Respiratory symptoms pulmonary function among welders working with aluminum, stainless and railroad

SJOGREN Respiratory symptomsand pulmonary function among weldersworking with aluminum, stainless steel and railroad tracks. Scand J Work Environ Health 11 (1985) 27-32. Sixty-fouraluminum welders, 46stainlesssteelwelders,and 149railroad track welderswereinvestigated regarding respiratory symptoms and pulmonary function (forced vital capacity and forced expiratory volume in 1 s). Referents consisted of nonwelding industrial workers and raiload workers. All groups of welders showed a higher frequency of chronic bronchitis symptoms than their respectivereferents. Respiratory symptomswererelated to ozone concentrations in weldersworking withaluminum. In stain lesssteeland railroad track weldersrespiratory symptomswererelated to chromium exposurerather than to total particleexposure. The pulmonary function wasnot affected in any of the weldinggroups studied.

Welding is a process of fusing metals together by melting. Electric-arc welding uses techniques to protect the melting metal from the oxidizing power of the air. In metal arc welding the melted metal is protected by material in the coating of the electrode, and in gas-shielded welding an inert gas acts as protection. Two different methods of gas-shielded welding are used. Metal inert-gas (MIG) welding uses a consumable electrode, and in tungsten inert-gas (TIG) welding a nonconsumable electrode of wolfram is used.
Welding generates particles and gases which are characteristic for the method used and the material welded. Gas-shielded welding .of aluminum produces ultraviolet radiation, which causes oxygen in the air to form ozone (7). Recently a case of pulmonary fibrosis was reported for a worker who had welded aluminum for 17 years, mainly in confined spaces (28). Manual metal arc welding with coated electrodes on stainless steel generates particles of easily soluble hexavalent chromium (19,25,27). The major component of welding fumes from low-alloy steel is iron oxide (23).
The purpose of this study was to investigate respiratory symptoms in relation to concentrations of air contaminants and pulmonary function in different groups of welders as part of a comprehensive study of Swedish welders.
Reprint requests to: Dr B Sjogren, Section of Occupational Medicine, Research Department, National Board of Occupational Safety and Health, S-171 84 Solna, Sweden.

Subjects
The three different exposed populations were identified in two different ways.
Enterprises consuming welding material were traced in a Register of Enterprises held by the National Bureau of Statistics. Representatives from companies with some welding material consumption were interviewed about the number of welders, the material welded, and the method used, as well as the welding intensity for each welder. At the time of the study about 200 welders in the country worked mainlyon aluminum with gas-shielded welding, and about 400 welders worked mainly on stainless steel with coated electrodes (11). Aluminum and stainless steel welders were included in the investigation if they had worked for at least one year with the method being studied and at least 3.5 h each workday and were present on the days of the examination. In the interview many welders stated less welding time than that estimated by the representatives of the companies. The welders worked in a geographic region extending from 600 km south to 300 km north of Stockholm.
One group of 64 welders working with gas-shielded welding (both MIG and TIG) on aluminum, agreed to participate in the study and none refused. The median exposure time of this group was 5 (range 1-24) years. All 46 welders who worked with coated electrodes on stainless steel and who fulfilled the entrance criteria agreed to participate in the study. Their median exposure time was 15 (range 1-39) years.
At the time of the study 215 persons were welding railroad tracks for the Swedish State Railways. Welders working in remote areas of the country were excluded from the study for practical reasons. No referents was 70, and they were neither individually matched for age nor for smoking habits.
All the welders and referents were males.

Methods
The welding fume samples were collected on cellulose ester membrane filters (Millipore filter AAWP , diameter 37 mm, mean pore size 0.8 I'm) with personal sampling pumps (MSA monitaire sampler pump, model G) with a calibrated flow. The filters were placed inside the welding helmet. Total particle welding fume amounts were measured gravimetrically. The sampling period was at least 4 h, and timeweighted average concentrations for the sampling period were calculated. Total chromium amounts were analyzed with a particle-induced X-ray emission (PIXE) technique at the Department of Nuclear Physics, Institute of Technology, Lund. The detection limit was O.ll'g/ m J (26), and the coefficient of variation was IO OJo (14).
Ozone was measured by a chemiluminiscence instrument inside the welding helmet. This instrument is practically specific for ozone, and the detection limit is about 0.001 ppm (26). These measurements were performed during at least two I5-min periods.
Ulfvarson (26) has thoroughly described all the methods used for the assessment of particles and gases in these studies , and all measurements were aimed to represent normal work conditions.
All the participants were interviewed in a standardized way. A modification of the British Medical Research Council's quest ionnaire on respiratory symptoms was used. Chronic bronchitis was defined as daily cough and production of phlegm for as much as three months each year . This definition is very close to that given by the World Health Organization (30).
Forced vital capacity (FVC) and forced expiratory volume in I s (FEV 1.10 were registered at least three times for each person with a vitalograph. The tests were performed with the subjects in the standing position and without a noseclip. The best value for each variable was used in the analysis, and all volumes were corrected to body temperature and pressure saturated with water vapor (BTPS). Regression equations were calculated in which age and height were independent variables and FVC and FEV1.0 were dependent variables.
Two vitalographs were used during the study. At the end of the data collection one spirometer showed plus 0.1 I at the 5-1 calibration and the other minus 0.1 I at the same volume.
Nonsmokers were defined as subjects who had never smoked or who had quit smoking more than two years ago. Subjects who stopped smoking less than two years ago were excluded.
A specific restrictions as to welding time per day were applied to this group. One hundred and forty-nine welders accepted participation and none refused. The welding work consisted mainly of joining rails by mold welding and overlay welding of worn or damaged rails . Basic electrodes were used in both operations. The electrode used for overlay welding contained 3.2 OJo chromium. The median exposure time of these welders was 18 (range 5-41) years.
An aged-matched (within four years) referent was chosen for each aluminum and stainless steel welder. These referents were nonwelding industrial workers chosen from the same company or from the geographically nearest company. If possible the referents were also matched for smoking habits.
Referents for the railroad track welders were nonwelding railroad workers from about the same geographic areas as the welders. The number of   Tabl e 2. Reported chronic bronchitis among all the welders and th eir referents divided into smokers and nonsmokers . Welders wi th exposure of less than 2 years were excluded ." The differences in the frequencies were tested by the chi-square test, and the differences in the means were tested by Student's t-test. The Mantel-Haenszel test was used for the analys is of stratification (20 Electric-arc welding generates fume in which the majority of the particles are smaller than I /lm and thus respirable (2,26). The distribution of the total particle concentrations for different welding operations without loca l exhaust ventilation is shown in figure I. The current Swedish occupational exposure limit for hexavalent chromium is 20 I-'g/m 3 • More than 80 rJ10 of the chromium concentrations exceeded that value when stain less steel was welded with coated electrodes.
The current Swedish 8-h occupational exposure limit for ozone is 0.1 ppm. Almost 50 rJ10 of the ozone concentrations exceeded that limit when MIG welding was performed on aluminum . The concentrations of oxides of nitrogen were typically below 5 ppm , regardless of the welding method .
The exposure cond itions have been extensively analyzed and pub lished by Ulfvarson (26).

Chronic bronchitis
All the groups of welders reported higher frequencies of chronic bronchitis than their respective referents, but the di fferences were not statistically significant (tab le I).
As smoking is strongly associated with chronic bronchitis , a stratification due to smoking was made (table 2). All the welders were included, as the number of welders with chronic bronchitis was very small in each subgroup. This analysi s sho wed smoking to be a stronger determinant for chronic bronchitis than welding fume expo sure . a Crude risk ratio due to welding: 2.14 (p = 0.12); ris k ratio after strati fication for smoking: 2.08 (p = 0.13). Crude risk ratio due to smok ing : 6.26 (p < 0.001); ris k ratio after stratification for weldi ng: 6 .15 (p < 0.001).

Resp iratory symptoms related to welding
All three groups of welders had significantly more respiratory sympto ms (eg, cough, phlegm, or a sense of irritation) in connection with work than their respective reference group . In general these symptoms were of shorter duration than the symptoms included in the definition of chro nic bronchitis. Welders working with gas-shielded welding on aluminum had more respiratory symptoms at higher ozone concentrations than at lower concentrations (figure 2). A comparison between exposure category I and category II gave a p-value of 0.03.
Respiratory symptoms were more comm on at higher than at lower chro mium concentrations among stainless steel, as well as among railroad track, welders (figures 3 & 4), but this tendency was not statistically significant.
Respiratory symptoms did not seem to show any relationship with the total particle concentrations of the three groups of welders.

Pulmonary function
As FVC and FEVl.o vary with age and height, all these variables are presented (table 3). There were no differences regarding FVC or FEVl.o between the three groups of welders and their respective reference groups. Neither did the subjects with long exposure periods have decreased pulmonary function variables . % with resp iratory s ympt oms

Chronic bronchitis
Publis hed reports on welders have explicitly or implicitly concerned electric-arc welding on low-alloyed steel, and they have shown an increased frequency of chronic bronchitis with risk ratios varying between 1.2 and 3.1 (I, 3,4,5,8,22). This study also showed somewhat increased frequencies of chronic bronchitis in all the subgroups of welders , close to the previously found risk ratios.
Stainless steel welding with coated electrodes has previously been associated with asthma (16). Th e present study did not reveal obstru ctive disease among the expo sed subjects, a findin g which could be expla ined by selection throug h job exchange for welders with severe respiratory disorders.

Respiratory symptoms related to welding
The present study showed ozone rat her tha n aluminum pa rticles to be responsible for the respirato ry symptoms among the alumin um welders, this finding is in accordance with the results of an experimental study (9) in which 0.35 ppm of ozone, but not   0.20 ppm , was associated with respiratory symptoms . In a previous cross-sectional study of stainles s steel and mild steel welders both groups had about the same frequency of respiratory symptoms (15); this finding agrees with the result s of the present study.
For the stainless steel welders, as well as for the railroad track welders, respiratory symptoms seemed to be more correlated with chromium exposure than with total particle exposure. Most of the chromium is in a hexavalent state (19,25,27), and such compounds are known to be highly irritative (18).

Pulmonary function
As mentioned in the Methods section two vitalographs were used. One showed plus 0.1 I and the other minus 0.1 1at 5 1. Each vitalograph was used to investigate an equal number of exposed welders, as well as referents in a particular workshop, ie, an equal proportion of welders and referents were investigated with the same spirometer . Thus the internal valid ity was adequate for the study. Furthermore, the regression equations, in which FVC and FEV l.o were calculated from height and age, were similar to previously reported equations (6).
In recent case reports pulmonary fibro sis and desquamative interstitial pneumonia have ·been discovered in two aluminum welders (12,28). These cross-sectional data do not support an y deterioration due to long -term exposure. The inherent selection weakness of a cross-sectional study must however be borne in mind.
The present study supports the findings of one published study showing no effect on pulmonary function in sta inless steel welders in a comparison with reference values (I5).
The railroad track welders in this study did not exhibit any impairment of pulmonary function. Obstructive changes have been observed previously among shipyard welders (I , 13). These differences concerning pulmonary function might be explained by the fact that shipyard welders often work in semiconfined or confined spaces , contrary to railroad track welders, and therefore are exposed to higher concentrations of air contaminants. In railroad track welding 25-50 070 of the measured total particle concentrations inside the welding helmet exceeded 5 mg/m! (figure 1), and in Danish shipyards 50 % of the measurements in confined spaces were above 7 mg/rn! (24). However, the results from our study are in accor dance with the spirometric results from welders working in engineering shops (3).