(years). 0 (n=77). 1-10 (n=29). 11-20 (n=41). >20 (n=20). Wartvirusinfection(n=87). 23 5. 44. 18. 16. 8. Cervical intraepithelial neoplasia: Grade I (n=34). 22-6. 15.
or regression of the lesion is not significantly affected by smoking (table below).
Epidemiologicalfeatures of patients with cervical abnormalities
Wartvirusinfection(n=87) Cervical intraepithelial neoplasia: Grade I (n=34) GradeII(n=26) Grade III (n=24)
No of cigarettes smoked daily
Mean age (years)
0 (n=77)
1-10 (n=29)
11-20 (n=41)
>20 (n=20)
23 5
44
18
16
8
22-6 245 27-3
15 11 7
5 4 2
8 8 9
4 2 6
V HARINDRA P SRISKANDABALAN A H DE SILVA Royal Bournemouth Hospital, Bournemouth BH7 7DW 1 Burger MPM, Hollema H, Gouw ASH, Pieters WJLM, Quint WGV. Cigarette smoking and human papiliomavirus in patients with reported cervical cytological abnormality. BMJ 1993;306:749-52. (20 March.)
smoked, suggests that a direct mutagenic effect of cigarette smoking on cervical epithelium should not be discounted. Although the mechanism is not clear, our findings agree with others showing a relation between high grade cervical intraepithelial neoplasia, age, and smoking more than 10 cigarettes a day but no relation between cervical intraepithelial neoplasia and number of sexual
1 Burger MPM, Hollema H, Gouw ASH, Pieters WJLM, Quint WGV. Cigarette smoking and human papillomavirus in patients with reported cervical cytological abnormality. BMJ 1993;306:749-52. (20 March.) 2 Winkelstein W. Smoking and cervical cancer-current status: a review. AmJEpidemiol 1990;131:945-57. 3 Layde PM. Smoking and cervical cancer: cause or coincidence.
JAMA 1989;261:1631-3.
4 Davey Smith G, Phillips AN. Declaring independence: why we should be cautious. J Epidemiol Community Health 1990;44: 257-8. 5 Halsey NA, Coberly JS, Holt E, Coreil J, Kissinger P, Moulton LH, et al. Sexual behaviour, smoking and HIV-1 infection in Haitian women. JAMA 1992;267:2062-6.
partners.2
As the evidence linking smoking to high grade cervical intraepithelial neoplasia and cervical cancer increases we believe that doctors should give advice on stopping smoking to all women attending for cervical cytological examination or colposcopy.
Cervical epithelium vulnerable in smokers EDITOR,-M P M Burger and colleagues provide further evidence that cigarette smoking is a risk factor for cervical neoplasia but did not find a relation between the number of cigarettes smoked daily and the grade of cervical intraepithelial neoplasia.' Our data differ from this. We performed a study to determine the relation between cigarette smoking and cervical abnormalities on histological examination. Altogether 171 patients were found to have abnormalities on colposcopic biopsy between 1 October 1988 and 31 September 1989. Of these, 87 had wart virus infection alone and 84 had cervical intraepithelial neoplasia (34 grade I, 26 grade II, and 24 grade III). There was a significant difference in mean age between those with wart virus infection alone (23-5 years) and those with cervical intraepithelial neoplasia grade III (27-3 years) (Student's t=3-22, p=0 01; table above). Epidemiological data were elicited by direct questioning before colposcopy. There was no significant difference among the groups in age at menarche, age at which intercourse first occurred, or number of sexual partners. Of the 167 patients who gave details of their smoking history, 90 were regular smokers. Smoking more than 10 cigarettes a day was associated with high grade cervical intraepithelial neoplasia (grade II or III) compared withwartvirus infection alone(X2=5*2, p=0 0226). There was a dose-response relation between the number of cigarettes smoked daily and grade of cervical intraepithelial neoplasia (XI for trend=3-949, p=00469). This trend was also significant when wart virus infection alone was included in the analysis (X2 for trend=4-631, p=0-0314). Our patients were much younger than Burger and colleagues' (mean age 24 v 34-4 years), and thus there had been less time for smoking to have a cumulative effect on the immunity of cervical epithelium. This, coupled with our finding of an association between the grade of cervical intraepithelial neoplasia and number of cigarettes
KAREN E ROGSTAD CAROL DIXON
I H AHMED-JUSHUF Department of Genitourinary Medicine, City Hospital, Nottingham NG5 IPB 1 Burger MPM, Hollema H, Gouw ASH, Pieters WJLM, Quint WGV. Cigarette smoking and human papillomavirus in patients with reported cervical cytological abnormality. BMJ 1993; 306:749-52. (20 March.) 2 Evans BA, Bond RA, MacRae KD. A colposcopic case-control study of cervical squamous intraepithelial lesions in women with angogenital warts. Geniwurinar Medicine 1992;68:300-4.
Disease progression unrelated to smoking EDITOR,-For the past seven years we have screened patients with genital warts colposcopically and histologically at intervals of six to eight months to study the progression or regression of infection with human papillomavirus and mild cervical intraepithelial neoplasia (grade I). We have screened 138 patients, of whom 50 had only infection with human papillomavirus and 88 had infection with human papillomavirus and grade I cervical intraepithelial neoplasia. We did not find any significant relation between the change in these conditions over time and the number of cigarettes smoked or the duration of cigarette smoking. We agree with M P M Burger and colleagues that cigarette smoking does not have any direct mutagenic effect on the cervical epithelium.' Progression or regression of these conditions in our patients did not seem to depend on the age at which intercourse first occurred, number of sexual partners, or use of contraceptive pill. From our study it seems that the progression to higher grade of cervical intraepithelial neoplasia or cervical carcinoma in patients with genital warts may depend on the oncogenicity of the human papillomavirus rather than on any associated risk behaviour. Cigarette smoking may initiate early changes in the cervical epithelium, but progression
Mycobacteria and sarcoidosis Clinical studies support link EDITOR,-Helen M Fidler and colleagues' paper adds to the growing number ofmolecular biological studies reporting the finding of mycobacterial DNA (or RNA) in material from similar proportions of patients with diagnoses of mycobacterial tuberculosis and of sarcoidosis. The authors refer to epidemiological, immunological, and microbiological evidence of a relation between sarcoidosis and mycobacterial infection but do not mention the evidence provided by long term clinical studies.23 These report many cases in which clinical and histological features of sarcoidosis and of caseating tuberculosis occur in the same patient in various temporal relations. In these, the manifestations of mycobacterial infection may be atypical and respond slowly to antimycobacterial treatment to which the organisms have been shown to be sensitive in vitro, sometimes only after corticosteroids have been added. Discussion of these issues has been hampered by some widely adopted definitions of sarcoidosis, which start with the declaration that it is a disease of unknown cause, especially if this is reinforced by a specific instruction that mycobacterial and fungal infections must be excluded. A nominalist definition based on clinical and histological features, leaving reference to uncertainty about causation to be part of the description of the diagnostic category so defined, imposes no restraint on discussion of the implications of evidence of the involvement of any agent, either already known or newly identified.3 It is then permissible to refer to cases of sarcoidosis in which evidence of the involvement of a mycobacterium is found as "mycobacterial sarcoidosis" without implying that all cases are of this sort. Although I have drawn attention for many years to the evidence favouring the involvement of mycobacteria in at least some cases of sarcoidosis, I think that advocacy of a large controlled trial of modem antituberculous chemotherapy for sarcoidosis is premature. The course of sarcoidosis is so variable that a large number of patients would have to be recruited to show any effect within the reasonably expectable range, and recruitment of a valid control group would be difficult. A new trial of antimycobacterial chemotherapy in sarcoidosis should await further evidence of the presence of the postulated cell wall deficient or other variant of mycobacteria in at least a high proportion of cases and of the effectiveness in in vitro and animal
Progression or regression of hwnan papillomavirus (HPV) infection and grade I cervical intraepithelial neoplasia over six to eight months in 138 patients according to smoking habit Grade I cervical intraepithelial neoplasia initially
Human papillomavirus infection initially
Cervical intraepithelial neoplasia
Cervical intraepithelial neoplasia No
abnormality Non-smokers Smokers: 5years
BMJ VOLUME 306
HPV infection
abnormality
HPV infection
Grade I
Grade II
Total
16
10
11
64
5 3 7 3
1 9 7
3 2 1
14 8 31 21
No Grade I
Grade II
Grade III
1
19
1
3
3
1
1 1 6 7
2 1 1
4
1 1 3 2
8MAY1993
2
1269
