Laparoscopic Versus Abdominal Myomectomy: Practice Patterns and Health Care Use in British Columbia

Transcription

1 GYNAECOLOGY Laparoscopic Versus Abdominal Myomectomy: Practice Patterns and Health Care Use in British Columbia Innie Chen, MD, FRCSC, 1,2 Sarka Lisonkova, MD, PhD, 3 K.S. Joseph, MD, PhD, 2,3 Christina Williams, MD, FRCSC, 3 Paul Yong, MD, PhD, FRCSC, 3 Catherine Allaire, MD, FRCSC 3 1 Department of Obstetrics and Gynecology, University of Ottawa, Ottawa ON 2 School of Population and Public Health, University of British Columbia, Vancouver BC 3 Department of Obstetrics and Gynaecology, University of British Columbia, Vancouver BC Abstract Objective: To examine the relative frequency and surgical outcomes of laparoscopic myomectomy compared with abdominal myomectomy in British Columbia. Methods: A linked database containing hospital admission, operating room, and emergency room data from 2007 to 2011 from eight Vancouver Coastal Health and Providence Health Region hospitals in British Columbia was used to conduct a retrospective cohort study of women who had myomectomy for uterine fibroids. All consecutive women who had abdominal or laparoscopic myomectomy at five hospitals were included in the study. Patients who had submucosal fibroids or hysteroscopic procedures were excluded. Abdominal and laparoscopic myomectomies were contrasted in terms of patient characteristics and surgical outcomes. Statistical significance was assessed using t tests, Wilcoxon, chi-square, and Fisher exact test; a two-sided P value < 0.05 was considered significant. Results: Of eight hospitals offering gynaecologic surgery, myomectomies were performed at five hospitals located in metropolitan areas. Of 436 women undergoing myomectomy, 88 cases (20.2%) were laparoscopic, 342 (78.4%) were abdominal, and 6 (1.38%) were laparoscopic with conversion to laparotomy. Women who had laparoscopic rather than abdominal myomectomies were slightly older (mean 38.7 vs years, respectively, P < 0.05). No significant difference was observed in median operative time (106 vs. 95 min), but length of stay was decreased for laparoscopic myomectomies (median 1 vs. 2 days, P < 0.01). No significant differences were observed between laparoscopic and abdominal routes in the rates of admission to intensive care, prolonged hospitalization (> 3 days), or rehospitalization. Key Words: Myomectomy, laparoscopy, practice pattern, health care use Competing Interests: None declared. Received on January 14, 2014 Accepted on April 15, 2014 Conclusion: Myomectomies are performed in urban, metropolitan areas in British Columbia, and a significant fraction of myomectomies are performed by laparoscopy. Compared with abdominal myomectomies, laparoscopic myomectomies in pre-selected patients are associated with decreased length of stay and comparable perioperative surgical outcomes. Résumé Objectif : Examiner la fréquence relative et les issues chirurgicales de la myomectomie laparoscopique, par comparaison avec la myomectomie abdominale, en Colombie-Britannique. Méthodes : Une base de données liées contenant des données obtenues au moment de l hospitalisation, dans la salle d opération et dans la salle d urgence entre 2007 et 2011 au sein de huit hôpitaux de la Vancouver Coastal Health and Providence Health Region, en Colombie-Britannique, a été utilisée pour mener une étude de cohorte rétrospective auprès de femmes ayant subi une myomectomie en raison de la présence de fibromes utérins. Toutes les femmes consécutives qui ont subi une myomectomie abdominale ou laparoscopique au sein de cinq hôpitaux ont été admises à l étude. Les patientes qui présentaient des fibromes sous-muqueux ou qui ont subi des interventions hystéroscopiques ont été exclues. Les myomectomies abdominales et laparoscopiques ont été comparées en fonction des caractéristiques des patientes et des issues chirurgicales. La signification statistique a été évaluée au moyen des tests t, de Wilcoxon, de chi carré et exact de Fisher; une valeur P bilatérale < 0,05 a été considérée comme étant significative. Résultats : Parmi les huit hôpitaux offrant des services de chirurgie gynécologique, les cinq hôpitaux qui ont mené des myomectomies étaient situés en région métropolitaine. Chez les 436 femmes ayant subi une myomectomie, 88 interventions (20,2 %) ont été menées par laparoscopie, 342 (78,4 %) l ont été par voie abdominale et 6 (1,38 %) ont été menées par laparoscopie ensuite convertie en laparotomie. Les femmes qui ont subi une myomectomie laparoscopique, par comparaison avec celles qui ont plutôt subi une myomectomie abdominale, étaient légèrement plus âgées (moyenne : 38,7 vs 37,4 ans, SEPTEMBER JOGC SEPTEMBRE

2 Gynaecology respectivement, P < 0,05). Bien qu aucune différence significative n ait été constatée en matière de temps opératoire médian (106 vs 95 min), la durée de l hospitalisation était moindre dans le cas des myomectomies laparoscopiques (médiane : 1 vs 2 jours, P < 0,01). Aucune différence significative n a été constatée entre les voies laparoscopique et abdominale pour ce qui est des taux d admission aux soins intensifs, d hospitalisation prolongée (> 3 jours) ou de réhospitalisation. Conclusion : En Colombie-Britannique, les myomectomies sont menées en région urbaine et métropolitaine; une proportion considérable de ces myomectomies sont menées par laparoscopie. Par comparaison avec les myomectomies abdominales, les myomectomies laparoscopiques menées chez des patientes présélectionnées sont associées à une durée d hospitalisation moindre et à des issues chirurgicales périopératoires comparables. J Obstet Gynaecol Can 2014;36(9): INTRODUCTION Fibroids (leiomyomata) are benign tumours that are common in women of reproductive age, affecting nearly 80% of premenopausal women. 1 Symptoms of fibroids include heavy menstrual bleeding and increased abdominal pressure, causing significantly decreased quality of life. 2 Surgical treatments for uterine fibroids tend to be invasive and expensive, but they are associated with increased quality of life. 3 While hysterectomy remains the definitive treatment for fibroids, myomectomy is the preferred option for women who wish to retain their uterus for childbearing. Myomectomy performed by laparoscopy is associated with decreased blood loss and a faster recovery than abdominal myomectomy. 4 However, laparoscopic myomectomy is an advanced laparoscopic skill that requires the ability to suture effectively and efficiently, and recent national surveys show that the majority of gynaecologists in Canada do not offer laparoscopic myomectomy. 5,6 While several studies to date have found differences in surgical outcomes for different routes of myomectomy, 4,7 little is known about actual practice patterns, geographic distribution, and health care use associated with laparoscopic versus abdominal myomectomies. The purpose of this study was to quantify the proportion of myomectomies performed by laparoscopy, to assess the geographic practice patterns of myomectomies in British Columbia, and to compare surgical time, length of stay in hospital, emergency visits, and re-hospitalizations for laparoscopic versus abdominal myomectomies. METHODS We conducted a study of all women undergoing laparoscopic or abdominal myomectomy for a benign indication between April 1, 2007, and September 15, 2012, in any hospital within the regions of British Columbia serviced by the Vancouver Coastal Health and Providence Health Care authorities. Together, these health authorities serve more than one million people (more than one quarter of the provincial population) and cover urban and rural geographic areas that include the city of Vancouver, Vancouver s North Shore, Richmond, the Sea-to-Sky Highway, Sunshine Coast, Bella Bella, Bella Coola, the Central Coast, and surrounding areas. Patient hospitalization records from the Discharge Abstract Database were linked to the Operating Room Management Information System (ORMIS) database and the Emergency Department Surveillance System. The Discharge Abstract Database is a national hospital administrative database that captures consecutive patients using consistent data collection procedures with routine validation and quality control, 8 and the majority of the information used for this study was derived from this data source. The ORMIS is an established and validated database system dedicated to the capture of detailed clinical and resource use information pertaining to each surgical episode, 9 and this dataset was used to obtain additional information regarding the surgical cases and the operative times. The Emergency Department Surveillance System was developed by the Vancouver Coastal Health Authority to capture emergency room visits from nine acute care hospitals across the region for the purpose of population surveillance, 10 and this information was useful for determining whether an emergency room visit occurred in the observation period for each case. Together, the linkage of these three databases provided a dataset that captured hospitalization, surgical, and emergency room information. Only women who had undergone an elective abdominal or laparoscopic myomectomy and who were residents of British Columbia at the time of surgery were included. Women with submucosal fibroids or any concurrent procedure that involved the use of hysteroscopy were excluded. The Canadian Classification of Health Interventions (CCI) and ORMIS procedure codes were used to distinguish between different types of partial uterine excision and routes of myomectomy. The International Classification of Diseases, 10th Revision, diagnostic codes were used to determine indication for surgery and the type of uterine fibroid. Myomectomies were classified into abdominal (CCI code 1RM87LAGX) and laparoscopic procedures (CCI code 1RM87DAGX). Conversion from the laparoscopic to the abdominal route of myomectomy was identified by concurrent abdominal myomectomy and laparoscopy codes (CCI codes 20T70DA and 2RM70DA). The type of fibroid was identified by ICD-10CA codes: D251 for intramural, D252 for subserosal, and D259 for fibroids of unspecified type. 818 SEPTEMBER JOGC SEPTEMBRE 2014

3 Laparoscopic Versus Abdominal Myomectomy: Practice Patterns and Health Care Use in British Columbia To characterize our study population and to evaluate potential confounders, patient and hospital characteristics potentially affecting the myomectomy route were also considered. Patient characteristics included age, urban or rural residence, socioeconomic status, and indication for surgery. Rural residences were defined by residential postal codes corresponding to areas with < inhabitants. 11 In studies of health outcomes in Canada, neighbourhood income level has been considered an adequate approximation of household income. 12 Therefore, socioeconomic status was inferred from residential postal codes using neighbourhood income quintile (from first being the lowest to fifth being the highest) relative to the income distribution in British Columbia in Each hospital s geographic location (urban or rural) was also examined. Information related to health care use, such as operating time, length of hospitalization, need for intensive care, return to emergency room, and readmission to hospital was also analyzed. The proportions of abdominal, laparoscopic, and laparoscopic with conversion to abdominal myomectomies were calculated. Laparoscopic with conversion to abdominal myomectomy was combined with laparoscopic myomectomy to form a planned laparoscopic group, which was then contrasted with the abdominal myomectomy group in terms of patient characteristics and health care use. Continuous variables were compared using a t test or Wilcoxon test to assess the statistical significance of differences, and categorical variables were compared using a chi-square or Fisher exact test. All analyses were carried out using SAS version 9.3. (SAS Institute Inc., Cary, NC) and a two-sided P value < 0.05 was considered significant. Ethics approval for this study was provided by the University of British Columbia Research Ethics Board. RESULTS Of 436 women undergoing myomectomy, 88 cases (20.2%) were laparoscopic, 342 (78.4%) were abdominal, and 6 (1.38%) were laparoscopic with conversion to laparotomy. Within the eight hospitals in the Vancouver Coastal Health and Providence Health Care regions that offered gynaecologic surgery, myomectomies were performed at five hospitals, all of which were located in the metropolitan area of Vancouver. Among the five hospitals, the proportion of myomectomies performed by laparoscopy ranged from 0% to 34.3%, and most laparoscopic myomectomies (93.2 %) were performed at only two of these hospitals. Women who had laparoscopic myomectomy were older than women who had abdominal myomectomy (mean age 38.7 vs years, respectively, P = 0.04) (Table 1). Laparoscopic myomectomy was also associated with a higher proportion of subserosal fibroids, while women who had abdominal myomectomy had a higher proportion of fibroids of unspecified type. The proportion of women with intramural fibroids and both intramural and subserosal fibroids was similar between the two groups, as was the proportion of women with infertility and menorrhagia. No differences were seen in the proportion of women living in urban or rural areas or in their socioeconomic status. Median operating time was similar for both laparoscopic and abdominal myomectomy (106 vs. 95 minutes, P = 0.13), but median length of stay in hospital was less for laparoscopic than abdominal myomectomy (1 vs. 2 days, P < 0.001) (Table 2). Prolonged hospitalization (> 3 days) was also more common after abdominal myomectomy (10.3% vs. 2.1%, P = 0.01). No significant differences were observed in the rates of admission to intensive care and rehospitalization. The proportion of women who returned to the emergency room was 8.5% in the abdominal myomectomy group and 5.3% in the laparoscopic group; this difference was not statistically significant (P = 0.39). DISCUSSION Our study showed that between 2007 and 2012, approximately 20% of myomectomies for benign fibroids in the Vancouver Coastal Health and Providence Health Care regions of British Columbia were performed by laparoscopy. Most myomectomies occurred in hospitals within urban and metropolitan areas in British Columbia; such centralization of surgery for myomectomy may be appropriate, because the morbidity associated with myomectomy, compared with other gynaecologic procedures, increases the need for blood transfusion. 13 In addition, we observed that myomectomy by the laparoscopic approach was performed predominantly at two of the five metropolitan area hospitals, suggesting a clustering of surgical skill at a few select hospitals. Currently, only approximately one quarter of Canadian gynaecologists perform laparoscopic myomectomy, and only 12.5% report that more than one half of the myomectomies they perform are done laparoscopically. 6 Major reported barriers to offering laparoscopic myomectomy are lack of training and operating time. 5,6 Given these constraints, the centralization of surgical expertise in this area may be appropriate, as it would allow adequate surgical volume for maintenance of competence in this procedure. Despite the centralization of myomectomy services, we did not find living in a rural area or having low socioeconomic status to be barriers for patients to undergo laparoscopic myomectomy. The finding that route of myomectomy was dictated by clinical (type of fibroid, Table 1) rather than social and geographical characteristics (socioeconomic status and SEPTEMBER JOGC SEPTEMBRE

4 Gynaecology Table 1. Patient and hospital characteristics for elective myomectomy, British Columbia, 2007 to 2012 Patient characteristics Abdominal* n = 342 Myomectomy Laparoscopic* n = 94 Age, years, mean (SD) 37.4 (5.2) 38.7 (5.9) 0.04 Rural residence 8 (2.3) 2 (2.1) > 0.99 Socioeconomic quintile (lowest) 143 (41.8) 40 (42.6) 2 29 (8.5) 7 (7.5) 3 (median) 50 (14.6) 16 (17.0) 4 53 (15.5) 14 (14.9) 5 (highest) 67 (19.6) 17 (18.1) Type of fibroid Both intramural and subserosal 43 (12.6) 14 (14.9) 0.56 Intramural 46 (13.5) 16 (17.0) 0.38 Subserosal 41 (12.0) 24 (25.5) Not specified 212 (62.0) 40 (42.6) Secondary diagnoses Bleeding 24 (0.1) 4 (0.04) 0.33 Infertility 31 (0.1) 12 (0.1) 0.29 Data shown as n (%) unless otherwise stated. *Including six conversions from laparoscopy to laparotomy. Derived from median neighbourhood income in British Columbia in P Table 2. Surgery time, hospital stay, emergency visits and re-hospitalization by surgical approach for elective myomectomy, British Columbia, 2007 to 2012 Outcome Abdominal* n = 342 Myomectomy Laparoscopic* n = 94 Median surgery time in minutes (quartile range) 95 (48) 106 (62) 0.13 Median length of stay in days (quartile range) 2 (1) 1 (0) < Length of stay > 3 days, n (%) 35 (10.3) 2 (2.1) 0.01 Emergency visits, n (%) 29 (8.5) 5 (5.3) 0.39 Re-hospitalization, n (%) 3 (0.9) 1 (1.1) > 0.99 *Including six conversions from laparoscopy to laparotomy. Including interhospital transfers. P rural residence, Table 1) is reassuring, and may suggest a willingness of gynaecologists to refer patients to surgeons who are able to offer a laparoscopic approach when appropriate. Although operating time has been considered by some to be a barrier to performing laparoscopic myomectomy, 5,6 we did not find significant differences in operating time in this study. This is in contrast to the findings of Jin et al. in a systematic review of randomized controlled trials, which showed a mean increase in operating time of 13 minutes for laparoscopic myomectomy compared with abdominal myomectomy. 4 In addition, we found no significant differences in other inpatient measures of hospital use except for length of hospital stay, which has also been shown in other studies to be significantly reduced for laparoscopic versus open approaches. 14,15 This study involved a linkage of inpatient records with outpatient emergency records and readmission records, and showed that the proportion of patients with a postoperative return to the emergency room or hospital was significant. As hospitals currently represent the largest component of health care spending, 16 such linkage of data enables better assessment of the duration of hospital stay, readmissions, and intensity of health care use. In a climate of rising health care expenditures and limited resources, results from studies using population-based linked data may better guide health care decision makers in allocating resources. 820 SEPTEMBER JOGC SEPTEMBRE 2014

5 Laparoscopic Versus Abdominal Myomectomy: Practice Patterns and Health Care Use in British Columbia In our analysis, we found a trend towards higher rates of return to the emergency department for the abdominal myomectomy group than for the laparoscopic group (8.5% vs. 5.3%, P = 0.39) (Table 2). The proportion of women with a length of hospital stay of more than three days was significantly higher in the abdominal myomectomy group (10.0% vs. 2.1%, P = 0.01) (Table 2). It is possible that women pre-selected for abdominal myomectomy had more complicated uterine pathology or increased comorbidity than women who were offered a laparoscopic approach and were therefore more prone to postoperative complications. However, it is also possible that laparoscopic myomectomy is associated with fewer postoperative complications than abdominal myomectomy because of the smaller incisions and faster recovery. It is likely that in pre-selected patients laparoscopic myomectomy and the associated earlier discharge from hospital for these patients do not lead to further increases in return to hospital. The major strength of this study is the use of a populationbased linked dataset that included multiple hospitals in urban and rural locations, as well as emergency and rehospitalization data. This allowed analysis of sociodemographic and hospital factors associated with the approach used for myomectomy, as well as quantification of postoperative health care use. The use of standardized databases also ensures that consecutive patients are captured using consistent data collection procedures. The major limitation of this study is the lack of detailed clinical information, such as size, location, and number of fibroids, which may affect decisions regarding the approach used for myomectomy. In addition, bleeding during surgery and the need for blood transfusion could not be assessed using our dataset. Other limitations include the potential underreporting and coding errors inherent in the use of hospital administrative data. CONCLUSION Myomectomies are mostly performed in urban metropolitan hospitals in British Columbia. Approximately 20% of myomectomies are performed by laparoscopy, and these procedures are performed at a few centres of surgical expertise. Compared with abdominal myomectomy, laparoscopic myomectomy in pre-selected patients is not associated with increases in need for operating time or for emergency or hospital resources. Instead, laparoscopic myomectomy is associated with a decreased length of hospital stay. ACKNOWLEDGEMENTS We are grateful to the Vancouver Coastal Health Authority Decision Support and Providence Health Care Decision Support for providing the data used in this study. Innie Chen is supported by a Frederick Banting and Charles Best Canada Graduate Scholarship Award from the Canadian Institutes of Health Research, and K.S. Joseph is supported by a Chair in maternal, fetal, and infant health services research from the Canadian Institutes of Health Research. REFERENCES 1. Day Baird D, Dunson DB, Hill MC, Cousins D, Schectman JM. High cumulative incidence of uterine leiomyoma in black and white women: ultrasound evidence. Am J Obstet Gynecol 2003;188(1): Lumsden MA, Wallace EM. Clinical presentation of uterine fibroids. Baillieres Clin Obstet Gynaecol 1998;12(2): You JH, Sahota DS, Yuen PM. Uterine artery embolization, hysterectomy, or myomectomy for symptomatic uterine fibroids: a cost-utility analysis. Fertil Steril 2009; 91(2): Jin C, Hu Y, Chen XC, Zheng FY, Lin F, Zhou K, et al. Laparoscopic versus open myomectomy a meta-analysis of randomized controlled trials. Eur J Obstet Gynecol Reprod Biol 2009;145(1): Chen I, Bajzak KI, Guo Y, Singh SS. A national survey of endoscopic practice among gynaecologists in Canada. J Obstet Gynaecol Can 2012;34(3): Liu G, Zolis L, Kung R, Melchior M, Singh S, Cook EF. The laparoscopic myomectomy: a survey of Canadian gynaecologists. J Obstet Gynaecol Can 2010;32(2): Pundir J, Pundir V, Walavalkar R, Omanwa K, Lancaster G, Kayani S. Robotic-assisted laparoscopic vs abdominal and laparoscopic myomectomy: systematic review and meta-analysis. J Minim Invasive Gynecol 2013;20(3): Canadian Institute for Health Information. Data quality documentation, Discharge Abstract Database multi-year information. Ottawa: CIHI; Martin JB, Cantrell ME, Fichman RG, Hunsberger PK, Radoyevich M. An integrated surgical suite management information system. J Med Syst. 1984;8(4): Vancouver Coastal Health, Public Health Surveillance Unit: an investment in regional public health, highlights from the first five years, Statistics Canada. Postal code conversion file (PDDF), reference guide Catalogue no. 92F0153GIE. Available at: Collection/Statcan/92F0153GIE/92F0153GIE pdf. Accessed April 6, Mustard CA, Derksen S, Berthelot JM, Wolfson M. Assessing ecologic proxies for household income: a comparison of household and neighbourhood level income measures in the study of population health status. Health Place 1999;5: Pundir J, Krishnan N, Siozos A, Uwins C, Kopeika J, Khalaf Y, et al. Perioperative morbidity associated with abdominal myomectomy for very large fibroid uteri. Eur J Obstet Gynecol Reprod Biol Apr;167(2): Palomba S, Zupi E, Falbo A, Russo T, Marconi D, Tolino A, et al. A multicenter randomized, controlled study comparing laparoscopic versus minilaparotomic myomectomy: short-term outcomes. Fertil Steril 2007;88(4): Malzoni M, Tinelli R, Cosentino F, Iuzzolino D, Surico D, Reich H. Laparoscopy versus minilaparotomy in women with symptomatic uterine myomas: short-term and fertility results. Fertil Steril 2010; 93(7): Canadian Institute for Health Information (CIHI). National Health Expenditure Trends, 1975 to Ottawa: CIHI; Available at: Accessed October 6, SEPTEMBER JOGC SEPTEMBRE