A the nose and paranasal sinuses with occupational

Transcription

1 ADENOCARCINOMA OF THE NOSE AND PARANASAL SINUSES IN WOODWORKERS IN THE STATE OF VICTORIA, AUSTRALIA P. IRONSIDE, MB, BS,* AND JANE MATTHEWS, PHD+ The case index of the Cancer Institute of Victoria (Australia) contained 9 cases of adenocarcinoma of the nose and paranasal sinuses. Eighteen of the cases were in men and in a woman. Routine questioning of these patients revealed an occupation involving woodworking in 7 cases, whereas among 80 cases of other malignant tumors of the nose and sinuses there were only' 4 who had been woodworkers. Among the patients with adenocarcinoma of the nose and sinuses, there was a significantly higher proportion of woodworkers than in the general population. The findings are consistent with European reports associating nasal adenocarcinoma with wood dust, but whereas the workers at risk in Europe are mainly in the furniture industry, some of the workers affected in Victoria have been sawmillers or carpenters. The specific salivary patterns of tumors of mucous glands are not associated with woodworking. Cancer 36:5-2, 975. N ASSOCIATION OF ADENOCARCINOMA OF A the nose and paranasal sinuses with occupational exposure to wood dust was discovered in England' and has been confirmed in Belgium,' Holland,' Denmark, '* and France. lo There is also an association with dusty occupations in the boot and shoe industry, and possibly with baking and Hour milling.s The tumors appear always to start on the middle turbinate or in the ethmoid sinus and characteristically extend laterally into the orbit and upwards into the anterior cranial fossa. Lymph node metastasis is rare. The English experience with these tumors in the furniture manufacturing areas of Oxfordshire and Buckinghamshire has been summarized by Hadfield." In view of the findings in other countries the case records of the Cancer Institute were examined to see whether adenocarcinoma of the nose and sinuses is associated with woodwork- From the Cancer Institute, hlelbourne, Australia. * Assistant Pathologist. ' Statistician. Address for reprints: P. Ironside, Cancer Institute, 48 Little Lonsdale Street, Melbourne 3000, Australia. The authors are indebted to Dr. K. Motteram for assistance wich interpretation of the biopsies and for reviewing the manuscript. The clinical information was obtained from case histories compiled while the patients were under the care of Dr. J. P. hladigan or Dr. K. A. Hurley. Dr. W. N. Eastham obtained and translated the Dutch and Flemish papers. Received for publication July 25, ing in the State of Victoria (Australia). The biopsies were reviewed to see what histologic types of tumor might be associated with occupational exposure to wood dust. The Cancer Institute has the only megavoltage radiotherapy machines in Victoria, and consequently most patients requiring radiotherapy for paranasal sinus tumors in Victoria are seen there. MATERIALS AND METHODS All case records of the Cancer Institute indexed as primary malignant tumors of the nose or paranasal sinuses were searched, and all available biopsy sections were obtained for review. Cases excluded were those for which satisfactory histologic material was not available and those in which the tumor appeared to originate in the hairy skin inside the anterior nares. Sections stained with hematoxylin and eosin were sufficient for classification of most of the tumors, but the following special stains were used occasionally: periodic acid-schiff with and without salivary digestion, and the Alcian blue methods for rnucin; Gordon and Sweet's method for reticulin; Perls' method for hemosiderin; and Fontana's method for melanin.' Malignant tumors were classed as adenocarcinorna if glandular differentiation was evident, but no specific salivary pattern, as described by Foote and Frazell,' could be discerned. Other

2 6 CANCER September 975 Vol. 36 TABLE. Histologic Diagnoses in 99 Cases of Cancer of the Nose and Paranasal Sinuses Diagnosis No. Cases Squamous cell carcinoma A denocarcinoma Undifferentiated carcinoma Transitional carcinoma Adenoid cystic carcinoma hlixed salivary tumor hlucoepidermoid carcinoma Acinic cell carcinoma Lymphoma Sarcoma Plasmacytorna Neuroblastoma hl elanoma Unclassitiable I 9 4 classes identified were tumors showing specific salivary patterns (adenoid cystic, mixed, mucoepidermoid, and acinic cell tumors), squamous cell carcinoma, transitional carcinoma, l3 undifferentiated carcinoma, sarcoma, lymphoma, melanoma, neuroblastoma, and plasmacytoma. The study of occupational risks summarized in this paper is based on information recorded for the series of 99 public patients included in the case index of the Cancer Institute with malignant tumors of the nose and paranasal sinuses. A breakdown of the histologic diagnoses in this group of patients is given in Table. When public patients first attend the Cancer In- stitute the admission clerk routinely asks them their current occupation. The replies to the admission clerks' questions form the basis of the statistical analysis given in this paper. Since the findings of Acheson's group have become known, close questioning of patients by the medical staff has sometimes elicited a history of exposure to wood dust in previous occupations (Table 2). These cases have been included in the study of histopathology of adenocarcinoma in woodworkers, as have four additional cases which were detected by means other than the case index. Occupational Hazards RESULTS Among 99 cases of malignant tumors of the nose or paranasal sinuses indexed in the Cancer Institute, there were 9 adenocarcinomas (Table ). The most striking observation on the 9 patients presenting with adenocarcinoma was that there was only female (Case No. 4). The probability of such an extreme sex distribution if in fact both males and females are equally likely to present at the Cancer Institute as a public patient with the disease is less than.000, i.e. less than chance in 0,000. The result is obviously highly significant statistically; we can feel safe in concluding that males are at greater risk than females. One possible explanation for TABLE 2. Occupations of 9 Patients with Adenocarcinoma Possible hazardous Case Year of Age Occupations recorded by occupations elicited no. registration (yr) admission clerks by physicians I Sawmill proprietor* Pensioner Carpenter* Timber worker* -I Body builder* Pensioner Pensioner Pensioner Telephone technician 0 I Estimator II Carpenter* Maintenance engineer Sheetmetal worker Home duties Hospital general reliever I Builder' Retired Boot factory Joiner* Wood machinist* Baker and pastrycook Storeman Wood turner. I Carpenter' * Occupation probably involves exposure to wood dust.

3 No. 3 NASAL ADENOCARCINOMA Woo IDWORKERS bonside and Matthews 7 this greater risk is that an occupational factor is involved. Looking at the occupations recorded at the time of presentation by the registration clerks, we see that only 3 of the 8 males had their most recent occupation recorded. These included 3 carpenters, timber worker, and sawmill proprietor. The 966 Victorian Census shows that only 3.8% of the male workforce fell into the group described as carpenters, joiners, cabinet makers and related workers, and 0.3% fell into the group described as timbergetters and other forestry workers, a total of 4.%. If males in these occupations are as likely to present to the Cancer Institute as males in other occupations, the probability of having as many as 5 in a sample of 3 patients is less than.0002, i.e. less than chance in 8000, again a highly significant result. It may reasonably be argued that the Victorian male workforce is not an appropriate control group for our 3 males, as they have been drawn from only the public patients at the hospital (which introduces a considerable occupation bias) and also because their age distribution is different from that of the general workforce. The 3 males had ages at presentation ranging from 45 to 68 years. As an alternative control group, the total male public patients registered at the Cancer Institute in 973 were considered (973 being the only year for which the required information was readily available); 288 of these males were aged between 40 and 70 and had their current or most recent occupation recorded. Forty-three of these patients, i.e. 3.34%, had occupations falling into the carpenters, joiners, cabinetmakers and related workers category, and 3, i.e. 0.23%, had occupations falling into the timbergetters and other forestry workers category, a total of 3.6%. Thus, using this control group, the result of 5 out of 3 is equally significant. (The levels of significance quoted have all been calculated using exact binomial probabilities.) It may be noted that in the 80 cases of other malignant tumors of the nose and sinuses there were again significantly more males than females (55 males: 25 females). However, close examination of the occupations recorded for 47 of the males did not indicate any high-risk group. There were only 2 carpenters and timber worker, which is consistent with the general population level. Thus it appears that woodworkers have a greater risk of developing adenocarcinoma of the nose and paranasal sinuses than other members of the population, but not of developing other malignant tumors of these areas. Country of Birth Since there is a substantial proportion of immigrants in the Australian workforce, countries of birth were obtained from the case records (Table 3). The figures do not show any excess of immigrants among the cases of adenocarcinoma, which suggests that the occupational hazard exists locally. This has been confirmed by questioning of the more recent patients, which revealed that their exposure was obtained locally and that native timbers were those most commonly used. Histopathology of Adenocarcinoma in Woodworkers Among the cases obtained from the index of diseases of the Cancer Institute, there were 0 cases of adenocarcinoma in woodworkers (Table 2). An additional 4 cases have been collected, including 2 Cancer Institute cases which have come to notice since the index was searched, and 2 cases from private practices. These 4 cases form the basis of the following description. Adenocarcinomas are composed of tall columnar cells lying on fibrous stroma, which contributes a substantial proportion of the bulk of the tumor. The epithelium either lines small cystic spaces or covers broad papillary processes. Where cystic spaces are formed, a resemblance to colonic carcinoma is produced and may be accentuated by the presence of goblet cells (Fig. ). Where the epithelium covers papillary processes it is usually deeply basophilic and substantially free of mucous vacuoles. It is of irregular thickness, varying from one to several cells; where it is several layers deep, glandular spaces form within the layers (Figs. 2 and 3). In five cases the tumor was mainly of the cystic TABLE 3. Country of Birth of 99 Patients with Cancer of the Nose or Paranasal Sinuses Australia U.K. Others Unknown Woodworkers with adenocarcinoma a Other patients with adenocarcinoma 7 0 Patients with tumors other than adenocarcinoma

4 8 CANCER September 975 Vol. 36 Lymph node metastasis was diagnosed clinically in 2 patients at the time of diagnosis, and in patient 8 months after the original diagnosis was made. In only patient was there evidence of metastasis beyond the regional lymph nodes; this was a case where the liver appeared clinically to be involved. Invasion of the anterior cranial fossa probably caused the death of 9 of the 7 patients who have died. Fir;. I. Adenocarcinoma of the nose and ethmoid sinus in a carpenter. The resemblance to colonic carcinoma is strengthened by the presence of goblet cells (H & E, X370). pattern and in five it was mainly of the papillary pattern. In one case there was an even mixture of these two patterns. There were three cases which did not fit the commoner range of appearances; these were characterized by partly solid and partly glandular islands of mucin-containing cells, dispersed in masses of fibrous stroma (Fig. 4). These occurred in a builder, a saw miller, and a carpenter. Clinical Behavior The clinical behavior of the 9 index cases of adenocarcinoma was usually consistent with that described in the English cases. l Although it was difficult to determine the site of origin when several sinuses were involved, the nasal cavity and ethmoid sinus were clinically involved at the time of diagnosis in all except one case. When that patient died 4 years later, the autopsy diagnosis was adenocarcinoma of the ethmoid with extension into brain, pituitary, and spinal cord. The cases reviewed generally support Hadfield s finding that adenocarcinornas originate on the middle turbinate or in the ethmoid sinus. l DISCUSSION The figures presented indicate that adenocarcinoma of the nose and paranasal sinuses is an occupational hazard for workers exposed to wood dust in Victoria. The number of workers exposed is large, yet the number of cases seen at this hospital is quite small, only 2 cases having been seen during the last 0 years of the period reviewed. If flour dust and leather dust are occupational hazards, as has been suggested by Acheson et al.,3 then the risk in Victoria for these workers must be very small, since Melbourne, the capital city of Victoria, has quite a large boot and shoe industry, and with a population of about two million has, of course, a large bakery industry. The importance of the occupational hazard lies in the possibility of preventing this rare but distressing tumor and in allowing just compensation to workers who are afflicted by it. In the latter respect it should be noted that adenocarcinoma of the nasal sinuses is prescribed in the United Kingdom national insurance regulations as a disease for which woodworkers may receive compensation. I The occurrence of adenocarcinoma of the nose and sinuses among carpenters and sawmillers in Australia is rather different from the English experience, where the finer dust produced in cabinet makers and furniture makers shops has been implicated. There are at least three possible explanations for the difference: ) saw milling may be a proportionately larger industry in Australia and furniture making the larger industry in England; 2) the wood used by Australian sawmillers and carpenters may be drier than that used by their English counterparts by reason of climate or treatment of the timber; and 3) the varieties of timber used may be different. Certainly the patients who could be questioned remarked on the intensity of dust exposure; one sawmiller stated: the dust was terrible. Because of these considerations and because most Australian builders have been apprenticed as carpenters, building has been included among the hazardous occupations.

5 No. 3 NASAL ADENOCARCINOMA WOODWORKERS Ironside and Mattherus 9 FIG. 2. Papillary adenocarcinorna of the nose and paranasal sinuses in a wood machinist. Epithelium from one to several layers thick covers broad papillary processes (H & E, X75) Some discussion has ranged around the precise nature of the It is possible that some varieties of wood contain carcinogens and others do not. Whatever is the case it appears from this study that heavy exposure to wood dust from Australian native timbers is a hazard, and that the coarse dust of sawmills may be a hazard, just as the fine dust of cabinetmakers shops is. In the differential histologic diagnosis, the FIG. 3. Same case as Fig. 2 showing glandular spaces within the epithelium where it is multilayered (H & E, X470). --

6 20 CANCER September 975 Vol. 36 FIG. 4. Adenocarcinorna of the nose and paranasal sinuses in a builder. Large partly solid and partly glandular masses are distributed in a fibrous strorna. Signet-ring cells are evident (H & E. X400). two groups to be considered are transitional carcinoma as described by Osborn, and the specific salivary types. O~born ~ uses the term transitional carcinoma for a differentiated tumor resembling those found in the bladder. Kingertz applied the term cyclindric cell carcinoma to the same tumor. In differentiating papillary adenocarcinoma from transitional carcinoma, the most useful features are the irregular and sometimes single layer of epithelium of adenocarcinoma, compared to the regular multilayered epithelium of transitional carcinoma, and the abundant fibrous stroma of adenocarcinoma, compared to the delicate fibrovascular cores, which are more usual in transitional carcinoma. The presence of mucin within the cells is of limited value, since it may be present in transitional carcinoma and may be difficult to demonstrate in adenocarcinoma. In differentiating adenocarcinorna from salivary types it must be realized that one may find small areas in either that imitate the other. Sections at many levels through the block may reveal the true nature of the tumor when the biopsy is small. In the papers of Acheson s it has not been clear whether salivary patterns are included under adenocarcinoma or other tumors. In the series described here no tumors of specific salivary type were associated with exposure to wood dust. REFERENCES I. Acheson. E. D., Hadfield, E. Ii.. and Macbeth, K. C.: cancer in the Northamptonshire boot and shoe industry. Br. Carcinoma of the nasal cavity and accessroy sinuses in dled. J. i: woodworkers. Lancet i:3-32, Debois, J. hl.: rumoren van de neusholte bij 2. Acheson, E. I),, Cowdell, K. H., Hadfield, E. H., and houtbewerkers. Tijdschr. Gcneeskd. 25:92-93, 969. hlacbrth, K. G.: Nasal cancer in woodworkers in the fur- 5. Delernarre. J. F. M., and Thernans, H. H.: Het niture industr). Br..Ilcd. J. ii: , 968. adenocarcinoma van de neusholten. Ned. Tijdschr. Gmeeskd. 3. Acheson, E. D., Cowdell, K. H., and Jolles, B.: Nasal 5: , 97.

7 No. 3 NASAL ADENOCARCINOMA WOODWORKERS Ironside and Matthews 2 6. Drury, R. A. B., and Wellington, E. A,: Carleton's Histological Technique. New York, Oxford University Press, Editorial: Nasal cancer in woodworkers. Br. Med. J. ii: , Editorial: Nasal cancer in woodworkers. Lancet ii:253, Foote, F. W., and Frazell, E. L.: Tumors of the major salivary glands. Atlas of Tumor Pathology. Washington, DC, Armed Forces Institute of Pathology, Gignoux. hi., Bernard, P., and Gignoux, B.: Cancer de I'ethnioide et categories socioprofessionnelles. J. Fr. Olorhinoluryngol. 20: , 97. II. Hadfield, E. H.: A study of adenocarcinoma of the paranasal sinuses in woodworkers in the furniture industry. Ann. R. Coll. Surg. Engl. 46:30-39, hlosbech, J.. and Acheson. E. D.: Nasal cancers in furniture makers in Denmark. Dun. ;Wed. Bull. 8:34-35, Osborn, L). A,: Nature and behavior of transitional tumors in the upper respiratory tract. Cunrer 25:50-60, Rinyertz, N.: Pathology of malignant tumors arising in the nasal and paranasal cavities and maxilla. Acto Otularyngol. JSupp.J (Stockh.) 27, 938.