All patients with FIGO Stage IB cervical cancer registered with the Department of Clinical Oncology at the ... Curies in 1898 and a woman with cervix cancer "well.
1992, The British Journal of Radiology, 65, 1018-1024
Stage IB cervical carcinoma: a clinical audit By M. N. Gaze, MRCP (UK), FRCR, C. G. Kelly, MRCP (UK), FRCR, *P. R. C. Dunlop, MD, FRCR, A. T. Redpath, BSc, PhD, G. R. Kerr, MSc and V. J. Cowie, MD, FRCR Department of Clinical Oncology, Western General Hospital, Edinburgh EH4 2XU, UK {Received 2 January 1992 and accepted 6 May 1992) Keywords: Cervical carcinoma, Clinical audit, Radiotherapy, Surgery, Treatment-related morbidity
Abstract. All patients with FIGO Stage IB cervical cancer registered with the Department of Clinical Oncology at the Western General Hospital, Edinburgh, during the 6 years from 1979 to 1984 have been reviewed, as part of a continuing programme of clinical audit. Of the 140 patients with Stage IB disease, 68 (49%) were treated by primary surgery of whom 44 (31%) also received adjuvant radiotherapy. Radical radiotherapy was the definitive treatment for 69 patients (49%). Three patients (2%) were not treated with curative intent. The crude 5-year survival rate for all cases was 72% and the cause-specific 5-year survival rate was 78%. Local tumour control at 5 years was 72%. There was no significant difference in outcome between the surgically treated and irradiated groups of patients. Age, histology and nodal status did not influence outcome. Irradiated patients with bulky tumours fared significantly worse than the other patients who received radical radiotherapy. Multivariate analysis of all patients revealed no significant independent prognostic variables. Primary surgery appears to confer no benefit over radical radiotherapy in terms of either survival or local control. Treatment-related late bladder and bowel morbidity was, however, significantly worse in irradiated patients.
Two new radical treatments for carcinoma of the uterine cervix were being pioneered at the turn of the century. Radium was first used shortly after its discovery by the Curies in 1898 and a woman with cervix cancer "well radiumised" in 1905 was reported cured 8 years later (Abbe, 1913). At the same time, a radical operation for cervical cancer was being developed (Wertheim, 1912). Since then, both surgery and radiotherapy have been important in the management of this disease. In the 6-year period from 1979 to 1984, 140 patients with FIGO Stage IB carcinoma of the uterine cervix were referred to the Regional Department of Clinical Oncology at the Western General Hospital, Edinburgh. The majority of patients were treated by radical hysterectomy or radical radiotherapy. This paper is an audit of the management of these patients in terms of survival, local tumour control and treatment-related morbidity. Possible important prognostic factors in Stage IB carcinoma of the cervix are evaluated and the place of lymphangiography in the assessment of new patients is discussed. Patients and methods
Patients and staging procedure The case notes of all 546 new patients with carcinoma of the uterine cervix registered in the Department of Clinical Oncology at the Western General Hospital, Address correspondence to Dr M. N. Gaze, Department of Radiation Oncology, University of Glasgow, CRC Beatson Laboratories, Alexander Stone Building, Garscube Estate, Glasgow G61 1BD, UK. * Present address: Department of Radiotherapy and Oncology, South Cleveland Hospital, Marton Road, Middlesbrough, Cleveland TS4 3BW, UK.
1018
Edinburgh, from 1979 to 1984 have been reviewed. This audit is of the 140 patients (26%) who were staged as having FIGO Stage IB carcinoma of the cervix. Routine staging investigations included examination under anaesthesia, cystoscopy, intravenous urography and chest radiography. Bipedal lymphangiography was not performed in all patients. Treatment policy Radical hysterectomy and pelvic lymphadenectomy tended to be the preferred treatment for younger, fitter patients. Some patients with bulky or poorly differentiated tumour received pre-operative radiotherapy, usually 20 Gy in 10 fractions over 2 weeks to the true pelvis using megavoltage X rays. If the histology of the excised pelvic lymph nodes revealed involvement with metastatic carcinoma, post-operative pelvic radiotherapy was administered. A dose of 40 Gy in 20 fractions over 4 weeks or its equivalent (including the preoperative treatment if that had been given) was prescribed. In a few patients the diagnosis of invasive carcinoma was made following simple hysterectomy. Most of these received post-operative pelvic radiotherapy. All patients treated surgically are analysed together irrespective of whether or not they received adjuvant pelvic radiotherapy. Patients considered unsuitable for radical hysterectomy were treated with radical radiotherapy. Radiotherapy technique A combination of external-beam therapy and intracavitary caesium was used in all patients. The radiotherapeutic technique changed during the period of this audit. From 1979 to 1981 patients were treated with the wedge technique. Megavoltage external-beam radiation The British Journal of Radiology, November 1992
Stage IB cervical carcinoma Table I. Treatment of patients with FIGO Stage IB cervical carcinoma
Radical hysterectomy and pelvic lymphadenectomy Simple hysterectomy Radiotherapy technique
No radiotherapy
Pre-operative radiotherapy
Post-operative radiotherapy
Pre- and post-operative Total radiotherapy
22
15
12
5
54
2
0
12
0
14
Wedge
Low-dose brick
High-dose brick
25
40
4
Total 69
Total
was administered to the whole pelvis using parallel opposed fields with midline lead shielding to give a dose of 20 Gy to the central pelvis and to point A, and 40 Gy to the pelvic sidewall in 20 fractions over 4 weeks. Intracavitary treatment was given using the Edinburgh line source applicator and a manual afterloading system with caesium sources (Campbell & Douglas, 1966; Redpath et al, 1976). The dose prescribed from intracavitary caesium was 55 Gy to point A in two insertions. From 1981 to 1984 the low-dose brick technique was used. External-beam therapy was given using a threefield brick technique usually prescribing 40 Gy in 20 fractions over 4 weeks with no central shielding. A single intracavitary caesium insertion was prescribed to give 35 Gy to point A. Patients considered to have advanced disease by virtue of abnormal lymphangiograms (UICC Stage IIIB), were treated with the highdose brick technique. A dose of 55 Gy in 30 fractions over 6 weeks was administered by a three-field brick technique to the pelvis followed by a single intracavitary caesium insertion prescribing a dose of 27 Gy to point A (Quilty, 1988). In 1984 a Selectron remote afterloading machine was installed and replaced the manual afterloading technique of Redpath et al (1976). The Edinburgh line source was adapted for use with the Selectron. The loading of the active sources was programmed to reproduce the dosimetry and dose rate at point A of the manual system. Patients treated by radical radiotherapy are analysed together irrespective of the radiotherapy technique used.
137
Results Treatment intent and modality This audit relates to 140 patients with FIGO Stage IB carcinoma of the cervix. Because of serious intercurrent illness or poor general condition, two patients received only palliative treatment and one was untreated. The treatments received by the 137 patients (98%) treated radically are shown in Table I. Age Patients treated surgically had a mean age of 41.2 (range 23-80) years and patients treated by radical radiotherapy had a mean age of 53.3 (range 26-79) years. Patients receiving radical radiotherapy were significantly older than those treated by surgery (t — 5.4, p < 0.0001). Histology Table II shows the distribution of pathological types and tumour grade. Histological confirmation of the diagnosis was available in all cases. Lymph node involvement Bipedal lymphangiography was performed in 84 patients (60%). The results are shown in Table III. Pre-operative lymphangiograms were performed in only 36 of the 54 patients who underwent radical hysterectomy and pelvic lymphadenectomy. Table IV shows the correlation between radiological abnormalities and Table II. Histological type and grade
Follow-up
Data are available until death or the 1989 anniversary of treatment in all except two patients who were lost to follow-up at 1 and 24 months.
Tumour type
Analysis The case notes of all patients were reviewed and standard data were entered into a computer for analysis. Survival curves were calculated by the life table method and compared using the log rank test. Proportional hazards analysis (Cox, 1982) was used to detect independent prognostic variables. The Mann-Whitney test was used to compare the incidence of morbidity in different treatment groups.
Squamous 8 carcinoma Adenocarcinoma 7 Adenosquamous 0 carcinoma Clear cell 0 carcinoma Total 15 (%) (11)
Vol. 65, No. 779
Degree of differentiation Well
Moderate Poor Not Total stated
(%)
28
63
17
116
(83)
2 1
2 2
7 1
18 4
(13) (3)
0
0
2
2
(1)
31 (22)
67 (48)
27 (19)
140 (100)
(100)
1019
M. N. Gaze et al 100
Table III. Results of lymphangiograms Not Normal Abnormal done appearances appearances Patients treated with radical hysterectomy and pelvic lymphadenectomy Patients treated with simple hysterectomy Patients treated with radical radiotherapy Total
18
24
12
Total
90-
54
80-
70-
14
8
3
27
28
14
69
53
55
29
137
60-
50-
40-
-Surgery (68) ' Radiotherapy (69)
30-
histological node involvement. The specificity of a test indicates the proportion of those without the disease who give a negative test result, and the sensitivity indicates the proportion of those with the disease who have a positive test result. The specificity of lymphangiography in detecting abnormal nodes in this group of patients was 73%, although the sensitivity was only 50%. Lymph node histology is available for 56 patients. Of the 54 patients treated by radical hysterectomy and pelvic lymphadenectomy, 42 patients were pathologically node negative and 12 had involved nodes. Two patients underwent laparotomy with a view to radical hysterectomy, but at operation nodal involvement was obvious and lymph node biopsy only was performed. Subsequent treatment was by radical radiotherapy. 42 patients underwent lymphangiography prior to radical radiotherapy. Outcome in the 14 patients with radiological evidence of nodal involvement was not significantly different from that in patients with normal lymphangiograms. Survival For all 140 patients, crude survival at 5 years was 71.8% (95% confidence interval (CI): 64.2-79.4%) and cause-specific survival (obtained by censoring 11 intercurrent deaths) was 77.5% (95% CI: 70.1-84.9%). However, patients continue to die of cervical cancer after 5 years and the cause-specific actuarial survival at 8 years was only 64.1% (95% CI: 50.3-77.9%). There was no significant difference (p = 0.302) in causespecific survival between patients treated with radical
20-
10-
12
Node histology Node histology positive negative Lymphangiogram abnormal Lymphangiogram normal Total
1020
36
48
60
Months
Figure 1. Cause-specific survival for patients treated with surgery and radical radiotherapy.
radiotherapy and those treated surgically (Fig. 1). The local recurrence-free survival curves similarly show no significant difference (p = 0.539) between the two treatment groups (Fig. 2).
70-
605040- Surgery (68) ' Radical radiotherapy (69)
30-
Table IV. Correlation of lymphangiographic appearances with pathological nodal involvement
24
20-
Total 10-
5
7
12
5
19
24
10
26
36
12
24
36
48
60
Months
Figure 2. Local recurrence-free survival for patients treated with surgery and radical radiotherapy. The British Journal of Radiology, November 1992
Stage IB cervical carcinoma Table V. Causes of death Intercurrent illness Patients treated by surgery Patients treated by radiotherapy alone Total
Pelvic disease only
2 9 14
11
Patterns of failure Of the 137 radically treated patients, 84 remain alive 5-10 years after presentation. Causes of death are shown in Table V. No patient died from treatmentrelated morbidity. Prognostic factors There was no discernible difference in survival or local control when the effect of age, histological type, differentiation and radiological and pathological node status were subjected to univariate analysis. Tumour size was not routinely measured, nevertheless the EUA report in 18 of the patients treated with radical radiotherapy describes the lesion as "bulky" or "extensive". Outcome for these patients was significantly worse (p = 0.021) than for the other 51 radically irradiated patients (Fig. 3). Multivariate analysis of the effect of age, cell type, tumour grade, lymphangiographic appearances, nodal status at operation and treatment modality failed to reveal any significant independent prognostic factors.
Total dead
Alive
7 12
16 35
52 34
19
51
86
Metastatic disease only
Tumour bulk could not be included in this analysis as data were not available for most patients.
Morbidity Late bladder and bowel morbidity was graded retrospectively according to the EORTC/RTOG scale Table VI. EORTC/RTOG scoring system for late radiation reactions Bladder Grade 1 Grade 2 Grade 3
Grade 4 Grade 5 Bowel Grade 1 Grade 2 Grade 3 Grade 4 Grade 5
60-
Pelvic and metastatic disease
Slight epithelial atrophy. Minor telangiectasia (microscopic haematuria) Moderate frequency, generalized telangiectasia. Intermittent macroscopic haematuria Severe frequency and dysuria. Severe generalized telangiectasia (often with petechiae). Frequent haematuria. Reduction in bladder capacity ( < 150 ml) Necrosis. Contracted bladder (capacity < 100 ml). Severe haemorrhagic cystitis Death directly related to radiation-induced effects Mild diarrhoea. Mild cramping. Bowel movement less than five times daily. Slight rectal discharge or bleeding Moderate diarrhoea and colic. Bowel movement greater than five times daily. Excessive rectal mucus or intermittent bleeding Obstruction or bleeding requiring surgery Necrosis, perforation or fistula Death directly related to radiation-induced effects
50-
Table VII. Treatment-related morbidity 40-
Grade 30-
Bulky tumours (18) Non-bulky tumours (51)
Late bladder morbidity Radical radiotherapy alone Surgery alone Combined treatment
20-
10-
12
24
36
48
60
Months
Figure 3. Cause-specific overall survival following radical radiotherapy for patients with "bulky" tumours compared with the other irradiated patients (p = 0.021).
Vol. 65, No. 779
Late bowel morbidity Radical radiotherapy alone Surgery alone Combined treatment
Total
0
1
2
3
4
5
54
3
5
3
4
0
69
24 39
0 1
0 0
0 2
0 2
0 0
24 44
47
8
5
4
5
0
69
24 36
0 3
0 1
0 3
0 1
0 0
24 44
1021
M. N. Gaze et al
(Table VI). The incidence of normal-tissue damage in each treatment group is shown in Table VII. Grade 3 or 4 bladder damage occurred in 10% of patients treated by radical radiotherapy compared with 6% of patients treated surgically (z = 2.30, p = 0.02). Grade 3 or 4 bowel damage occurred in 13% of patients treated by radical radiotherapy compared with 6% of patients treated surgically (z = 2.82, p = 0.005). Within the group treated by surgery, no morbidity was documented in patients treated by surgery alone whereas Grade 3 or 4 bladder and bowel damage each occurred in 9% of patients receiving surgery and adjuvant radiotherapy (z = 1.76, p = 0.08 for bladder morbidity, z = 2.27, p = 0.02 for bowel morbidity). There was no difference in the morbidity between the patients receiving radical radiotherapy and those treated by surgery with adjuvant radiotherapy. Discussion Clinical audit is defined as "a systematic critical analysis of the quality of medical care, including the procedures used for diagnosis and treatment, the use of resources, and the resulting outcome for the patient" (H. M. Secretaries of State, 1989). It has recently been enccouraged by the Government, although there is a long tradition of audit in clinical oncology. This dates back to the establishment in 1929 of The National Radium Commission, which made record keeping and submission of annual reports of treatment results a condition for lending radium sources to hospitals for the treatment of malignant disease (Raven, 1990). In oncology, survival has traditionally been the principal measure chosen to reflect outcome of treatment. Other important outcome measures include local control and treatment-associated morbidity. Disease-related factors are, however, not the only aspects of patient management that must be considered. In times of increasing financial accountability, audit takes into account use of available resources and the cost of treatment. Also, in an era of informed consent, patient preference may influence treatment choice. If a cancer or a particular stage of the disease can be treated by different modalities with an equivalent chance of cure, morbidity, patient preference and economics become important factors. An audit of cervical cancer treatment in this department for the period from 1960 to 1976 has been published previously (Brand & Kerr, 1982). The 5-year survival rate was 72.0% for 442 patients with Stage IB disease treated radically by surgery or radiotherapy. Although conclusions drawn from comparison with historical controls can only be tentative, the present crude overall survival rate of 71.3% is remarkably similar. These results are, however, poor compared with those reported from other centres. For patients treated with radical radiotherapy, 5-year survival rates greater than 85% are commonly reported (Fletcher, 1971; Park et al, 1973; Perez et al, 1986). Similar figures are also given for patients treated by surgery (Park et al, 1973; Hoskins 1022
et al, 1976). The reasons for this discrepancy are not clear. There appears to be no difference in survival rates, either in the current series or in the literature, between patients treated with surgery or radiotherapy (Hopkins & Morley, 1991) or between patients treated with radiotherapy alone or combined modalities (Mendenhall et al, 1991). It would be unwise, however, to attempt to draw conclusions about the relative efficacy of surgery and radiotherapy from the present study. It must be emphasized that this is not a randomized trial but an audit of clinical practice in which the two groups of patients are not comparable. Although it has been suggested that there is a more aggressive form of cervical cancer carrying a poorer prognosis emerging in younger patients (Prempree et al, 1983; Lybeert et al, 1987, Elliot et al, 1989), there is no evidence to support this in the present data. Age is not a significant prognostic factor in our series and this is in keeping with other recent reports (Russell et al, 1987; Junor et al, 1989). The proportion of patients with adenocarcinoma (13%) is slightly higher than that reported from other centres (Park et al, 1973; Perez et al, 1986; Alcock & Toplis, 1987), which is from 6 to 8%. Although cervical adenocarcinoma was once thought to be less radiosensitive than squamous carcinoma (Fletcher, 1971) prognosis in our series is independent of cell type, and accords well with the findings of Grigsby et al (1988): Lymphangiography is generally recognized to be an investigation of high specificity (85-100%) but somewhat poorer sensitivity (27-77%) (Piver et al, 1971; de Muylder etal, 1984; Smales etal, 1986). Our figures of 73% and 50% for sensitivity and specificity are not markedly different. These results may not reflect the accuracy of lymphangiography in all patients who underwent the investigation, however, as no pathological data were obtained for those patients treated with radiotherapy because of positive lymphangiograms. Thus the true sensitivity may be greater than indicated. Lymph node status is usually found to be a major prognostic factor, whether assessed radiologically (Smales et al, 1986) or pathologically (Alcock & Toplis, 1987). It is surprising, therefore, that we have not demonstrated either radiological or histological lymph node involvement to be an adverse factor. Prognosis in the surgically treated group was similar whether or not the nodes were histologically involved, and not different from that achieved by radical radiotherapy. The routine use of lymphangiography does not, therefore, seem to be justified. The increase in tumour size that generally occurs along with increasing FIGO stage is recognized to correlate with worsening prognosis. The FIGO system does not take account of tumour volume within each stage, although it has been recognized by Fletcher (1971), Grigsby et al (1988) and Kovalic et al (1991) as well as the present series to be a relevant factor. Most series fail to report measurements of tumour volume, and this is clearly a weakness of the present staging system. The reason for the poorer prognosis in patients with bulky The British Journal of Radiology, November 1992
Stage IB cervical carcinoma
tumours is not clear, as the patterns of failure in these patients are not significantly different from those seen in patients with non-bulky tumours. 19 of 40 patients dying from their malignancy had distant metastases in the absence of local recurrence. This indicates that cervical carcinoma, even when at an apparently early 'stage, may be a systemic disease. Although improvements in local methods of treatment may diminish mortality, effective systemic therapy is needed. The incidence of late morbidity is an important factor in the assessment of results of treatment of cervical carcinoma. However, only limited weight can be given to a retrospective analysis of unclassified clinical observations recorded in case notes. While severe morbidity is likely to be relatively well documented, milder degrees may not be uniformly recorded by different observers. Sismondi etal (1989), surveying 96 papers that refer to complications of cervix cancer treatment, found that 59 made no attempt to grade severity. 22 different classifications were used in the 39 papers in which some sort of scale was used to quantify problems. It is against this background that one must evaluate the observation that significant late bladder and bowel morbidity is documented for patients receiving radiotherapy and no morbidity in patients treated by surgery alone. While our figures for severe radiotherapy-related morbidity are similar to the 11 % national benchmark figure reported in the Patterns of Care Outcome Survey (Hanks etal, 1982), they are greater than is desirable. This relatively high incidence is not accounted for by the excessive late radiation damage noted in patients treated with the high-dose brick technique (Quilty, 1988). This audit has all the deficiencies inherent in any retrospective analysis. For example, some important data were missing from the case notes, pathological and radiological findings were not uniformly reported and tumour size, which may be a particularly important prognostic factor, was not systematically recorded. In addition, radiotherapy techniques changed and morbidity was not always documented. To try to improve the clinical service and standards of care for our patients, a new practice has evolved in Edinburgh. We attempt to ensure that all new patients with carcinoma of the cervix, irrespective of where in south-east Scotland they may have presented, are jointly assessed together with a central review of pathology in a combined gynaecological/oncology clinic. This is staffed by consultant gynaecological, radiation and medical oncologists, all with special expertise and experience in the management of gynaecological malignancy. Protocols for staging, treatment and follow-up have been agreed. This enables multidisciplinary discussion with a view to optimal and consistent management of all patients. Prospective recording of clinical and pathological data, results of staging investigations and full treatment details are performed. At each follow-up attendance, standardized documentation of bowel, bladder and sexual function and morbidity and disease status is made. This will ensure that future audits are more reliable and that patient care is improved.
Vol. 65, No. 779
Acknowledgments We wish to thank Dr J. R. B. Livingstone, Dr S. M. Ludgate and Dr G. E. Smart under whose care the majority of these patients were treated.
References ABBE, R., 1913. The use of radium in malignant disease. Lancet, ii, 524-527. ALCOCK, C. J. & TOPLIS, P. J., 1987. The influence of pelvic
lymph node disease on survival for stage I and II carcinoma of the cervix. Clinical Radiology, 38, 13-16. BRAND, M. A. J. T. & KERR, G. R.,
1982. The results of
treatment of carcinoma of the uterine cervix using a linear vaginal source and 4MV X rays. British Journal of Radiology, 55, 352-355. CAMPBELL, E. M. & DOUGLAS, M., 1966. The treatment of
carcinoma of the cervix using a linear vaginal source and 4MeV X rays. British Journal of Radiology, 39, 537-546. Cox, D. R., 1982. Regression models and life tables. Journal of the Royal Statistical Society Series, 34, 187-220. DE
MUYLDER, X., BELANGER, R., VAUCLAIR, R., AUDET-LAPOINTE, P., CORMIER, A. & METHOT, Y., 1984.
Value of lymphography in stage IB cancer of the uterine cervix. American Journal of Obstetrics and Gynecology, 148, 610-613. ELLIOT, P. M., TATTERSALL, M. H. N., COPPLESON, M., RUSSELL, P., WONG, F., COATES, A. S., SOLOMON, H. J., BANNATYNE, P. M., ATKINSON, K. H. & MURRAY, J. C ,
1989. Changing character of cervical cancer in young women. British Medical Journal, 298, 288-290. FLETCHER, G. H., 1971. Cancer of the uterine cervix. American Journal of Roentgenology, 111, 225-242. GRIGSBY, P. W., PEREZ, C. A., KUSKE, R. R., CAMEL, H. M., KAO, M. S., GALAKATOS, A. E. & HEDERMAN, M. A., 1988.
Adenocarcinoma of the uterine cervix: lack of evidence for a poor prognosis. Radiotherapy and Oncology, 12, 289-296. HANKS, G. E., KRAMER, S., DIAMOND, J. J. & HERRING, D. F.,
1982. Patterns of care outcome survey: national outcome data for six disease sites. American Journal of Clinical Oncology, 5, 349-353. H. M. SECRETARIES OF STATE FOR HEALTH, WALES, NORTHERN
IRELAND
AND SCOTLAND,
1989. Working for
Patients
(HMSO, London), p. 39. HOPKINS, M. P. & MORLEY, G. W., 1991. Radical hysterectomy
versus radiation therapy for Stage IB squamous cell carcinoma of the cervix. Cancer, 68, 212-211. HOSKINS, W. J., FORD, J. H., LUTZ, M. H. & AVERETTE, H. E.,
1976. Radical hysterectomy and pelvic lymphadenectomy for the management of early invasive cancer of the cervix. Gynecologic Oncology, 4, 278-290. JUNOR, E. J., SYMONDS, R. P., WATSON, E. R. & LAMONT,
D. W., 1989. Survival of younger cervical carcinoma patients treated by radical radiotherapy in the West of Scotland 1964-1984. British Journal of Obstetrics and Gynaecology, 96, 525-528. KOVALIC, J. J., PEREZ, C. A., GRIGSBY, P. W. & LOCKETT,
M. A., 1991. The effect of volume of disease in patients with carcinoma of the uterine cervix. International Journal of Radiation Oncology, Biology, Physics, 21, 905-910. LYBEERT,
M. L. M.,
MEERWALDT,
J. H.
&
VAN PUTTEN,
W. L. J. 1987. Age as a prognostic factor in carcinoma of the cervix. Radiotherapy and Oncology, 9, 147-151. MENDENHALL, W. M., MCCARTY, P. J., MORGAN, L. S., CHAFE,
W. E. & MILLION, R. R., 1991. Stage IB or IIA-B carcinoma of the intact uterine cervix > 6cm in diagmeter: is adjuvant
1023
M. N. Gaze et al extrafascial hysterectomy beneficial? International Journal of Radiation Oncology, Biology, Physics, 21, 899-904. PARK, R. C , PATOW, W. E., ROGERS, R. E. & ZIMMERMAN,
E. A., 1973. Treatment of stage I carcinoma of the cervix. Obstetrics and Gynecology, 41, 117-122. PEREZ, C. A., CAMEL, H. M., KUSKE, R. R., KAO, M.-S., GALAKATOS, A., HEDERMAN, M. A. & POWERS, W. E., 1986.
Radiation therapy alone in the treatment of carcinoma of the uterine cervix: a 20-year experience. Gynecologic Oncology, 23, 127-140. PIVER, M. S., WALLACE, S. & CASTRO, J. R., 1971. The accuracy
of lymphangiography in carcinoma of the uterine cervix. American Journal of Roentgenology, Radium Therapeutics and Nuclear Medicine, 111, 278-283. PREMPREE, T., PATANAPHAN, V., SEWCHAND, W. & SCOTT,
R. M., 1983. The influence of patients' age and tumour grade on the prognosis of carcinoma of the cervix. Cancer, 51, 1764-1771. QUILTY, P. M., 1988. A report of late rectosigmoid morbidity in patients with advanced cancer of the cervix, treated by a six week pelvic brick technique. Clinical Radiology, 39, 297-300.
1024
RAVEN, R. W., 1990. The Theory and Practice of Oncology: Historical Evolution and Present Principles (Parthenon Publishing, Carnforth), p. 232. REDPATH, A. T., DOUGLAS, M. A. J. T. & ORR, J. A., 1976. The
Edinburgh afterloading technique for carcinoma of the uterine cervix. British Journal of Radiology, 49, 963-965. RUSSELL, J. M., BLAIR, V. & HUNTER, R. D., 1987. Cervical
carcinoma: prognosis in younger patients. British Medical Journal, 295, 300-303. SlSMONDI, P . , SlNISTRERO, G . , ZOLA, P., VOLPE, T., FERRARIS, R., CASTELLI, G. L. & GIAI, M., 1989. Complications of
uterine cervix carcinoma treatments: the problem of a uniform classification. Radiotherapy and Oncology, 14, 9-17. SMALES, E., PERRY, C. M., MACDONALD, J. S. & BAKER, J. W.,
1986. The value of lymphography in the management of carcinoma of the cervix. Clinical Radiology, 37, 19—22. WERTHEIM, E., 1912. The extended abdominal operation for carcinoma uteri. American Journal of Obstetrics, 66, 169-232.
The British Journal of Radiology, November 1992
