed. Depending on the type of data and comparison between groups or conditions, either a t-test or Mann–Whitney test was used. A two-way ANOVA with post hoc comparison (Tukey’s test, corrected for multiple comparisons) was also performed to establish whether pupil size was significantly different between exposed and non-exposed individuals for different lighting conditions. Correlation analysis between two variables was assessed using Spearman’s Rho. In all cases, a p < 0.05 value was considered statistically significant. The analysis was conducted with the Graph Pad software, version 9.

Results

Indicators of neurotoxicity and pesticide exposure

There was a statistically significant difference between non-exposed and exposed groups for AChE activity (t (36) 2.99, p = 0.008). In particular, the group of exposed workers had lower activity of AChE 3381 U/L SD ± 1306 U/L than the non-exposed group 4765 U/L SD ± 1300 U/L. Cohen’s d, was 1.06, indicating a moderate to large effect size. Modified Q16 neurotoxic symptoms questionnaire scores were also significantly different between the two groups (t (36) = 7.659, p ≤ 0.0001), with the exposed group reporting on average higher symptom scores of 49.85 SD ± 8.94 (which indicated medium-high neurotoxicity levels) than the non-exposed group 27.25 SD ± 8.86. The average duration of exposure was 12.6 years (SD ± 5.10). Cohen’s d, was 2.53, indicating a large effect size.

There was no correlation between the modified neurotoxic symptoms questionnaire Q16 score and years of exposure (rho = 0.10 (p = 0.65)) for the exposed group (see Table 1). However, modified Q16 neurotoxic symptoms questionnaire scores were negatively correlated with AChE activity in bloodrho = −0.56 (p = 0.0009) for the exposed group but not for the non-exposed group p = 0.39. This result suggests that higher neurotoxicity symptoms may be associated with lower AChE activity (see discussion) in individuals exposed to pesticides.^8,32

Table 1. Correlation matrix between pupil size and neurotoxicity measures for the exposed group.

	Neurotoxicity Symptoms	p value*	Pupil Size Mesopic	p value*	Pupil Size Scotopic	p value*
Years of Exposure	0.10	0.65	−0.477	0.034	−0.613	0.004
Neurotoxicity Symptoms (Modified Q16)			−0.100	0.650	−0.110	0.610
Cholinesterase Levels	−0.56	0.0009	0.589	0.010	0.504	0.033

*Spearman Correlation.

Pupil size measurements

The average pupil size for the exposed group for mesopic and scotopic light levels was smaller than the non-exposed group (Figure 1). The Shapiro–Wilk normality test showed that all pupil size data were normally distributed. A two-way ANOVA was conducted with Lighting Conditions (mesopic and scotopic) and Group (non-exposed and exposed) as factors. This analysis showed a significant effect of Lighting Condition (F (3, 57) = 43.50, p = 0.0136, η2 = 0.69) with pupil size on average larger under scotopic than mesopic conditions. A main effect of Group was also observed, with pupil sizes on average smaller for the exposed group than the non-exposed group (F (19, 55) = 2.152, p = <0.0001, η2 = 0.42). There was no significant interaction effect.

Figure 1. Average pupil size differences for the exposed and non-exposed group for both mesopic and scotopic light levels. Tukey’s post hoc comparison tests ***p < 0.002.

Tukey’s post hoc comparison tests showed that pupil size for the exposed group was significantly smaller than the non-exposed group for Mesopic (p < 0.002) and Scotopic (p < 0.003) conditions (Figure 1). Years of pesticide exposure were negatively correlated with pupil size under mesopic and scotopic conditions (p < 0.05).

Significant correlations were observed between AChE activity and pupil size for mesopic (rho = 0.589, p = 0.010) and scotopic conditions (rho = 0.504, p = 0.033), such that lower AChE activity is associated with smaller pupil sizes. There was no significant correlation between modified Q16 scores and pupil size regardless of the lighting condition rho = −0.100 (p = 0.650) mesopic and rho = −0.110 (p = 0.610) scotopic (Table 1).

Discussion

The present study quantified the relationship between pesticide exposure and pupil size and associated pupil size with neurotoxic symptoms (as measured by the modified Q16) and AChE in blood. Exposure to pesticides was associated with significant alterations in pupil size, which was significantly smaller (in both mesopic and scotopic conditions) for the exposed group compared to the non-exposed group.

The exposed group, on average, reported significantly more neurotoxic symptoms and lower levels of AChE in blood, indicating overstimulation of the nervous system that manifests in a spectrum of neurotoxic symptoms, several of which are assessed using the modified Q16 questionnaire. These symptoms associated with AChE inhibition encompass involuntary muscle twitching and weakness (Questions 11–15), cognitive impairment, including confusion or other neurological deficits (Questions 1–8), and in severe cases, organ failure, particularly respiratory failure (Question 10).

Pesticides inhibit AChE levels which can impact nicotinic and muscarinic processes, including pupil size.¹¹ As reported in the present study, the exposed group had smaller pupil size, and this may occur due to the accumulation of ACh at nerve synapses due to the inhibition of the AChE enzyme. The excessive amounts of ACh produced result in symptoms or signs of cholinergic poisoning.^9,11 Previous studies on the effects of these chemical compounds on the eye suggest that the inhibition of AChE is a direct result of the vapours on ocular tissues rather than systemic absorption and distribution of the pesticide.^19,24,31

The localised increase in ACh leads to the contraction of the pupillary sphincter muscle, resulting in dose-dependent miosis.^24,27 In patients intoxicated with Sarin vapours (one of the most toxic organophosphorus compounds), after 24 h of exposure, the acetylcholinesterase activity was significantly lower in patients with miotic pupils than those with normal-sized pupils.³⁷ In summary, the findings of a negative medium association between AChE levels and pupil size indicate that miosis may be a sensitive index of exposure to pesticides at lower levels that can lead to other identifiable effects or poisoning. However, future research is needed to further establish and quantify this relationship.

Years of pesticide exposure were negatively correlated with pupil size under mesopic and scotopic conditions. These results are consistent with signs of acute poisoning from exposure to organophosphates, previously described as a direct consequence of the inhibition of AChE activity. Whilst AChE levels are correlated with pupil size, modified Q16 symptom scores were not (Table 1). This finding may be because the modified neurotoxic symptoms Q16 questionnaire registers more general symptoms of neurotoxicity that are not necessarily related to vision or visual impairment.

Perhaps another contributing factor is that it is well-known that significant individual differences exist in how people subjectively report their perception of pain and discomfort. Accordingly, the subjective nature and variability of individual reports of neurotoxic symptoms will likely affect the correlation strength with a specific outcome measure. Note, however, that the modified Q16 was correlated with AChE, which suggests that AChE is related to general neurotoxicity symptoms.

It is important to highlight that the exposed group in this study exclusively consisted of males, in contrast to the non-exposed group, which included 12 males and 8 females. Consequently, gender differences, along with gender-related factors such as body mass index (BMI) may be contributing factors to pupil size. Although previous studies have demonstrated the independence of pupil size from gender,^38,39 a further analysis (t-test) was performed to determine whether the pupil size and levels of neurotoxicity of female control participants differed from their male control counterparts. This analysis revealed no statistical difference in pupil size under both mesopic (p = 0.21) and scotopic conditions (p = 0.45), AChE activity (p = 0.49) as well and Q16 neurotoxicity levels (p = 0.32), suggesting that gender may not be a contributing factor in accounting for the present findings.

It is important to consider age-related changes in pupil size, which tend to be more pronounced in significantly older individuals, with pupil size diminishing by 0.6 to 1 mm every 20 years.^38,39 In the current study, both groups were approximately age-matched, with the exposed group averaging 31 years and the non-exposed group 27 years. The modest 4-year difference (estimated to be 0.12–0.2 mm) is anticipated to have minimal impact on the findings of this study, given the gradual rate of pupil size change associated with age.

This study demonstrates that pupil size may be used as an important biomarker or indicator for pesticide exposure; specifically, pupils size measurement can be used to detect and monitor the presence of a pesticide in the body and serve as a sign of adverse health effects caused by this exposure. This highlights the utility of pupil size measures in the broader context of other clinical tests and the medical history of the patient. As such, pupil size measurement could potentially guide clinicians towards improved diagnosis and better case management.^22,27

The use of cost-effective and easy-to-measure biomarkers like these might help evaluate potential pesticide exposures and allow for decisions to be made about the regulation of the use of pesticides to protect human health and the prevention of pesticide toxicity. Pupil size measures can be used in conjunction with other measures of neurotoxicity (e.g., modified neurotoxic symptoms Q16) to contribute to a more comprehensive assessment of the consequences of pesticide exposure and used as a means of monitoring recovery.

Acknowledgements

This work was supported by Universidad de La Salle, Bogotá, Colombia, in collaboration with The University of New South Wales. We express our sincere gratitude to the potato and strawberry farmworkers from Belén Boyacá, Colombia. We also acknowledge the research group named Critical Zone from the La Salle Environmental Engineering program for providing information regarding the pesticides used in the geographical area of the study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The work was supported by the Universidad de La Salle [19072017].