The relationship between serum trace element changes and visual function in heavy smokers


Dr Ömer Akyol
Inonu University
Faculty of Medicine
Department of Biochemistry
Turgut Ozal Medical Centre
44069 Malatya
Tel: + 90 422 341 0660
Fax: + 90 422 341 0728


Purpose:  The aim of this prospective study was to evaluate serum manganese (Mn), zinc (Zn) and copper (Cu) levels and visual functions including visual acuity, colour vision, pattern visual evoked potentials (PVEPs), and contrast sensitivity in heavy smokers and to compare these with the equivalent levels and functions in non-smokers.

Methods:  Data were recorded in 24 healthy, chronic and heavy cigarette smokers and 16 healthy, non-smoking control subjects. Serum Zn, Cu and Mn concentrations in all subjects were measured by atomic absorption spectrophotometry.

Results:  Both study and control subjects had normal visual acuity and colour vision. Pattern visual evoked potentials were normal in all study and control subjects. Contrast sensitivity was significantly reduced in heavy smokers compared to non-smokers (p < 0.023), despite the fact that central vision and PVEP responses were not affected. Mean serum Mn and Zn levels were significantly lower in smokers than in non-smokers (p < 0.0001 and p < 0.005, respectively).

Conclusions:  Reduced contrast sensitivity values associated with low levels of serum Mn and Zn, which function as cofactors of superoxide dismutase in erythrocyte and other nucleated cells, suggest a possible role of trace elements in smoking-induced early retinal toxicity.


Smoking is a significant risk factor in several debilitating and fatal diseases including coronary artery disease, myocardial infarctions, cerebrovascular accidents and sudden death. Among the eye diseases in which smoking is known to be implicated is a bilateral optic neuropathy, namely, tobacco-toxic optic neuropathy (TTON) (Dreyfus 1977; Samples & Younge 1981; Rizzo & Lessell 1993). In addition, smoking is still a major controllable risk factor for thyroid ophthalmopathy (Hamilton 1999), cataract (West et al. 1989; Leske et al. 1991; Cekic 1998; Ojofeitimi et al. 1999), strabismus (when the mother smokes throughout pregnancy) (Hakim & Tielsch 1992), and colour vision defects (Erb et al. 1999). Tsao et al. (1999) recently showed that smoking may be an important risk factor in some pedigrees with Leber's hereditary optic neuropathy.

Tobacco-toxic optic neuropathy is a chronic and symptomatic toxicopathy caused by longterm, heavy smoking (Wang et al. 1992). Blood serum zinc (Zn) levels have been found to be lower in patients with TTON than in normal subjects, while levels of other elements such as iron (Fe), copper (Cu), manganese (Mn) and chromium (Cr) have been found to be lower in some subjects only (Wang et al. 1992).

The present study aimed to examine the influence of smoking on some serum trace element levels and on contrast sensitivity, and to study the relationship between levels of trace elements and visual function in non-symptomatic, healthy, chronic cigarette smokers.

Material and Methods

Study and control subjects were selected from the outpatient clinic (Turgut Ozal Medical Center, Department of Ophthalmology) on a volunteer basis. Informed consent was obtained from all subjects.

The study group comprised 24 chronic, heavy (Cekic 1998) smokers (20 males, four females), who had smoked at least 20 cigarettes per day for the last 10 years. The control group comprised 16 healthy, non-smoking subjects (14 males, two females). A dietary and smoking questionnaire was administered (see appendix). Thus, the subjects in the study group reported a history of heavy smoking. The standard pack year method was used to measure the incidence and degree of smoking. According to pack-year method, you have a 10 pack-year history, if you have smoked one pack a day for ten years, two packs a day for 5 years, etc. In both groups, only the subjects with small refractive errors (− 1 to + 1 D spherical equivalent) and normal visual acuity (VA) according to Snellen chart tests were eligible for the study. Volunteers were excluded if they had large refractive errors, were under local or systemic medication, had been exposed to any kind of toxic substances, had a history of eye or systemic disease, or a history of alcohol consumption. Subjects who reported any cessation of smoking in the preceding 10 years were also excluded from the study. We deliberately chose young, healthy, heavy smokers who maintained an adequate diet and had no history of alcohol consumption in order to rule out the effects of alcoholism and malnutrition (Victor & Dreyfus 1965).

A general physical examination and analysis of clinical laboratory data disclosed no evidence of systemic disease in any of the study or control subjects. Gamma glutamyl transferase levels were not available for objective assessment of any changes in liver metabolism. All subjects also underwent a complete ophthalmological examination, including indirect ophthalmoscopy of the retina and optic disc. No pathological findings such as media opacities, precataractous lens changes or accelerated lens yellowing were found in either study or control subjects.

Contrast sensitivity testing was then carried out on both eyes of each subject, using a Humprey automatic refractor/keratometer (Model 599) (San Leandro, CA, USA). Contrast sensitivity testing with the Humprey automated refractor/keratometer uses a standard Snellen VA chart, but with a lower contrast between the letters and background. Once each subject's refractive error had been subjectively refined, all subjects with normal VA (≥ 20/20) were required to read the letters again in the low contrast mode. The VA result at the lowest contrast the patient was able to reach was then divided by the VA result at normal contrast and expressed as the contrast sensitivity value. Pattern visual evoked potentials (PVEPs) testing was performed with a Nicolet Compact Four electrodiagnostic system (Madison, WI, USA). The reversing checkerboard pattern covered a field of 8 × 10 degrees. The average luminance on the screen was 50 cd/m2 and the contrast was 80% for all recordings. Three spatial frequencies (60, 30 and 15 min of arc check) and a reversal rate of 1.9 times/second (1 Hz) were used. The bandwidth used was 1–100 Hz. The electrode was placed on Oz, Fz, A1 according to the EEG international 10–20 electrode systems. The input impedance was not more than 5 KΩ The averaged responses to 100 reversals were recorded. The latency of major positivity (P100) and amplitude (N75−P100) were recorded.

Blood samples were taken in the morning after a fasting period. Samples of 10 ml of blood were taken by venipuncture of an antecubital vein without using any anticoagulant. Each sample was transferred to a deionized glass tube and then centrifuged at 3000 g for 10 min. Centrifuged material was put into deionized tubes. Sera were digested in a nitric acid/perchloric acid mixture (1/5, v/v) by heating mildly until all perchloric acid fumes had disappeared (Kitapci et al. 1994). Inorganic sediment obtained was dissolved in a certain amount of tridistilled water (6mL) and the last solution was used for Zn, Cu and Mn analyses. Element analyses were performed by atomic absorption spectrophotometry (Shimadzu AA-6701F) (Kyoto, Japan) using Kirgbright's standard additional technique (Kirgbright 1980). Each measurement was performed three times and averages were used for analysis. Results were expressed in microgram per decilitre (µg/dl) of serum.

The distributions for each group were compared with one sample Kolmogrov−Smirnov test. Both groups showed normal distribution, thus parametric statistical methods were used to analyse the data. Student's t-test was performed for pairwise comparisons. Bivariate comparisons were examined using Pearson's rank correlation coefficients (r). Results are presented as means ± standard deviations. P-values < 0.05 are regarded as statistically significant.


The mean age of the smokers was 34.91 years (range 30–52 years). Mean smoking time was 17.83 years (range 10–38 years) and mean effective pack year was 1.25. The mean age of control subjects was 34.87 years (range 29–48 years). There were no significant age or sex differences between the groups (p > 0.05).

Both study and control subjects had normal VAs. The mean contrast sensitivity value of the smokers (0.77 ± 0.14) was significantly lower than that of the control subjects (0.89 ± 0.14) (p < 0.018). The mean serum Mn level was significantly lower in the smoking group (2.11 ± 1.03 µg/dl) than in the control group (4.01 ± 1.65 µg/dl) (p = 0.004). The mean serum Zn level was also significantly lower in the smoking group (141.5 ± 17.5 µg/dl) than in the control group (160.5 ± 23.6 µg/dl) (p < 0.005). However, Cu levels did not differ (p > 0.05) between the two groups. Test results are shown in Table 1 and Table 2.

Table 1.  Serum manganase (Mn), zinc (Zn) and copper (Cu) levels in healthy, heavy smokers and control subjects.
 Mn (µg/dl)Zn (µg/dl)Cu (µg/dl)
  1. ns = non-significant

Smokers (n = 24)2.11 ± 1.03141.5 ± 17.5191.1 ± 26.2
Controls (n = 16)4.01 ± 1.65160.5 ± 23.6183.1 ± 19.7
Table 2.  Contrast sensitivity and PVEP (P100 latency and N75−P100 amplitude) test results in healthy, heavy smokers and control subjects.
 ContrastPVEP P100PVEP N75-P100
 sensitivitylatency (ms)amplitude (µV)
  1. ns = non-significant

Smokers (n = 24)0.77 ± 0.1499.2 ± 3.97.58 ± 2.85
Controls (n = 16)0.89 ± 0.14101.1 ± 4.89.24 ± 3.96

In order to study the effects of the daily dose of tobacco on the observed visual changes, we divided the subjects into four subgroups consisting of those who smoked one packet/day, those who smoked 1.5 packets/day, those who smoked two packets/day, and those who smoked 2.5 packets/day, respectively. There were no statistically significant differences between the subgroups. However, because of the small number of subjects in each subgroup, these findings have little power.


Our study showed that healthy, chronic, heavy cigarette smokers had decreased contrast sensitivity although central vision was preserved. These findings are not consistent with those in clinically established TTON. Mild forms of a number of diseases affecting the optic nerve, including toxic optic neuropathy, may allow the patient to read the 20/20 line yet demonstrate a diminished contrast sensitivity function (Lorance et al. 1987). Contrast sensitivity testing is also an accurate method of following the progress of optic neuritis (Fleishman et al. 1987; Lorance et al. 1987) and has been reported to be effective in revealing subtle visual losses in optic neuritis (Optic Neuritis Study Group 1991). Others have proposed that contrast sensitivity testing might be effective in detecting subclinical toxic optic neuropathy (Salmon et al. 1987). On the other hand, visual contrast sensitivity was found to be significantly reduced in epileptic patients (Tomson et al. 1988). In that study, there was a significant negative correlation between the plasma concentration of carbamazepine and contrast sensitivity, indicating that the reduced contrast sensitivity was due to the drug therapy. The presence of higher gamma-glutamyl transpeptidase and mean corpuscular volume levels in patients who had visual contrast sensitivity abnormalities indicative of alcohol-tobacco amblyopia suggests that alcohol consumption is involved in the development of these abnormalities (Roquelaure et al. 1995). Moreover, Fine & Kobrick (1987) observed that habitual smoking is associated with alterations in visual contrast sensitivity in military volunteers. The results of the present study support this finding in that significantly decreased contrast sensitivity was observed in chronic, heavy smokers compared to controls. In the present study, the method of contrast sensitivity testing by Humprey automated refractor/keratometer used a low contrast version of a typical Snellen chart. Although control subjects demonstrated modest reductions in VA for low contrast letters, chronic, heavy smokers with no evident abnormalities showed profound reductions in VA at low contrast mode, despite showing good levels of VA for standard high contrast letters. Thus, it is reasonable to assume that this approach to contrast sensitivity testing is able to reveal early disturbances of the visual functions that are not revealed by standard VA testing. The advantage of a low contrast acuity chart in a clinical setting is that it is a test procedure highly familiar to most patients.

Trace elements exist in very low concentrations in the body. They play important functional roles in the peripheral and central nervous systems, although they are minor building components in the tissues. Zinc plays an important role in nucleic acid metabolism and is necessary for normal neuronal development (Ilhan et al. 1999). Severe Zn depletion causes impaired nervous system maturation in fetal and neonatal animals. Manganase and Zn together are essential for the activity of mitochondrial (Mn) and cytoplasmic (Cu, Zn) superoxide dismutase (SOD) izoenzymes. Both SOD izoenzymes carry the above-mentioned trace elements as prosthetic groups. Superoxide dismutases catalyse the conversion of toxic superoxide radicals to hydrogen peroxide and thus protect cellular and extracellular structures and molecules from free oxygen radicals. Therefore, depletion of Mn and Zn results in accumulation of harmful reactive oxygen species in the retina. This may decrease retinal function in heavy smokers. In this regard, the positive relationship between low contrast sensitivity and low Mn levels was especially noteworthy.

How can we explain the changes of serum trace elements in smokers? Changes in the levels of elements such as Cd and Pb in smokers (Lewis 1972) may affect levels of other trace elements in serum, because there is a steady state balance among the elements. Therefore, diminished Mn and Zn levels may be secondary to changes in the levels of other elements, resulting from the close relationships between toxic and nontoxic trace elements (Akyol et al. 1997).

Cigarettes contain cyanogen. Tobacco toxication is the consequence of exposure to the cyanide in smoke (Dreyfus 1965; Lessell 1971) as some biochemical tests and pathological findings have proved (Chisholm et al. 1967). Cyanide is a neurotoxin that has been implicated as the cause of TTON, especially in conjunction with disorders in vitamin B12 (cyanocobalamin) metabolism, including in malnutrition and alcohol abuse (Halliday et al. 1972; Ikeda et al. 1978). However, other toxic compounds such as nicotine and carbon monoxide may play important roles in retinal sensitivity loss.

Wang et al. (1992) found that serum Zn levels were diminished in all patients with TTON. Zinc has been shown to be a component of several enzyme systems involving nucleic acids, protein and carbohydrate metabolisms. Therefore, low Zn levels in TTON may reflect a malnourished state. Low serum Zn levels may also be due to deficient absorption of Zn caused by a tobacco chelation effect, as has been observed in other toxic optic neuropathies (Leopold 1978; Wang 1990). We found significant differences in serum Zn levels between smokers and controls. Marcheggiani et al. (1990) also found an increase in average levels of serum Cd and Pb in heavy smokers, but they found no differences in Cu and Zn levels in smokers as compared to non-smokers. Some research has indicated significant accumulations of Cd in the blood and of Cd, Cu and Pb in the lenses in smokers (Cekic 1998). The total quantity of Zn in the ejaculate (seminal plasma) of smokers has also been found to be significantly lower than in non-smokers (Oldereid et al. 1994).

In conclusion, we found reduced contrast sensitivity values to be associated with low levels of serum Mn and Zn. These metals function as cofactors of superoxide dismutase in erythrocytes and other cells, suggesting a possible role of trace elements in smoking-induced early retinal toxicity.


Appendix 1.

Table Appendix1.. 
Questionnaire used to obtain information about dietary and smoking habits.
How frequently do you eat the following foods?
  Fruit and/or fruit juice
  Milk and/or milk products
Do you drink alcohol?
Do you smoke?
If yes, for how many years have you smoked?
How many packs per day do you smoke?
What brand of cigarettes (regular or light)?
When did you begin smoking?
Have there been any changes in the amount you have smoked over time?
How long has it been since you ceased smoking?
Do you have any systemic diseases related to cigarette smoking?