Seven-year follow-up of white-finger symptoms and radiographic wrist findings in lumberjacks and referents.

S, ZITTING Al. Seven-year follow-up of white-finger symptoms and radiographic wrist findings in lumberjacks and referents. Scand J Work Environ Health 1994;20: 10 1-6. In 1978, a cross-sectional study of 279 lumberjacks exposed to hand-arm vibration and 178 unexposed referents was conducted. The aim of the present study was to provide a seven year follow-up on (i) the changes in the white-finger symptoms among the lumberjacks and the refer ents, (ii) the effect of white-finger symptoms on the professional prognosis of the lumberjacks, and (iii) the relationship between hand-arm vibration and changes in wrist bones. METHODS - The methods consisted of a questionnaire, a routine clinical examination, and radiographs of the wrists and hands. The examinations were conducted as a field study using a mobile unit. RESULTS - The prevalence of white-finger symptoms was 18% among the lumberjacks and 3% among the referents in the original study. Seven years later 213 lumberjacks and 140 referents participated in the follow-up. The prevalence of white-finger symptoms was 24.9 among the lumberjacks and 5.7 among the referents, and the seven-year cumulative incidence was 14.7% among the lumberjacks and 2.3% among the referents. After allowance for age, there was no difference in the incidence of white finger symptoms between the lumberjacks with fewer than 15 years of exposure and the referents, but the risk increased with increasing duration of exposure (risk ratio 8.9, 95% confidence interval 2.5 28.9 for those exposed at least 25 years). There was no difference between the two groups in the prev alence of radiographically detectable translucencies or osteoarthrotic changes in the wrists and hands. CONCLUSIONS - According to the results, white-finger symptoms are still a problem among lumber jacks who started chain-saw work before 1970.

Chain-saw vibr atio n can caus e w hite-fi nger sym ptom s. It ha s been cl aimed that it also cau ses rad iographica lly detect abl e tran slucen cie s in the carpal bones of exp osed workers (1-3), but the results of various studies hav e been contro vers ia l (4)(5)(6) .
The pr eval ence of wh ite -fin ger sy mpto ms among Finni sh lumberjack s was at its high est in th e early 1970s, when it was 40-60% (7,8). Du ring the I 970s, the prevalence decreased, and in studies conducted at the end of the decade, it was 13-18% (6,9) .
In 1978 we stud ied 279 lumberjacks whos e me an ex pos ure time to cha in-saw vibration was 10.4 years (6). On e hundred and seventy-eight peat workers who we re not exp osed to hand-arm vibr ation serve d as the referen ce group. The prevalence of white-fin ger I LEL Employment Pension Fund, Helsinki, Finland. 2 Finnish Institute of Occupational Health, Helsinki, Finland. .1 Kuopio Regional Institute of Occupational Health, Kuopio, Finland. 4 Department of Public Health, Helsinki University, Helsinki, Finland. Reprint requests to: Dr 1 Kivekas, LEL Employment Pension Fund, PO Box 20, FIN-0024I Helsinki, Finland. sy mptoms was 18% among the lumberj acks and 3 % among the referents, re sp ecti vel y. The overall prevalence of circumscribed tran slu cen cies in the w ris t bones of the lumberj acks was 9 % for each hand. In the reference group, the respective prevalences were 8 and 9%. Th e pre val ence of bone trans lucencies was not rel ated to either the duration of exposure or th e oc currence of white-finge r symptom s.
In 1985 we conducted a follow-up study of th e same gro ups. The objective w as to study (i) the chang es in whi te-finger symptom s among the lumberjacks and the referents, (ii) the effect of whitefin ger sy mptoms on the pro fessional pro gnosis of the lumberjacks, and (iii ) the relationship between handarm vibrat io n and cha nges in wri st bones.

Subje cts
In the foll ow -up phase , 2 13 lumberjacks (76% of the original gro up) and 140 referents (7 8% of the orig inal group) we re ex amined. Th ose 46 lumberjacks who had left their previous occupation during the follow-up peri od we re included. The pre val ence of white-finger sy mpto ms in 1978 was slig htly lower (17 versus 18%) for the lumberj acks and higher (5 versus 3%) for the referents who participated in the foll ow-up study as compared with the original groups.
At the end of the follow-up, the mean age of the lumberj acks was 41.9 (SD 9.2) years and that of the referents was 39.7 (SD 8.9) years. The lumberjacks' mean expo sure time to hand-arm vibration was 19.7 (SD 12.4 ) years, with a range of 8-35 years. Data on exposure were not obtained for four lumberj acks.
Peat workers drive a tractor full -time dur ing the har vesting season (3-5 months per year), and during the winter they mainly do maintenance and repair work . Referents exposed to chain-saw vibration for more than one year had been excluded from the stud y in 1978.

Methods
The study program con sisted of a questionnaire, a routin e clinical examin ation and radi ogr aphs of the wrists and hands. The examinations were conducted as a field study using a mobile unit as in the previous study.
The survey included que stions on work history, general state of health, and symptoms of white fingers. Most of the questions were the same as in the previou s study. The answers were checked for completene ss. The prevalence of white-finger symptoms was estimated on the basis of the answers to the question "H ave you had finger blanching?" Because of technical difficulties, radiographs of the hands and wrists were obtained from only 188 lumb erjacks and 126 referent s in 1985.
The occurrence of circumscribed translucencie s in the wrist bones was determined from the radiograph s tak en in 1978 and 1985. The film s were read independ entl y by two radiologists who did not know the subject's age, occupation, or when the film s were taken. In cases of discrep ant classifications, j oint rereadings were made. The agreement rate of the first interpretation was 82%. The size and shape (regular or irregular) of the translucencies, as well as the possible sclerosis of their borders, were regi stered.
The degree of osteoarthrotic chan ges in the joints of the hand and wrist bones was classified according to Kellgren & Lawrence ( 10) on the basis of referen ce radiographs. According to this classification, 102 the findings were regi stered on a five-point scale (none, possibl e, mild , moderate , severe) for each joi nt. In the analysis, the occurrence of mild, moderate, or seve re osteoarthrosis was considered osteoarthrosis.

Data analysis
Separate analy ses were condu cted for the prev alence data and the seve n-year cumul ative incid ence data (first-time event s). Cumulative incidence was approximated as the prop ortion of those who had contracted white-fi nger symptoms of those previously symptomless . In the stratified analysis, risk ratios (RR) and 95% confidence interv als (95% CI) (based on the standard errors) were estimated (II ). The test for trend in proportions was analyzed according to Nurminen (12).

Prevalence of white-finge r symptoms
In the beginning of the follow-up, the prevalence of white-finger symptoms was highe r amon g the lumberjacks than among the referent s. There was a significant increase in the lumberjacks' prevalence, whereas no change was found for the referents (table I ).
White-fin ger symptoms were rare amon g the men younger than 37 years of age ; the risk was five-to sixfold in the two older age groups (table 2). After allowance for age, there was a significant trend for the prevalence of white-finger symptoms according to the duration of exposure (chi square 30.6, P<O.OO I ). The risk ratio for at least 25 years of exposure was 6.8 (95% CI 3.0-15. 8).

Incidence of white-finge r symp toms
The seven-yea r cumulative incidence of white-finger symptoms was 14.7% among the lumberjacks and 2.3% among the referents (RR crudc 6.5, 95% CI 2.4-17.5). For the 23 new cases of white-finger symptoms among the lumberjacks the mean exposure time from the onset of exposure to the beginning of symptoms was 17.5 (SD 2.8, range 13-25) years. The incidence was highest in the middle-age group (table 3). After allow ance for age, there was a significant incre asing trend in the incidence of whitefinger symptoms acc ording to the duration of exposure (chi square 16.6, P<O .OOI ). The risk ratio for at least 25 years of exposure was 8.9 (95% CI 2.5-28.9).
White-finger symptoms developed in only one previously asymptomatic lumberj ack younger than 37 years and one exposed for less than 15 year s. The number of lumberjacks who reported white-finger symptoms in 1978 was 78, and 72 of the 78 reported white-finger symptoms also in 1985. Five of the six "recovered" lumberj acks had reported white-fi nger symptoms only in one hand in 1978. Table 2. White·finger sym ptoms acco rding to duration of exposure to chain-saw vib ration and age in 1985. (A = all su bjects in the exposu re categ ory, B =number of subjec ts wit h white-finger sym ptom s, 95% CI =95% con fid ence int erval)

Radiographic fi ndings
There were no significant differences in the occurrence rates of translucencies in the wrist bones between the lumberj acks and the refere nts (table 4). The pre valence of translucenc ies increa sed as age increased. Lunatoma lacia was found in 1978 in three lumberjacks and in 1985 in four lumberjacks. Two of the four had previously injured the affected wrist. Among the refe rents, no one had lunatoma lacia in 1978, and one had it in 1985.
The prevalence of osteoarth rosis of the hands and wrists increased from 5.3% in 1978 to 16.0% in 1985 among the lumberjacks and from 6.4 to 12.7% among the referents. There was no difference between the hands. In both groups there was a strong relationship between osteoart hrosis and age (table 5). After allowance for age, the prevalence of osteoarthrosis was not related to exposure time among the lumberj acks.
White-finger symptoms in the beginning of the follow-up was not a significant predict or for selection out of occ upation among the lumberj acks. Twent yfour of 36 lumberj acks with white-finger symptoms in 1978 were still work ing as lumberj acks in 1985, four had changed occupations, and eight had retired . Likewise, there was no evidence for a relationship between the radiographically detectable tra nslucencies in the wrist bones and turno ver among the lumberjacks.

Discussion
According to this study, white-finger symptoms are still a problem for lumberjacks, at least among the older men exposed for a long time. The seven-year cumulative incidence of white -finger symptoms among the lumberjacks exposed for at least 25 years was as high as 30.6%, and the prevalence at the end of the follow-up was 49.0%. In a recent Japanese study the prevalence of white-fi nger sy mptoms among worke rs with a duration of exposure of 30 years or more was 20.9% (13). In this study only one new case of white-finger symptoms occurred in the least-exposed group of lumberjacks, who started working with chain saws after 1970. The differential participation in our follow-up study with regard to white-finger symptoms in 1978 among the two groups may have caused bias. This bias would tend to dilute the occupational effect and thus would not substantially influence the conclusions drawn from our study.
In our study, the prevalence and incidence of white-finger symptoms was determined from the workers' subjective reports of finger blanching without information on its origin or a specified period of observation. Thus there may have been some cases of finger blanching due to circulatory problems in both groups. The proportion of such cases was assumed to be similar in both groups. The same question concerning the last two years has been used in studies in which the prevalences of white-finger symptoms have been slightly smaller (8,9). Our study was a prospective follow-up of workers with a long exposure history. The mean exposure time was 19.7 (range 8-35) years, which is eight years longer than in the aforementioned studies (8,9), and this difference may explain the higher prevalence.
The prevalence of white-finger symptoms decreased during the 1970s, and the most important factor contributing to this decrease was the damping of chain-saw vibration. New chain-saw models that generate less vibration appeared in logging work in the late 1960s and early 1970s (14)(15)(16)(17). In addition, the vibration measured from the wrists of lumberjacks has been lower with these third-generation chain saws.
In this study, there was a group of lumberjacks who only or mainly had used third-generation chain saws. Only one of these men reported symptoms of white-fingers at the end of the follow-up. In Japan, Futatsuka & Ueno (18) found white-finger symptoms in only 1.6% of the lumberjacks who had been exposed for seven to eight years and had started using chain saws in 1972-1973. The mean exposure time until white-finger symptoms first occurred was 17.5 years among the new cases. This length of time is much longer than the four to five years reported in the early 1970s (8).
It seems that white-finger symptoms develop much more slowly when modern saws are used instead of older on es. Among the lumberjacks with an exposure hi story of at least 15 years , however, the pr e valence and incidence of white-finger symptoms inc reased with an increase in the duration of exposure. This find ing indicates that the vi bratio n of mod ern sa ws can elicit or maintain sympto ms in itiated by former exposure to more intense hand-arm vibration.
Two out of three of the lumberj acks who reported white-finger symptoms in 1978 were still in the same occupatio n in 1985 . Most of th em still had sy mptom s, but the sy m pto ms were less se vere. T ypi cally, the symptoms a ppea red when the men ' s hands ca me into contact with cold and moist condition s, for instance , while fis hing during lei sure tim e .
There was no di fference bet we en the study and the reference groups in the pre val en ce of radiographically detectable tran slucencies in the wri st bones. The pr evalence of tra ns luce ncies inc re as ed as ag e increased in both groups.
The prior evidence for a connection between translucencies in th e wrist bones and hand-arm vibration is ba sed on studies made in the early 1970s ( I, 2, (9). The lumberja ck s examined in these st udies had a long history of working with the old, hea vy c ha in sa ws , whose vibratio n wa s po orly damped .
A similar lack of correlati on between hand-arm vibration and translucencies, as in thi s study, has been reported in man y earlier studi es (4-6). According to these studies , e xpo sure to hand-arm vibratio n, at least with the modern type of ch ain saws, docs not ca use radiographically detect able translucencies in wrist bones.
In this study the prevalence o f osteoarthrosi s was gr eatly age-dependent, but it was not dependent o n exposure to hand-arm vibratio n. In both gro ups , the pre valence was less than in the Fi nnis h agri cultural population (20). The determination of oste oarth rosi s in thi s study was ba sed o nly on radi ographic ch an ge s in joints of the wrists and hands.
Findings on the connection bet ween osteoarthrosis a nd physical load ha ve been co ntro ve rsi al. Gordon (2 1) and Zinn (22) found osteoarthros is more ofte n in farm ers than in other populations. However, in many studies there has been no connection betw een the pre valence of osteoarthrosis and physical load (23-25) o r vibration (26) .