Recent trends in lung cancer mortality in Canada A.B. MILLER, MB, FRCP[C]
In the last decade lung cancer mortality has continued to Increase in men, especially those aged 35 to 64 In the province of Ouebec, and It has Increased strikingly in women. The percentage of regular smokers of cigarettes has decreased recently among men but has increased among women. It is concluded that It may be some years before mortality begins to decrease in either sex, though decreases may be anticipated earliest in men in Ontario, the Prairies and British Columbia. More antismoking efforts directed to individuals are required, especially for young women. Au cours de Ia derniere decennie Ia mortalite reli6e au cancer du poumon a continue d'augmenter chez les hom. mes, particulierement chez ceux de 35 a 64 ans de Ia province de Quebec, et elle a augment6 de fa9on frappante chez les femmes. Le pourcentage de fumeurs reguliers a baiss6 r6cemment chez l'homme mais ii a augment6 chez Ia femme. On conclut qu'il faudra possiblement quelques annees avant que Ia mortalite commence a d6croitre chez les representants des deux sexes. bien qu'une baisse puisse-t-.tre anti. cipee plus t6t pour les hommes de l'Ontario, des Prairies et de Colombie. Britannique. Une lutte accrue contre l'usage du tabac s'adressant aux mdi. vidus et particulierement aux jeunes femmes est requise. In 1966 Springett1 demonstrated from mortality figures in England and Wales that, at least in those countries, the beginning of the end of the increase in mortality from carcinoma of the lung was in sight. In 1972 Schneiderman and Levin2 produced evidence suggesting that similar changes in the United States might be expected shortly, and Gray and Hill3 in 1975 produced evidence of a reduction in mortality among men in Australia. It is thus timely to examine data for Canada to see whether any decrease in mortality has occurred here. Methods Data on numbers of deaths from lung cancer in 5-year age groups are Presented in part at the symposium on oncogenesis, jointly sponsored by the Canadian Society for Clinical Investigation and the Royal College of Physicians and Surgeons of Canada, Quebec City, Jan. 21, 1976 Reprint requests to: Dr. AB. Miller. Director. Epidemiology Unit, National Cancer Institute of Canada, 121 St. Joseph St., Toronto, Ont. M55 2R9
available from the publications of Statistics Canada (formerly the Dominion Bureau of Statistics). Because of changes in the International Classification of Diseases, care was taken to ensure that, as far as possible, a similar grouping was used throughout. Age-specific mortality rates for Canada were calculated for 5-year periods except for the period 1935 to 1939 and 1973, when rates were calculated for a 4-year period and a single year, respectively. These rates were based on the appropriate population for Canada, obtained from the relevant census of Canada data, or on the population estimates published by Statistics Canada. Age standardization was performed with the "European" population for all ages and the "truncated" population for ages 35 to 64 years,4 this group being selected because it was anticipated that any effect of a reduction in cigarette smoking might first be seen in younger people. Mortality rates for the provinces were based on numbers of deaths from lung cancer, obtained from unpublished tabulations supplied to the National Cancer Institute of Canada by Statistics Canada. The same "truncated" population was used in calculating standardized rates for ages 35 to 64 as for the data on Canada. Since 1965, data on cigarette smoking has been published regularly by the health protection branch of Health and Welfare Canada, based on replies to questions in annual sample surveys of the labour force conducted by Statistics
were derived from these data. Results
Trends in mortality for lung cancer in both men and women in Canada, calculated as age-standardized rates, are shown in Fig. 1. Although mortality increased among men from the 1930s, the greatest increase has occurred in the last decade, with no indication so far of a levelling off. In women the increase was slow up to the beginning of the last decade and more rapid thereafter. In the age group 35 to 64 mortality increased throughout the period considered in men, while in women there was hardly any increase until the end of the 1950s, after which it was rapid. In neither sex was there any indication of a decrease. Age-specific rates for men at ages 40 to 44, to 65 to 69 in 5-year age groups are shown in Fig. 2. For no age group other than the oldest was there any indication of a decrease. This is surprising in view of Springett's demonstration in England and Wales that by 1963 mortality was decreasing in all age groups from 40 to 44 up to 55 to 59.1 Instead an inexorable increase is seen for each age group, except ages 65 to 69 in 1973, and this apparent decrease may be an artefact of a single year's calculation. Age-specific rates in terms of birth cohorts are plotted in Fig. 3. With each successive birth cohort there is an increase in mortality except for the group aged 40 to 44 in 1971, which had a
Years Years
FIG. 1-Mortality rates for lung cancer in Canada; age-standardized rates.
28 CMA JOURNAL/JANUARY 8, 1977/VOL. 116
FIG. 2-Changes in age-specific mortality rates for lung cancer in Canada; men, 1935-38 to 1973.
Age
FIG. 3-Mortality rates for lung cancer in Canada; 5-year cohorts of men.
slightly lower mortality than the group aged 45 to 49 in 1971 had at age 40 to 44 in 1966. The slope of the increase in mortality in those aged 45 to 49 in 1971 also suggests that over the next 5-year period this particular age group may show a reduction compared with those aged 50 to 54 in 1971. If so, the findings in this and the preceding age group could be the first indication of the beginning of a reduction in mortality in men in Canada. Examination of age-specific rates in women both within the same age group and by birth cohorts showed no evidence of either a reduction in agespecific rates or a cohort effect, with an increase in mortality in all birth cohorts from the age of 40 or more in the last 10 years. The mortality in women at present is similar to the mortality in men in the early 1940s. Canada is a heterogeneous country and it is conceivable that mortality changes may be occurring to a different extent in different regions of the country. Although the mortality among men
in the age group 35 to 64 over the period 1954-58 to 1969-73 increased in all regions, the increase has been greatest in the last decade in Quebec and the Atlantic provinces, as the upper half of Table I documents. Indeed, Quebec has moved from third to first in mortality in this age group. For the same age group of women, British Columbia had the highest mortality throughout these years, and in all areas of the country mortality has been increasing in the last decade, as the lower half of Table I indicates. Recently among men there has been a reduction in the percentage of regular smokers of cigarettes, with possibly the greatest reduction in Ontario and the least in Quebec, which has the highest rate throughout (Table II). Among women there has been an increase except in Ontario and British Columbia. In contrast, for smokers of 25 cigarettes or more a day (data not tabulated here) the changes have been less striking in men, with a reduction for Canada as a whole from 5.8 to 5.2%, though in the Atlantic region there has been an increase from 4.6 to 4.9%; in women the proportion has increased in all regions, and for Canada as a whole it has increased from 1.3 to 2.0%. Most of the increase in cigarette smoking is occurring among younger women. Discussion Current mortality rates for lung cancer in Canada, both across the country and by region, give little cause for optimism that we are about to see a substantial reduction in one of the most important causes of premature death among men and a rapidly increasing cause of premature death among women. That the increase is not restricted to Caucasians has been strikingly emphasized by the studies of Schaefer and his colleagues6 in Inuit of the Northwest Territories, which have demonstrated a substantial recent increase in mortality from lung cancer, particularly among
women, such that Canadian Inuit women aged 40 to 59 now seem to have one of the highest mortality rates for lung cancer in the world. Whether this is due entirely to cigarette smoking - which, as Schaefer pointed out, has increased greatly in this population or partly to other factors needs further evaluation. Nevertheless, there is some indication from surveys of cigarette smoking that over the next decade a reduction in mortality can be anticipated for men in Ontario and in the Prairies, with possibly a reduction in the rate of increase of mortality for women in these areas. However, there can be little hope for similar changes in Quebec and the Atlantic provinces; in fact, these two regions seem destined to have high mortality rates for lung cancer for some years. If attempts at education of the public have not succeeded greatly in persuading cigarette smokers to take advantage of the reduction in mortality that follows cessation of cigarette smoking,7 more efforts directed to the individual are necessary. Those who have evaluated antismoking clinics have demonstrated some success but not as much as one would have hoped for.8 A relatively easy assay of individual susceptibility seemed likely to arise from the studies of the inducibility of arylhydrocarbon hydroxylase, but this method failed to live up to initial expectations.9 Further endeavours in this area are proceeding. Sputum cytology has yet to be fully evaluated as a means of identifying those with atypical cells who are at increased risk for lung cancer;10 studies are in progress in the United States. It is possible that with better predictive methods, attempts at prevention in individuals may be more effective. Nevertheless, we should not overlook
I
C.j.ada
.' I
CMA JOURNAL/JANUARY 8, 1977/VOL. 116 29
the fact that cigarette smoking is an important risk factor for many more conditions than just cancer of the lung. The incidence of bladder cancer, laryngeal cancer and possibly oral and esophageal cancer is increased in heavy smokers of cigarettes. Furthermore, the toll of smoking-induced respiratory disease and coronary heart disease is well known. It is possible that persons at risk of dying from these conditions if they smoke heavily are not the same as those at risk of dying from lung cancer.1' Thus, increased efforts at prevention of cigarette smoking in the young and reduction of cigarette smoking in adults must be pursued. Canada is not alone in showing substantial increases for women in mortality rates for what is undoubtedly
smoking-induced lung cancer over a decade in which knowledge of the association has been most prevalent. Similar findings have recently been reported from Denmark.12 I thank the staff of the health division of Statistics Canada for supplying the unpublished tabulations used in the comparisons of mortality rates, and Dr. H. N. Colburn of the food and drug directorate ot the health protection branch of Health and Welfare Canada for supplying the data on smoking in Canada.
References 1. SPRINGETr VH: The beginning of the end of the increase in mortality in carcinoma of the lung. Thorax 21: 132, 1966 2. SCHNEIDERMAN MA, LEvIN DL: Trends in lung cancer. Mortality, incidence, diagnosis, treatment, smoking, and urbanization. Cancer 30: 1320, 1972 3. GRAY N, HILL D: Cigarette smoking, tar
content, and death-rates from lung cancer in Australian men (C). Lancet 1: 1252, 1975 4. DOLL R: Comparison between registries, in Cancer Incidence in Five Continents, vol 2, DOLL R, MusR CS, WATERHOUSE JAH (eds), Geneva, Union Internationale contre le Cancer, Springer-Verlag, 1970, pp 333-38 5. HACKLAND SP: Smoking Habits of Canadians, 1965-74, Health and Welfare Canada, nonmedical use of drugs directorate, research bureau, 1975 6. SCHAEFER 0, HILDES JA, MEDD LM, et al: The changing pattern of neoplastic disease in Canadian Eskimos. Can Med Assoc 1 112: 1399, 1975 7. DOLL R, HILL AB: Mortality in relation to smoking: ten years' observations of British doctors. Br Med 1 1: 1399, 1964 8. DELARUE NC: A study of smoking withdrawal: the Toronto smoking withdrawal study centre - description of activities. Can I Public Health 64 (suppi): S5, 1973 9. SHAW CR, KELLERMAN G: Hydroxylase variation and susceptibility to bronchogenic carcinoma in man. Presented at XIth International Cancer Congress, Florence, 1974 10. MILLER AB: Screening for lung cancer. Can I Public Health 64 (suppl): S32, 1973 11. CAPLIN M, FESTENSTEIN F: Relation between lung cancer, chronic bronchitis and airways obstruction. Br Med 1 3: 678, 1975 12. JENSEN OM: Lung cancer and smoking in Danish women. mt I Cancer 15: 954, 1975
Percutaneous radiofrequency lumbar rhizolysis (rhizotomy) JOHN A. MCCULLOCH,* MD, FRCS[C]; L.W. ORGAN,t MD, BA SC
Low back pain may arise from degenerative changes in the posterior joints of the lumbar spine. These joints are innervated by a branch of the posterior primary remus, which follows an anatomically constant course. Pain impulses from these joints can be Interrupted by coagulating this nerve with a radiofrequency wave, the probe having been placed In the area of the nerve percutaneously. Percutaneous lumbar rhizolysis was carried out under local anesthesia on an outpatient basis in 82 patIents, most of whom had multiple level rhizolysis. Rhizolysis was successful In 670/0 of patients with mechanical low back pain without evidence of disc herniatlon and nerve-root compression or psychogenic pain, who had not previously undergone an operation for relief of the pain. Des douleurs lombaires basses peuvent apparaitre a Ia suite d'alt6rations d6g6n6ratives des articulations posterleures de Ia colonne vert6brale. Ces articulations sont innerv6es par une ramification de Ia branche primaire post6rieure, qul suit un parcours anatomique constant. Les impulsions douloureuses provenant de ces articulations peuvent .tre bloquses par coagulation du nerf par radlo-onde, en pla.ant Ia sonde en percutan6e dans Ia r6gion du nerf. From the *departments of surgery and anatomy, University of Toronto and St. Michael's Hospital, and tthe department of physiology, University of Toronto Reprint requests to: Dr. JA. McCulloch, Department of surgery, St. Michael's Hospital, 30 Bond St., Toronto, Ont. M5B 1W8
Une rhizolyse lombaire percutanee a 6te faite sous anesthesie locale en clinique externe chez 82 patients dont plusleurs subirent UN. rhizolyse a differents niveaux. La rhizolyse a ste utilisee avec succes chez 6701o des patients souffrant de douleurs lombaires mecaniques sans signe de hernie discale et de compression de Ia racine nerveuse ou de douleur psychogene, et qui n'avaient pas subi d'operation prealable destinee a soulager Ia douleur.
The most recent proposal for the treatment of backache has come from Rees' and Shealy,2 who described the technique of percutaneous lumbar rhizolysis. On the theory that low back pain may arise from irritated or degenerated lumbar articular facet joints, they proposed that the facet joint be denervated at several levels. Rees produced denervation with a knife cut in the region of the facet joint and reported immediate relief of pain in 998 of 1000 patients with "intervertebral disc syndrome Shealy introduced the procedure to North America but, after encountering a number of large hematomas, chose to induce radiofrequency coagulation through a probe placed in the region of the nerve supply to the facet joint. In one of his few reports on the subject Shealy2 described a "90% success rate in 'virgin' back patients". He suggested that the procedure was useful for true disc herniations and "discogenic pain syndrome". In this paper we describe the procedure and report our results with it in 82 patients.
30 CMA JOURNAL/JANUARY 8, 1977/VOL. 116
Value of facet joint denervation The theory of facet joint denervation is attractive. In a number of patients who underwent lumbar fusion without discectomy and reported immediate relief of pain Macnab (personal communication, 1975) attributed the relief to facet joint denervation. Other surgeons have reported that some types of back pain can be relieved by an injection of local anesthetic into the region of the facet joint.3 The role of the facet joint in the production of low back pain remains unknown. Pedersen, Blunck and Gardner4 first demonstrated the sensory nerve supply to the facet joints in 1956. It then became a simple matter to denervate these joints either with a knife or a radiofrequency current. Denervation procedures alone for the relief of joint pain have not had a lasting good reputation among orthopedic surgeons. First, it is difficult to be sure, especially when doing a procedure percutaneously, that all the nerve supply to a joint has been cut. Probably the facet joint in the lumbosacral region receives branches from more than one nerve root;5 thus the single nerve lesion may not produce total denervation. Second, many structures in the lumbar region can give rise to pain and simple denervation of the facet joint will not affect some of these. Third, total denervation of an extremity or joint will often fail to relieve chronic pain. Fourth, low back pain is peculiar: if the supervising surgeon waits long enough and brings enough factors to bear on a situation (medication, brace support, weight reduction, job change,
In the last decade lung cancer mortality has continued to Increase in men, especially those aged 35 to 64 In the province of Ouebec, and It has Increased strikingly in women. The percentage of regular smokers of cigarettes has decreased recently among men but has increased among women. It is concluded that It may be some years before mortality begins to decrease in either sex, though decreases may be anticipated earliest in men in Ontario, the Prairies and British Columbia. More antismoking efforts directed to individuals are required, especially for young women. Au cours de Ia derniere decennie Ia mortalite reli6e au cancer du poumon a continue d'augmenter chez les hom. mes, particulierement chez ceux de 35 a 64 ans de Ia province de Quebec, et elle a augment6 de fa9on frappante chez les femmes. Le pourcentage de fumeurs reguliers a baiss6 r6cemment chez l'homme mais ii a augment6 chez Ia femme. On conclut qu'il faudra possiblement quelques annees avant que Ia mortalite commence a d6croitre chez les representants des deux sexes. bien qu'une baisse puisse-t-.tre anti. cipee plus t6t pour les hommes de l'Ontario, des Prairies et de Colombie. Britannique. Une lutte accrue contre l'usage du tabac s'adressant aux mdi. vidus et particulierement aux jeunes femmes est requise. In 1966 Springett1 demonstrated from mortality figures in England and Wales that, at least in those countries, the beginning of the end of the increase in mortality from carcinoma of the lung was in sight. In 1972 Schneiderman and Levin2 produced evidence suggesting that similar changes in the United States might be expected shortly, and Gray and Hill3 in 1975 produced evidence of a reduction in mortality among men in Australia. It is thus timely to examine data for Canada to see whether any decrease in mortality has occurred here. Methods Data on numbers of deaths from lung cancer in 5-year age groups are Presented in part at the symposium on oncogenesis, jointly sponsored by the Canadian Society for Clinical Investigation and the Royal College of Physicians and Surgeons of Canada, Quebec City, Jan. 21, 1976 Reprint requests to: Dr. AB. Miller. Director. Epidemiology Unit, National Cancer Institute of Canada, 121 St. Joseph St., Toronto, Ont. M55 2R9
available from the publications of Statistics Canada (formerly the Dominion Bureau of Statistics). Because of changes in the International Classification of Diseases, care was taken to ensure that, as far as possible, a similar grouping was used throughout. Age-specific mortality rates for Canada were calculated for 5-year periods except for the period 1935 to 1939 and 1973, when rates were calculated for a 4-year period and a single year, respectively. These rates were based on the appropriate population for Canada, obtained from the relevant census of Canada data, or on the population estimates published by Statistics Canada. Age standardization was performed with the "European" population for all ages and the "truncated" population for ages 35 to 64 years,4 this group being selected because it was anticipated that any effect of a reduction in cigarette smoking might first be seen in younger people. Mortality rates for the provinces were based on numbers of deaths from lung cancer, obtained from unpublished tabulations supplied to the National Cancer Institute of Canada by Statistics Canada. The same "truncated" population was used in calculating standardized rates for ages 35 to 64 as for the data on Canada. Since 1965, data on cigarette smoking has been published regularly by the health protection branch of Health and Welfare Canada, based on replies to questions in annual sample surveys of the labour force conducted by Statistics
were derived from these data. Results
Trends in mortality for lung cancer in both men and women in Canada, calculated as age-standardized rates, are shown in Fig. 1. Although mortality increased among men from the 1930s, the greatest increase has occurred in the last decade, with no indication so far of a levelling off. In women the increase was slow up to the beginning of the last decade and more rapid thereafter. In the age group 35 to 64 mortality increased throughout the period considered in men, while in women there was hardly any increase until the end of the 1950s, after which it was rapid. In neither sex was there any indication of a decrease. Age-specific rates for men at ages 40 to 44, to 65 to 69 in 5-year age groups are shown in Fig. 2. For no age group other than the oldest was there any indication of a decrease. This is surprising in view of Springett's demonstration in England and Wales that by 1963 mortality was decreasing in all age groups from 40 to 44 up to 55 to 59.1 Instead an inexorable increase is seen for each age group, except ages 65 to 69 in 1973, and this apparent decrease may be an artefact of a single year's calculation. Age-specific rates in terms of birth cohorts are plotted in Fig. 3. With each successive birth cohort there is an increase in mortality except for the group aged 40 to 44 in 1971, which had a
Years Years
FIG. 1-Mortality rates for lung cancer in Canada; age-standardized rates.
28 CMA JOURNAL/JANUARY 8, 1977/VOL. 116
FIG. 2-Changes in age-specific mortality rates for lung cancer in Canada; men, 1935-38 to 1973.
Age
FIG. 3-Mortality rates for lung cancer in Canada; 5-year cohorts of men.
slightly lower mortality than the group aged 45 to 49 in 1971 had at age 40 to 44 in 1966. The slope of the increase in mortality in those aged 45 to 49 in 1971 also suggests that over the next 5-year period this particular age group may show a reduction compared with those aged 50 to 54 in 1971. If so, the findings in this and the preceding age group could be the first indication of the beginning of a reduction in mortality in men in Canada. Examination of age-specific rates in women both within the same age group and by birth cohorts showed no evidence of either a reduction in agespecific rates or a cohort effect, with an increase in mortality in all birth cohorts from the age of 40 or more in the last 10 years. The mortality in women at present is similar to the mortality in men in the early 1940s. Canada is a heterogeneous country and it is conceivable that mortality changes may be occurring to a different extent in different regions of the country. Although the mortality among men
in the age group 35 to 64 over the period 1954-58 to 1969-73 increased in all regions, the increase has been greatest in the last decade in Quebec and the Atlantic provinces, as the upper half of Table I documents. Indeed, Quebec has moved from third to first in mortality in this age group. For the same age group of women, British Columbia had the highest mortality throughout these years, and in all areas of the country mortality has been increasing in the last decade, as the lower half of Table I indicates. Recently among men there has been a reduction in the percentage of regular smokers of cigarettes, with possibly the greatest reduction in Ontario and the least in Quebec, which has the highest rate throughout (Table II). Among women there has been an increase except in Ontario and British Columbia. In contrast, for smokers of 25 cigarettes or more a day (data not tabulated here) the changes have been less striking in men, with a reduction for Canada as a whole from 5.8 to 5.2%, though in the Atlantic region there has been an increase from 4.6 to 4.9%; in women the proportion has increased in all regions, and for Canada as a whole it has increased from 1.3 to 2.0%. Most of the increase in cigarette smoking is occurring among younger women. Discussion Current mortality rates for lung cancer in Canada, both across the country and by region, give little cause for optimism that we are about to see a substantial reduction in one of the most important causes of premature death among men and a rapidly increasing cause of premature death among women. That the increase is not restricted to Caucasians has been strikingly emphasized by the studies of Schaefer and his colleagues6 in Inuit of the Northwest Territories, which have demonstrated a substantial recent increase in mortality from lung cancer, particularly among
women, such that Canadian Inuit women aged 40 to 59 now seem to have one of the highest mortality rates for lung cancer in the world. Whether this is due entirely to cigarette smoking - which, as Schaefer pointed out, has increased greatly in this population or partly to other factors needs further evaluation. Nevertheless, there is some indication from surveys of cigarette smoking that over the next decade a reduction in mortality can be anticipated for men in Ontario and in the Prairies, with possibly a reduction in the rate of increase of mortality for women in these areas. However, there can be little hope for similar changes in Quebec and the Atlantic provinces; in fact, these two regions seem destined to have high mortality rates for lung cancer for some years. If attempts at education of the public have not succeeded greatly in persuading cigarette smokers to take advantage of the reduction in mortality that follows cessation of cigarette smoking,7 more efforts directed to the individual are necessary. Those who have evaluated antismoking clinics have demonstrated some success but not as much as one would have hoped for.8 A relatively easy assay of individual susceptibility seemed likely to arise from the studies of the inducibility of arylhydrocarbon hydroxylase, but this method failed to live up to initial expectations.9 Further endeavours in this area are proceeding. Sputum cytology has yet to be fully evaluated as a means of identifying those with atypical cells who are at increased risk for lung cancer;10 studies are in progress in the United States. It is possible that with better predictive methods, attempts at prevention in individuals may be more effective. Nevertheless, we should not overlook
I
C.j.ada
.' I
CMA JOURNAL/JANUARY 8, 1977/VOL. 116 29
the fact that cigarette smoking is an important risk factor for many more conditions than just cancer of the lung. The incidence of bladder cancer, laryngeal cancer and possibly oral and esophageal cancer is increased in heavy smokers of cigarettes. Furthermore, the toll of smoking-induced respiratory disease and coronary heart disease is well known. It is possible that persons at risk of dying from these conditions if they smoke heavily are not the same as those at risk of dying from lung cancer.1' Thus, increased efforts at prevention of cigarette smoking in the young and reduction of cigarette smoking in adults must be pursued. Canada is not alone in showing substantial increases for women in mortality rates for what is undoubtedly
smoking-induced lung cancer over a decade in which knowledge of the association has been most prevalent. Similar findings have recently been reported from Denmark.12 I thank the staff of the health division of Statistics Canada for supplying the unpublished tabulations used in the comparisons of mortality rates, and Dr. H. N. Colburn of the food and drug directorate ot the health protection branch of Health and Welfare Canada for supplying the data on smoking in Canada.
References 1. SPRINGETr VH: The beginning of the end of the increase in mortality in carcinoma of the lung. Thorax 21: 132, 1966 2. SCHNEIDERMAN MA, LEvIN DL: Trends in lung cancer. Mortality, incidence, diagnosis, treatment, smoking, and urbanization. Cancer 30: 1320, 1972 3. GRAY N, HILL D: Cigarette smoking, tar
content, and death-rates from lung cancer in Australian men (C). Lancet 1: 1252, 1975 4. DOLL R: Comparison between registries, in Cancer Incidence in Five Continents, vol 2, DOLL R, MusR CS, WATERHOUSE JAH (eds), Geneva, Union Internationale contre le Cancer, Springer-Verlag, 1970, pp 333-38 5. HACKLAND SP: Smoking Habits of Canadians, 1965-74, Health and Welfare Canada, nonmedical use of drugs directorate, research bureau, 1975 6. SCHAEFER 0, HILDES JA, MEDD LM, et al: The changing pattern of neoplastic disease in Canadian Eskimos. Can Med Assoc 1 112: 1399, 1975 7. DOLL R, HILL AB: Mortality in relation to smoking: ten years' observations of British doctors. Br Med 1 1: 1399, 1964 8. DELARUE NC: A study of smoking withdrawal: the Toronto smoking withdrawal study centre - description of activities. Can I Public Health 64 (suppi): S5, 1973 9. SHAW CR, KELLERMAN G: Hydroxylase variation and susceptibility to bronchogenic carcinoma in man. Presented at XIth International Cancer Congress, Florence, 1974 10. MILLER AB: Screening for lung cancer. Can I Public Health 64 (suppl): S32, 1973 11. CAPLIN M, FESTENSTEIN F: Relation between lung cancer, chronic bronchitis and airways obstruction. Br Med 1 3: 678, 1975 12. JENSEN OM: Lung cancer and smoking in Danish women. mt I Cancer 15: 954, 1975
Percutaneous radiofrequency lumbar rhizolysis (rhizotomy) JOHN A. MCCULLOCH,* MD, FRCS[C]; L.W. ORGAN,t MD, BA SC
Low back pain may arise from degenerative changes in the posterior joints of the lumbar spine. These joints are innervated by a branch of the posterior primary remus, which follows an anatomically constant course. Pain impulses from these joints can be Interrupted by coagulating this nerve with a radiofrequency wave, the probe having been placed In the area of the nerve percutaneously. Percutaneous lumbar rhizolysis was carried out under local anesthesia on an outpatient basis in 82 patIents, most of whom had multiple level rhizolysis. Rhizolysis was successful In 670/0 of patients with mechanical low back pain without evidence of disc herniatlon and nerve-root compression or psychogenic pain, who had not previously undergone an operation for relief of the pain. Des douleurs lombaires basses peuvent apparaitre a Ia suite d'alt6rations d6g6n6ratives des articulations posterleures de Ia colonne vert6brale. Ces articulations sont innerv6es par une ramification de Ia branche primaire post6rieure, qul suit un parcours anatomique constant. Les impulsions douloureuses provenant de ces articulations peuvent .tre bloquses par coagulation du nerf par radlo-onde, en pla.ant Ia sonde en percutan6e dans Ia r6gion du nerf. From the *departments of surgery and anatomy, University of Toronto and St. Michael's Hospital, and tthe department of physiology, University of Toronto Reprint requests to: Dr. JA. McCulloch, Department of surgery, St. Michael's Hospital, 30 Bond St., Toronto, Ont. M5B 1W8
Une rhizolyse lombaire percutanee a 6te faite sous anesthesie locale en clinique externe chez 82 patients dont plusleurs subirent UN. rhizolyse a differents niveaux. La rhizolyse a ste utilisee avec succes chez 6701o des patients souffrant de douleurs lombaires mecaniques sans signe de hernie discale et de compression de Ia racine nerveuse ou de douleur psychogene, et qui n'avaient pas subi d'operation prealable destinee a soulager Ia douleur.
The most recent proposal for the treatment of backache has come from Rees' and Shealy,2 who described the technique of percutaneous lumbar rhizolysis. On the theory that low back pain may arise from irritated or degenerated lumbar articular facet joints, they proposed that the facet joint be denervated at several levels. Rees produced denervation with a knife cut in the region of the facet joint and reported immediate relief of pain in 998 of 1000 patients with "intervertebral disc syndrome Shealy introduced the procedure to North America but, after encountering a number of large hematomas, chose to induce radiofrequency coagulation through a probe placed in the region of the nerve supply to the facet joint. In one of his few reports on the subject Shealy2 described a "90% success rate in 'virgin' back patients". He suggested that the procedure was useful for true disc herniations and "discogenic pain syndrome". In this paper we describe the procedure and report our results with it in 82 patients.
30 CMA JOURNAL/JANUARY 8, 1977/VOL. 116
Value of facet joint denervation The theory of facet joint denervation is attractive. In a number of patients who underwent lumbar fusion without discectomy and reported immediate relief of pain Macnab (personal communication, 1975) attributed the relief to facet joint denervation. Other surgeons have reported that some types of back pain can be relieved by an injection of local anesthetic into the region of the facet joint.3 The role of the facet joint in the production of low back pain remains unknown. Pedersen, Blunck and Gardner4 first demonstrated the sensory nerve supply to the facet joints in 1956. It then became a simple matter to denervate these joints either with a knife or a radiofrequency current. Denervation procedures alone for the relief of joint pain have not had a lasting good reputation among orthopedic surgeons. First, it is difficult to be sure, especially when doing a procedure percutaneously, that all the nerve supply to a joint has been cut. Probably the facet joint in the lumbosacral region receives branches from more than one nerve root;5 thus the single nerve lesion may not produce total denervation. Second, many structures in the lumbar region can give rise to pain and simple denervation of the facet joint will not affect some of these. Third, total denervation of an extremity or joint will often fail to relieve chronic pain. Fourth, low back pain is peculiar: if the supervising surgeon waits long enough and brings enough factors to bear on a situation (medication, brace support, weight reduction, job change,
