Swedish gynecologists

Transcription

1 Acta Obstetricia et Gynecologica. 2009; 88: ORIGINAL ARTICLE Attitudes to mode of hysterectomy Swedish gynecologists a survey-based study among PÄR PERSSON 1,2, THOMAS HELLBORG 1, JAN BRYNHILDSEN 1, MATS FREDRIKSON 3 & PREBEN KJØLHEDE 1 1 Department of Clinical and Experimental Medicine, Division of Obstetrics and Gynecology, Faculty of Health Sciences, University Hospital, Linköping, Sweden, 2 Department of Obstetrics and Gynecology, University Hospital, Uppsala, Sweden, 3 Department of Clinical and Experimental Medicine, Division of Occupational and Environmental Medicine, Faculty of Health Sciences, University Hospital, Linköping, Sweden Abstract Objective. To determine gynecologists attitudes to mode of hysterectomy on benign indication. Design. Cross-sectional study. Setting. Sweden. Population. Members of the Swedish Society of Obstetrics and Gynecology. Methods. A postal questionnaire. Questions examined attitudes to mode of hysterectomy based on three clinical scenarios with different conditions of the uterus. Gynecologists were also asked to estimate how the distribution of the different modes of benign hysterectomy should be overall. The modes to choose were total abdominal, subtotal abdominal, laparoscopic or vaginal hysterectomy (VH). Analyses were performed with multiple logistic regression and multivariate analysis of covariance. Main outcome measures. Preferred mode of hysterectomy in the three scenarios and distribution of modes of hysterectomy. Results. VH was the most preferred method in general as well as when the uterus was of normal size, whereas subtotal and total abdominal hysterectomy were the most favored methods when the uterus was enlarged. VH was more often preferred by male compared to female gynecologists as a personal preference. The choice and distribution of mode varied significantly between place of work, seniority and in the quantity of yearly performed hysterectomies. The minimal invasive methods, vaginal and laparoscopic hysterectomy, were recommended in more than 50% of the overall suggested distribution. Conclusion. Personal choice of mode of hysterectomy does not seem to strictly follow evidence-based recommendations, but varies significantly between gynecologist s gender, type of clinical setting in which the gynecologist works, seniority and by how many hysterectomies the gynecologist does annually. Key words: Attitudes, gynecology, hysterectomy, survey Introduction Hysterectomy is the most common major gynecological operation. The rate of hysterectomy for benign conditions varies widely in the industrialized world. In 2003, 169/100, 000 women had a hysterectomy in Sweden (1). Hysterectomy is carried out on benign and malignant indications and even for obstetrical reasons (2). There are no generally accepted recommendations as to which mode to use for hysterectomy on benign indications. The scientific background for recommendations is incomplete. Recently, a large review was published concerning the choice of mode of hysterectomy for benign gynecological disease (3). Vaginal hysterectomy (VH) was recommended when possible instead of abdominal hysterectomy because of shorter duration of hospital stay and fewer postoperative complications. In cases where VH was not possible, laparoscopic hysterectomy should be considered as an alternative to abdominal hysterectomy. The authors concluded that research with randomized controlled studies and long-term follow-up was required. Little research has been published on the attitudes of gynecologists regarding mode of hysterectomy on benign indication. In a Danish survey the abdominal Correspondence: Pär Persson, Department of Obstetrics and Gynecology, University Hospital, Uppsala, Sweden. par.persson@akademiska.se (Received 21 August 2008; accepted 25 November 2008) ISSN print/issn online # 2009 Informa UK Ltd. (Informa Healthcare, Taylor & Francis AS) DOI: /

2 268 P. Persson et al. methods were preferred, with abdominal subtotal hysterectomy as the mostly preferred method (4). In two surveys from England and USA (5,6), selecting between total and subtotal abdominal hysterectomy, a majority of the gynecologists preferred the total abdominal mode. The aims of this study were to investigate attitudes toward mode of benign hysterectomy among Swedish gynecologists and to analyze whether differences exist with respect to gender of the gynecologist, years in the specialty, place of work and the surgical experience of the gynecologist in relation to the quantity of hysterectomies done yearly. Material and methods A postal questionnaire with an informative letter was sent to members of the Swedish Society of Obstetrics and Gynecology with a private address in Sweden. A single reminder letter was sent to those who did not respond within eight weeks. In order to display present clinical practice in Sweden, the results of the questionnaire comprised only the members who stated that they were clinically active. The questionnaire consisted of 36 questions. The participants were asked about gender, age, years within the specialty, employment, employer (public or private practice), type of hospital or clinic, the skill to perform hysterectomy and the annual number of hysterectomies done by the individual gynecologist. Twelve questions addressed the personal choice of mode of hysterectomy for male gynecologists regarding their spouse, and for female gynecologists for themselves. The questions were repeated in each of three scenarios with different clinical conditions in which hysterectomy was indicated due to benign gynecological disease: Scenario A. Normal to slightly enlarged uterus, no uterine descensus and no previous cervical dysplasia. Scenario B. Normal to slightly enlarged uterus, no uterine descensus but previous treatment of cervical dysplasia up to moderate degree (CIN II). Scenario C. Enlarged uterus (larger than gestational week 1213), no uterine descensus and no previous cervical dysplasia. The gynecologists were asked to choose between total abdominal hysterectomy (TAH), subtotal abdominal hysterectomy (SAH), laparoscopic or laparoscopically assisted hysterectomy (LH) and VH. The participants were also requested to estimate from their personal view how the overall distribution of the four modes of hysterectomy should be in Sweden for all hysterectomies on benign indications. The answer was to be given in percentage for each mode with a sum of 100%. The questionnaire was validated with a readability and context validation by eight gynecologists of varying seniority in the specialty. The comments and suggestions of these readers were considered and consensus obtained for the final version of the questionnaire. The study was approved by the regional ethics board of Linköping University. Statistical analysis Statistical analyses were carried out in the software StatView(r) for Windows, Copyright(c), , Version (SAS Institute Inc., SAS Campus Drive, Cary, NC 27513, USA). Nominal effect variables were analyzed by means of multiple logistic regression models. Consequently, results were presented as odds ratios and 95% confidence intervals. The suggested distribution of each of the four modes of hysterectomy was compared by using multivariate analyses of covariance (MANCOVA) models followed by individual ANCOVAs. Subsequent post hoc testing was done with Fisher s PLSD test. In the multivariate models, adjustments were carried out simultaneously for gender of the gynecologists, years in the specialty, number of hysterectomies performed yearly by the gynecologist and type of working place. A 5% level of significance was accepted. Results The response rate was 65% (1,010/1,556). The study group consisted of the 840 (83%) who reported that they were clinically active. The responders did not always answer all questions. The numbers and percentages in the text and tables comprise only those who answered the specific question. The descriptive data of the responders are shown in Table I. The proportion of female and male gynecologists was comparable to the non-responders (data not shown). Otherwise no analyses were performed with regard to the non-responders. Of the clinically active gynecologists 573 (73%) did or had done hysterectomies independently. Among the methods they practiced independently, TAH was performed by 90% (517/573), SAH by 87% (498/573), VH by 53% (301/573) and LH by 20% (117/573). Three hundred and ninety gynecologists presently performed hysterectomy. The median number carried out in this group was 15/year (range 1160); 82% (319/390) did ]10 yearly and 8.7% (34/390) operated 40 hysterectomies/year. The distribution of the recommended/preferred modes of hysterectomy in the three scenarios and the

3 Gynecologists attitudes to mode of hysterectomy 269 Table I. Descriptive data of the 840 clinically active gynecologists of the Swedish Society of Obstetrics and Gynecology. Specific question characteristics Number and (%) or median and (range) Gender Female gynecologists 507 (60%) Male gynecologists 333 (40%) Age All 51 (2773) Female 47 (2769) Male 56 (2973) Years in specialty All 18 (144) Female 14 (144) Male 25 (1.540) Seniority B15 years in specialty 324 (39%) ]15 years in specialty 507 (61%) Place of work Outpatient clinic 219 (27%) County hospital 112 (14%) Central hospital 218 (27%) University hospital 263 (32%) All gynecologists did not answer all questions. The figures in the table encompass those who answered the specific question. associations with gender, years in the specialty, number of individually performed hysterectomies/ year and place of work are depicted in Table IIAC. The minimal invasive methods (i.e. VH and LH) were the most favored methods being recommended in 5558% of the cases when the uterus was of normal size or slightly enlarged (scenarios A and B). The multivariate logistic regression analysis showed that the recommendation of VH was significantly positively associated with the number of hysterectomies done yearly by the gynecologist (the more hysterectomies per year, the more he/she will favor use of VH). Gender was in cases without previous cervical dysplasia (scenarios A and C) also an independent factor for favoring VH and male gynecologists were significantly more likely to favor VH than female gynecologists. The abdominal methods (i.e. TAH and SAH) were chosen in 83% when the uterus was considerable enlarged (scenario C). Number of hysterectomies/year was the only factor that was positively associated with the recommendation of TAH in case of a substantially enlarged uterus (scenario C). The recommendation of TAH was significantly positively associated with years in the specialty when uterus was of normal size or slightly enlarged. SAH and TAH were preferred in almost equal proportions (45% vs. 38%) when the uterus was enlarged and without cervical dysplasia. When the uterus was of normal size or slightly enlarged and without previous cervical dysplasia, SAH was favored by more than three times as many as TAH (27% vs. 8%). In contrast, when cervical dysplasia had been present (scenario B) the vast majority recommended a total hysterectomy. There was no association between those who had preference for VH in the normal sized uterus without previous cervical dysplasia and place of work, whereas the association was significant for those who chose SAH. Those who recommended SAH were predominantly from county and central hospitals and less often from outpatient clinics and university hospitals. Preference of TAH was strongly negatively associated with outpatient clinics. The outpatient clinic gynecologists favored TAH less than gynecologists in the other three hospital settings. Concerning the scenario with the enlarged uterus, no associations were found between TAH or SAH and place of work. The preference of LH in cases with normal to slightly enlarged uterus, independent of occurrence of previous cervical dysplasia, was significantly positively associated with outpatient clinic and university hospital gynecologists meaning that gynecologists in these settings did significantly more often favor LH than those in county and central hospitals. The group who stated that the mode of hysterectomy did not matter or they could not decide, revealed a significant negative association with number of hysterectomies/year, indicating that those gynecologists did less hysterectomies annually than those who had decided about method. Otherwise no associations were found between this group and gender, years in specialty or place of work. Four hundred and eighty-five gynecologists completed the questionnaire concerning suggested distribution of modes of hysterectomy on benign indications. Table III shows the distribution in relation to gender and place of work. In general, VH had the highest rating in all institutions. The least favored method was LH except in university settings. The minimally invasive methods constituted about 55% and the abdominal methods approximately 45%. Initially the MANCOVA was performed with full interaction effects. Since no interaction terms were significant, the final analyses included interactions up to the depth of one. This revealed that there was a difference in preference of mode of hysterectomy between the places of work (Pillai Trace p B0.0001), the number of years in the specialty and number of hysterectomies done yearly (Pillai Trace p and p , respectively), but not between male and female gynecologists (Pillai Trace p ). The ANCOVAs that followed the MANCOVAs showed that place of work was significantly associated with preference for LH (pb0.0001) and VH (p ), but not for TAH and SAH. Number of years in the specialty was significantly positively associated with preference for SAH (p )

4 Table II. Distribution of the recommended mode of hysterectomy by the male gynecologist to his spouse or the preferred mode of hysterectomy of a female gynecologist for herself and associated factors in the three clinical scenarios. Multiple logistic regression. Factor associated with recommended/preferred mode of hysterectomy Distribution Gender ß/à Years in specialty No. of annual hysterectomies Place of work Mode of hysterectomy N and (%) OR and (95% CI) OR and (95% CI) OR and (95% CI) Type OR and (95% CI) (A) Clinical condition: Normal or slightly enlarged uterus, no uterine descensus and no previous cervical dysplasia. TAH A Reference B 1.49 ( ) 56 (8%) 0.89 ( ) 1.04 ( ) 1.01 ( ) C 2.25 ( ) D 2.51 ( ) SAH A Reference B 2.14 ( ) 198 (27%) 0.76 ( ) 1.01 ( ) 0.98 ( ) C 1.96 ( ) D 1.04 ( ) VH A Reference B 0.95 ( ) 308 (41%) 1.57 ( ) 0.97 ( ) 1.02 ( ) C 0.93 ( ) D 0.64 ( ) LH A Reference B 0.21 ( ) 107 (14%) 0.91 ( ) 1.01 ( ) 1.00 ( ) C 0.19 ( ) D 1.04 ( ) Doesn t matter or can t decide A Reference 75 (10%) 0.70 ( ) 1.00 ( ) 0.97 ( ) B 1.08 ( ) C 1.06 ( ) D 1.34 ( ) Missing answer 96 (11%) 270 P. Persson et al. (B) Clinical condition: Normal or slightly enlarged uterus, no uterine descensus and previous treatment of cervical dysplasia of at most moderate degree (CIN II). TAH A Reference B 1.98 ( ) 246 (33%) 0.90 ( ) 1.01 ( ) 0.98 ( ) C 2.16 ( ) D 1.72 ( ) SAH A Reference B 1.96 ( ) 21 (3%) 1.13 ( ) 1.03 ( ) 0.97 ( ) C 0.49 ( ) D 1.78 ( ) VH A Reference B 0.94 ( ) 369 (49%) 1.17 ( ) 0.98 ( ) 1.02 ( ) C 0.86 ( ) D 0.54 ( ) LH A Reference B 0.12 ( )

5 Table II (Continued) Factor associated with recommended/preferred mode of hysterectomy Distribution Gender ß/à Years in specialty No. of annual hysterectomies Place of work Mode of hysterectomy N and (%) OR and (95% CI) OR and (95% CI) OR and (95% CI) Type OR and (95% CI) 70 (9%) 1.13 ( ) 1.00 ( ) 1.00 ( ) C 0.12 ( ) D 0.99 ( ) Doesn t matter or can t decide A Reference 46 (6%) 0.64 ( ) 1.02 ( ) 0.97 ( ) B 0.65 ( ) C 1.31 ( ) D 1.36 ( ) Missing answer 88 (10%) (C) Clinical condition: Enlarged uterus (larger than gestational week 1213), no uterine descensus and no previous cervical dysplasia. TAH A Reference B 1.30 ( ) 288 (38%) 0.94 ( ) 1.00 ( )) 1.01 ( ) C 1.03 ( ) D 1.27 ( ) SAH A Reference B 1.24 ( ) 348 (45%) 0.89 ( ) 1.01 ( ) 0.99 ( ) C 1.21 ( ) D 0.90 ( ) VH A Reference B 0.15 ( ) 42 (6%) 3.29 ( ) 0.98 ( ) 1.02 ( ) C 0.42 ( ) D 0.33 ( ) LH A Reference B 0.71 ( ) 39 (5%) 1.24 ( ) 0.98 ( ) 1.00 ( ) C 0.95 ( ) D 2.24 ( ) Doesn t matter or can t decide A Reference 44 (6%) 0.65 ( ) 0.98 ( ) 0.93 ( ) B 1.18 ( ) C 1.54 ( ) D 0.83 ( ) Missing answer 79 (9%) LH: laparoscopic (assisted) hysterectomy. TAH: total abdominal hysterectomy. SAH: subtotal abdominal hysterectomy. VH: vaginal hysterectomy. AOutpatient clinic; BCounty hospital; CCentral hospital; DUniversity hospital. ORs adjusted for gender of the gynecologists, years in the specialty, number of hysterectomies performed yearly by the gynecologist and type of working place. Gynecologists attitudes to mode of hysterectomy 271

6 272 P. Persson et al. Table III. The suggested overall distribution, in percentage, of the modes of benign hysterectomy by 485 clinically active gynecologists in relation to gender and place of work of the gynecologists. Gender of gynecologists Place of work of gynecologists Mode of hysterectomy All à ß Outpatient clinics County hospitals Central hospitals University hospitals TAH SAH VH LH LH: laparoscopic (assisted) hysterectomy. TAH: total abdominal hysterectomy. SAH: subtotal abdominal hysterectomy. VH: vaginal hysterectomy. (a positive association, i.e. the longer the gynecologist had worked in the specialty the more he/she favored SAH) and VH (negative association) (p0.0003). The number of hysterectomies carried out yearly was significantly associated with preference for all four modes of hysterectomy: positive associations with preference for TAH and VH (p and p0.0013, respectively), and negative associations with SAH and LH (p and p0.0028, respectively). The post hoc test of place of work showed significant differences regarding preference for LH between gynecologists in outpatient clinics and county and central hospitals (p B and pb0.0001, respectively). The gynecologists in outpatient clinics had in both cases a higher preference for LH than those in county and central hospital settings. There was also a difference between gynecologists in university hospitals and county and central hospitals where those in university hospitals had a higher preference for LH (p and pb0.0001, respectively). Regarding preference for VH there was a significant difference between gynecologists in university and central hospitals compared to central hospital gynecologists who had a higher preference for VH (p0.0002). The central hospital gynecologists rated even the vaginal mode higher than county hospital and outpatient clinic gynecologists (p and p B0.0001, respectively). There was a striking similarity between genders with regard to preferred method if they were divided with regard to place of work. Discussion Patients and doctors may have different attitudes to the mode of hysterectomy and many factors should be considered when choosing how the uterus should be removed in the individual case. The method of choice should idealistically comply with evidence-based recommendations. It is therefore important to investigate gynecologists attitudes toward mode of hysterectomy on benign indication. This study indicates that the choice of mode of hysterectomy for benign conditions is influenced by gender, seniority, quantity of hysterectomies carried out per year and the place of work of the gynecologist, and is not just based on evidence-based recommendations. Only a few studies dealing with gynecologists attitudes to mode of hysterectomy have been published (46). These surveys had different designs and hypotheses. The results must therefore be interpreted carefully before comparisons can be made. Only one study had similar design as the present study with the same modes of hysterectomy represented (4). The response rate in the present study was equal or slightly higher than in the previously published. Validation of a questionnaire is important. In this study the questionnaire was content and readability validated. We decided to ask about how the gynecologist would counsel in case of a close relative (male gynecologist) or for herself (female gynecologist) and further how the gynecologist would consider the distribution of the modes of hysterectomy in the general population of patients with benign disorders in order to obtain the most personal as well as general attitude to the mode of hysterectomy. No specific analysis of the non-responders was done in this study except that the gender of the responders was equal to the gender-distribution of members in the specialty according to statistics from the Swedish Medical Association (7). There is a risk that surgically interested gynecologists may have answered the survey to a larger extent than notsurgically interested. A gynecologist who does not perform hysterectomy currently might have thought that his or her opinion was not up-to-date or not of interest and thus did not answer the survey. However, no such analyses have been executed in any of

7 Gynecologists attitudes to mode of hysterectomy 273 the other above-mentioned studies (46). In two of these studies the non-responder rates were higher than in our study. Their results might therefore be more unreliable for generalization. Several factors may covariate with a gynecologist s attitude to mode of hysterectomy. It is therefore important to adjust the results for these potential confounders. Multivariate analyses were conducted in this study and adjustments for four potential confounding factors included. Similar statistical tests were used in the Danish study (4), whereas the British study used only descriptive statistics (5) and the American study used univariate statistics (6). This study showed that VH was the most preferred method of benign hysterectomy for personal choice as well as in general. By contrast, Gimbel et al. found that SAH was the preferred mode (4). The difference might be explained by the fact that the Danish survey was conducted in 2000 and ours in During that period the proportion of VH increased from 24 to 36% of hysterectomies in Sweden according to the Swedish National Register of Gynecological Surgery. In 2004, 2,053 hysterectomies on benign indications were included in the register. The distribution of the four modes of hysterectomy that year (corresponding to the time of the survey) was: TAH 37%, SAH 23%, LH 4% and VH 36% (unpublished data, personal communication from Mats Löfgren, Swedish National Register of Gynecological Surgery). This distribution differs from the results of the present survey concerning the proportion of TAH and LH. In particular, TAH was done more often in reality than what was suggested for benign hysterectomy overall in the survey, in contrast to LH that was less often used. VH is a popular mode of hysterectomy in reality and as proposed by the gynecologists in the survey. Thus, it seems that Swedish gynecologists generally comply with the recommendations from the Cochrane Database (3). The laparoscopic approach has not gained high popularity, as shown by the low frequency of LH in the Swedish National Register of Gynecological Surgery for 2004 and the low number of gynecologists conducting laparoscopic hysterectomy according to the survey. VH was the most preferred mode of benign hysterectomy among Swedish gynecologists when the uterus was of normal size, significantly more often recommended by male than female gynecologist which is in accordance with the Danish study (4). With a significantly enlarged uterus the abdominal route was the preferred method of choice. Subtotal hysterectomy was preferred especially when no cervical dysplasia had been found. Senior gynecologists seemed to prefer SAH significantly more often than the younger gynecologists in the suggested distribution of benign hysterectomy overall, but not in their personal choice, in particular if they personally had low surgical activity concerning hysterectomies. This might partly be explained by the impact of the studies by Kilkku (810) concerning SAH which gained much attention during the 1980s, i.e. during the time when the present senior gynecologists were in training. The recommendation of total hysterectomy in cases with previous cervical dysplasia was almost unanimous in the survey. This probably reflects the fact that the risk of recurrence of dysplasia and thus cervical cancer is higher in a population with previous dysplasia than in a population without previous dysplasia (1113). The preferred mode of hysterectomy varied between the types of hospital. Similar findings were reported in the Danish study, where gynecologists in the capital area recommended SAH and LH more often than the gynecologists in the province (4). The laparoscopic method is a relatively new method, introduced in 1989 by Reich et al. (14). The method has not gained widespread popularity in Sweden, as seen in the present study by the number of gynecologists performing LH. New surgical methods are by tradition usually first introduced in hospitals with academic traditions. This may partly explain why university hospitals had the highest preference for LH. VH is not a new method. It regained popularity shortly after the introduction of LH. The reasons why VH became a popular method at central and county hospitals might be lack of technical resources, education and surgical skill to perform LH, or since a substantial part of the LH was conducted through the vaginal route, the gynecologists found it unnecessary to perform the laparoscopic part. Besides, the gynecologists may have been interested in taking advantage of the reported experience of short hospital stay and sick-leave in LH. Last but not least the laborious work of a single gynecologist from a central hospital, who, through dedicated programs with live hands-on teaching and training, taught gynecologists the technique of VH (15), may have had substantial impact of the expansion of VH in Sweden. The survey also revealed that gynecologists not involved in surgery comprising hysterectomy favor minimal invasive (LH) and less radical surgery (SAH). The reason for this may either be a belief that these treatments may be less harmful to the patient or a clinical impression which these clinicians observe when they have seen patients after surgery. Interestingly, there was a striking difference concerning gender between gynecologists personal

8 274 P. Persson et al. preference of mode of hysterectomy and that for the distribution of the modes of hysterectomy in the general population. Male gynecologists were more prone to recommend their spouse or partner to have a VH than the female gynecologist would prefer for herself, whereas such difference between the genders did not exist in the suggested general distribution of benign hysterectomies overall. The reason for this discrepancy may be the construction of the survey with only three scenarios of clinical conditions for hysterectomy for a personal preference, whereas the suggested distribution of modes of hysterectomy should encompass all benign clinical conditions. However, other reasons such as external influence or limited clinical knowledge may play a role. It is obvious that settled reports from national registers such as the Swedish National Register of Gynecological Surgery may influence gynecologists attitude to mode of hysterectomy. However, randomized controlled studies are needed to determine longterm effects of the various modes in order to develop evidence-based recommendations for mode of hysterectomy for benign conditions. Acknowledgements The study was supported financially by the County of Östergötland and Linköping University. Declaration of interest: None of the authors reported any conflicts of interest. References 1. Socialstyrelsen. Epidemiologiskt Centrum-Folkhälsa i siffror [in Swedish]. Available at Statistik/statistikdatabas/ 2. Carlson KJ, Nichols DH, Schiff I. Indications for hysterectomy. N Engl J Med. 1993;328: Johnson N, Barlow D, Lethaby A, Tavender E, Curr E, Garry R Surgical approach to hysterectomy for benign gynaecological disease. Cochrane Database Syst Rev. 2006; 19:CD Update of: Cochrane Database Syst Rev. 2005; (1):CD Gimbel H, Ottesen B, Tabor A. Danish gynecologists opinion about hysterectomy on benign indication: results of a survey. Acta Obstet Gynecol Scand. 2002;/81:/ Thakar R, Manyonda I, Robinson G, Clarkson P, Stanton S. Total versus subtotal hysterectomy: a survey of current views and practice among British gynaecologists. J Obstet Gynaecol. 1998;/18:/ Zekam N, Oyelese Y, Goodwin K, Colin C, Sinai I, Queenan JT. Total versus subtotal hysterectomy: a survey of gynecologists. Obstet Gynecol. 2003;/102: / Swedish Medical Association. Physicians in Sweden Available at webb.pdf (Retrieved June 3, 2005). 8. Kilkku P, Hirvonen T, Grönroos M. Supra-vaginal uterine amputation vs. abdominal hysterectomy: the effects on urinary symptoms with special reference to pollakisuria, nocturia and dysuria. Maturitas. 1981;/3:/ Kilkku P. Supravaginal uterine amputation vs. hysterectomy. Effects on coital frequency and dyspareunia. Acta Obstet Gynecol Scand. 1983;/62:/ Kilkku P, Grönroos M, Hirvonen T, Rauramo L. Supravaginal uterine amputation vs. hysterectomy. Effects on libido and orgasm. Acta Obstet Gynecol Scand. 1983;/62:/ Pettersson F, Malker B. Invasive carcinoma of the uterine cervix following diagnosis and treatment of in situ carcinoma. Record linkage study within a National Cancer Registry. Radiother Oncol. 1989;/16:/ Viikki M, Pukkala E, Hakama M. Risk of cervical cancer subsequent to a positive screening cytology: follow-up study in Finland. Acta Obstet Gynecol Scand. 2000;/79:/ Mitchell H, Hocking J. Influences on the risk of recurrent high grade cervical abnormality. Int J Gynecol Cancer. 2002;/ 12:/ Reich H, DeCaprio J, McGlynn F. Laparoscopic hysterectomy. J Gynecol Surg. 1989;/5:/ Ottosen C. Dare to perform the surgery vaginally! Vaginal hysterectomy is to be preferred when there is no indication for the abdominal intervention [in Swedish]. Lakartidningen. 1997;/94:/21836.