Quantitative evaluation of ECG components of workers exposed to carbon disulfide.

The components of electrocardiograms (ECG) of 253 workers exposed to carbon disulfide (CDS) and those of 99 controls were quantitatively measured and evaluated. ECG of the exposed workers showed a statistically significant higher prevalence of ECG pathological changes, higher P amplitude and Macruz index, longer P duration, longer both crude and corrected Q-T intervals and R-R intervals and shorter P-R segments and QRS intervals than that of the controls. On the other hand, P-R intervals and heart rates of the two groups were not significantly different. Among both the the exposed and control groups, values of P duration were significantly negatively correlated with that of P-R segment: r = -0.216 and -0.132, respectively, p greater than 0.05. Values of the ECG components were not related to duration of exposure to CDS. Moreover, no significant difference was observed between duration of exposure of the exposed workers with and of those without pathological ECG changes. The method used in our study may be useful in evaluating the ECG of exposed workers before the appearance of the known pathological abnormalities.


Introduction
Carbon disulfide (CDS) is an industrial solvent used mainly in the manufacture of artificial silk and rubber. In spite of its reported toxic effects, it is still used in many countries, including Egypt. The cardiotoxic effect of CDS had been reported by many authors (I). Although CDS related electrcardigraphic changes were demonstrated by some investigators (2)(3)(4)(5), others observed the absence of such a relationship (6,7). Since electrocardiography (ECG) is a simple, quick, nonexpensive, and noninvasive technique that may be helpful in early detction ofthe cardiac toxic effects of CDS, more investigation may be needed to evaluate the ECG patternofCDS-exposed workers. Accordingly, this study was initiated as a trial for quantitative evaluation ofthe measured components ofECG of CDS-exposed workers, which my provide an indication ofthe cardiac effect ofCDS even in the absence of the typical known ECG abnormalites.

Subjects and Methods
The study was carried out on a random sample of253 workers selected out of 503 workers exposed to CDS in the viscose rayon department of ESCO Factory at Bahteem (a suburb of Cairo).
This fiktry is old and lacks the sufficient environmental preventive measures. The workers do not use respiratory masks. Records of the factory indicated that the level of workers' exposure to CDS ranged from 20 to 45 ppm.
A control group of 99 workers was selected from Wooltex company for spinning and weaving of wool. They were of the same age groups and occupational grade as that ofthe CDS exposed workers. Each member ofthe study and control group was interviewed about personal data, smoking habit, occupational history (which included site, nature, and duration ofthe present or pastjobs), complaints (ifany), cardiovascular-related symptoms, and past history of other diseases. This interview was followed by a short, general clinical examination. Those with past history of diabetes, liver, kidney, and rheumatic heart diseases were excluded from the study. Electrocardiography ECG tracing was in a quiet room in the factory clinic, using a Philips apparatus connected with the standard 12 leads (6 limb leads and 6 precordial leads). Subjects were instructed to avoid physical activity and smoking for at least 1 hr betfre the recording. The apparatus was regularly calibrated so that every 1 mv produced 10 mm deflexion. The speed ofpaper movement was 25-mm/sec.
Each electoardiogram was examined for rate, heart rate, rhythm, p-wave duration and amplitude, P-R segment, P-R interval, QRS duration, Q-T interval, S-T segment displacement, and other pathological abnormalities (8). The criteria of these parameters are described in the footnotes of Table 1. 3KMAL ETAL. aP-wave parameters in lead II: duration, number of small squares included between the beginning and the end of P-wave. Each small square equals 0;04 sec; amplitude, number of small squares included between the top of P-wave and the base line; each small square represents I mm.
bP-R segment, number of small squares from the end of P-wave to the beginningpof QRS complex; each one repreents 0.04 sec.
CP-R interval, numbers ofsmall squares from the beginning ofP-wave to the beginning ofQRS complex; each small square represents 0.04 sec. This was measured in standard lead II.
dQRS duration, number of small squares from the beginning of Q-wave to the end of S-wave; each one represents 0.04 sec.
eQ-T interval, number of small squares from the beginnning of QRS complex to the end of T-wave; each one represents 0.04 sec. The corrected Q-T interval was calculated as Q-T/R-R. The Macruz index was calculated as P duration/P-R segment (9). Statistical analysis of the results was done on IBM computer supplied with the Microstat package of statistics. Statistical tests used were Student's t-test, chi-square test, and Z test for two proportions.

Results
The mean ages ofthe CDS-exposed workers and controls were 39.37 + 9.36 years and 41.2 ± 10.3 years, respectively. The difference is statistically insignificant, p > 0.05.
Smokers constituted 49 and 47% ofthe CDS-exposed workers and controls, respectively. The difference is statistically insignificant, p > 0.05. Moreover, frequency distribution ofboth groups as regards their systolic or diastolic blood pressure did not reveal significant differences between the two groups.
Duration of exposure of the study group ranged from 4 to 29 years (mean ± SD = 15.4 ± 8.13 years). The prevalence of the ECG pathological changes among the exposed workers was significantly higher than among the controls ( Table 2).
Frequency distribution of the two groups according to the values of the measured components of the ECG (Table 1) demonstrated statistically significant trends toward higher P amplitude and Macruz index, longer p duration, longer crude and corrected Q-T interval and R-R interval (border line significance,p = 0.044), and shorter P-R segment and QRS interval among the exposed group than among the controls. On the other hand, no significant difference was observed between both groups as regards P-R interval and heart rate. Among both the exposed and control groups, values ofP duration was significantly negatively correlated with that ofP-R segment: r = -0.216 and -0.132, respectively, p < 0.05. No statistically significant association or correlation could be observed between duration ofexposure to CDS or smoking habit and values ofthe ECG components. Moreover, no significant difference was observed between duration ofexposure ofthe exposed workers with and ofthose without pathological ECG changes

Discussion
The absence of significant difference between the exposed workers and the controls in regards to age, smoking and systolic or diastolic blood pressure most probably exclude the confounding effects ofsuch factors. The higher prevalence ofpathological ECG changes among CDS-exposed workers than among controls is in agreement with data of previous reports (2)(3)(4)(5).
The term nonspecific S-Tchanges (Table 2) is a cautious term used in the presence ofS-T depression and/or T-wave inversion, which may occur in conditions other than ischemic heart disease, e.g., electrolyte imbalance, ventricular hypertrophy, metabolic disorders, acid-base balance changes, hyperventilation and drugs. These ECG changes can be considered specific when they occur during exercise or in the presence of concomitant chest pain. However, such changes should be followed up, as they may precede manifest ischemic changes (10). Elimination ofthis ECG abnormality from Table 2 may lead to an impression ofabsence of significant difference between the CDS-exposed workers and the controls. Recently, Knapikowa et al. (7) used the Minnesota code, which is a very specific code, for evaluation ofthe ECG of CDS-exposed workers, and they could not find significant differences between the exposed workers and the controls. Accordingly, we tried in this study to evaluate the the ECG of CDSexposed workers using the values of its measured components, which may indicate the cardiac effect of CDS before the appearance of the specific signs and symptoms of myocardial ischemia. The longer P duration and the higher P amplitude among CDS-exposed workers than among controls may point to ischemic changes. Patients with coronary heart diseases may show a broad P-wave without actual left atrial enlargement (11). The mechanisms ofsuch prolongation may be related to delayed intra-atrial conduction (11), which may be a part of systemic affectation of peripheral nerves (12).
It seems that prolongation ofP duration occurs at the expense ofthe P-R segment. This suggestion may be documented by the observed shortening ofthe P-R segment ofthe exposed workers without significant change of the P-R intervals and by the statistically significant negative correlation between the values of P duration and P-R segment among both study and control groups. Accordingly, the use of Macruz index in indicating changes that occurred in P duration, P-R segment, and P-R interval is highly recommended because it expresses the degree of prolongation of P duration at the expense of the P-R segment. This was obvious from the markedly higher mean Macruz index ofthe exposed workers than that ofthe controls (2.0904 ± 1.054 and 1.456 ± 0.593, respectively, t = 7.133, p < 0.00001). It should be mentioned that high Macruz index was assumed to connote impaired atrial electrgenesis even in the absence ofconcomitant pathological ECG changes (13).
Interestingly, the observed mean value ofthe Macruz index of the controls is nearly similar to that observed by Taccola et al. (13) among their control group (1.43). However, the mean value among our study group is much higher than that of Macruz index of their exposed group (1.65). This may indicate a higher exposure effect ratio in our study group, which is most likely due to less effective environmental preventive measures and the unsatisfactory usage of personal protective devices (14).
The significance of the observed longer crude and corrected Q-T intervals among CDS-exposed workers than among the controls is difficult to establish from this study. However, changes ofthe Q-T interval have been reported to be associated with or precede polymorphous ventricular tachycardia (15) or even serious ventricular arrhythmias as reported in cases of organophosphorus insecticides poisoning (16).
The nonsimilarity betweentheresultsofcomparisonsbetween the exposed and control groups regarding the R-R interval and heart rate mostprobably aredueto the factthattde two parameters were not measured exactly at the same moment. Both can be affectedby many factors suchasmovementemotional state, respiration, etc. However, the borderline significantdifferencebetween the R-R interval ofthe exposed and control groups (p = 0.044), togetherwith the absence ofsignificantdifferencebetweenthetwo groups in regardto heartrate mostprobably limits thevalue ofthis interval in indicating a myocardial effect ofCDS-exposure. The shorter QRS duration among exposed workers than among the controls may point to acceleration ofventricular depolarization.
The absence of significant association between duration ofexposure and the values of the ECG components, together with absence of significant difference between duration ofexposure of the exposed workers with and of those without pathological ECG changes may indicate that the duration of exposure is not the determinant risk factor for the development ofthe cardiotoxic effect ofCDS. This most probably supports the suggestion that CDS only manifests its coronary effect in the presence ofothers predisposing factors (1).
According to the results of our study, it seems that CDS had different effects on different parts ofthe heart. An exact explanation of such findings cannot be provided because there are no available data about studies concerning this point. However, cells of different parts of the heart have different transmembrane potentials, different thresholds, and action potentials ofdifferent shapes (17). This may lead to different responses ofthe different parts of the heart to CDS.
It would be interesting to investigate such changes among workers who developed pathological ECG changes. Unfortunately, this could not be demonstrated because the number of individuals in the group was too small to validate significance of the statistical analysis. It can be concluded that the method used in our study may be useful in evaluating the ECG of exposed workers before the appearance of the known pathological abnormalites.