Comparison of Morbidity Associated With Laparoscopic Myomectomy and Hysterectomy for the Treatment of Uterine Leiomyomas

Transcription

1 Comparison of Morbidity Associated With Myomectomy and Hysterectomy for the Treatment of Uterine Leiomyomas Madeleine Lemyre, MD, 1,2 Emmanuel Bujold, MD, MSc, 2 Ruth Lathi, MD, 1 Lisa Bhagan, MD, 1 Jian Qun Huang, MD, 1 Camran Nezhat, MD 1,3 1 Department of Obstetrics and Gynecology, Stanford University Medical Center, Palo Alto CA 2 Department of Obstetrics and Gynecology, Faculty of Medicine, Laval University, Quebec City QC 3 Department of Surgery, Stanford University Medical Center, Palo Alto CA Abstract Objective: To compare short-term morbidity and quality of life after laparoscopic hysterectomy (LH) and laparoscopic myomectomy (LM) for the treatment of symptomatic uterine leiomyomas. Method: We performed a prospective, observational study of women who were eligible for both surgical procedures. After informed consent was obtained, each participant was asked to complete the SF-12v2 Health Survey before surgery and to repeat it seven days and 28 days after surgery. Data on short-term morbidities, such as operative time, blood loss, length of hospital stay, and surgical complications, were collected by an obstetrician-gynaecologist. Women who underwent LH were compared by non-parametric statistical analyses with those who underwent LM. Results: Sixty-one women were recruited between January 1 and December 31, 2008, including 40 who underwent LM and 21 LH. Women who underwent LH were older, had higher parity, and were less likely to have infertility than those who chose LM. Median LH operative time of 223 minutes (IQR 214 to 241) was slightly longer than for LM (188 minutes, IQR 154 to 239; P = 0.02). However, we found no difference between the two groups in terms of SF-12v2 fluctuation, blood loss, hospital stay, and short-term complications. Conclusion: myomectomy is a viable alternative to laparoscopic hysterectomy for women with symptomatic leiomyomas who want conservative surgery. The procedures have similar morbidity and impact on quality of life. Key Words: Hysterectomy, myomectomy, fibroids, myoma, laparoscopy, morbidity, sf-12, minimally invasive surgery Competing Interests: None declared. Received on July 12, 2011 Accepted on September 12, 2011 Résumé Objectif : Comparer la qualité de vie et la morbidité à court terme à la suite de l hystérectomie laparoscopique (HL) et de la myomectomie laparoscopique (ML) visant la prise en charge de léiomyomes utérins symptomatiques. Méthode : Nous avons mené une étude observationnelle prospective auprès de femmes qui étaient admissibles à ces interventions chirurgicales. À la suite de l obtention du consentement éclairé des participantes, nous avons demandé à chacune d entre elles de remplir le questionnaire sur la santé SF-12v2 à trois moments distincts : avant la chirurgie, sept jours après celle-ci et vingthuit jours après la chirurgie. Les données sur les morbidités à court terme (comme la durée opératoire, la perte sanguine, la durée de l hospitalisation et les complications chirurgicales) ont été recueillies par un obstétricien-gynécologue. Les femmes ayant subi une HL ont été comparées, au moyen d analyses statistiques non paramétriques, aux femmes qui ont subi une ML. Résultats : La participation de 61 femmes (40 qui ont subi une ML et 21 qui ont subi une HL) a été sollicitée entre le 1 er janvier et le 31 décembre Les femmes qui ont subi une HL étaient plus âgées, présentaient une parité accrue et étaient moins susceptibles de connaître une infertilité que les femmes qui ont choisi de subir une ML. La durée opératoire médiane de l HL (223 minutes, ÉIQ de 214 à 241) s est avérée légèrement plus longue que celle de la ML (188 minutes, ÉIQ de 154 à 239; P = 0,02). Cependant, nous n avons constaté aucune différence entre les deux groupes en matière de fluctuation du SF-12v2, de perte sanguine, de durée de l hospitalisation et de complications à court terme. Conclusion : La myomectomie laparoscopique constitue une solution de rechange viable à l hystérectomie laparoscopique pour les femmes présentant des léiomyomes symptomatiques qui souhaitent subir une chirurgie conservatrice. Ces interventions sont similaires du point de vue de la morbidité et des effets qu elles exercent sur la qualité de vie. J Obstet Gynaecol Can 2012;34(1):57 62 JANUARY JOGC JANVIER

2 INTRODUCTION Uterine leiomyomas, also called uterine fibroids, affect about 25% of women in their reproductive years. 1 Although most fibroids are asymptomatic, treatment may be required for abnormal bleeding, pain, compressive symptoms, or fertility issues. 1 Traditionally, laparotomy is performed to remove either the fibroids selectively or the entire uterus. To date, hysterectomy is the most common surgical procedure for treating symptomatic leiomyomas, but myomectomy is becoming a popular and safe alternative. 2 5 Myomectomy is associated with longer operative time than hysterectomy, but mean blood loss and the febrile morbidity rate are potentially reduced. 3,4 However, definitive conclusions about the benefits and risks of either procedure are limited by the power of previous studies. In the last two decades, laparoscopy has widely replaced laparotomy for gynaecological surgery because of its advantages in many situations. 6 With regard to hysterectomy, the laparoscopic approach has become a well-accepted option when vaginal hysterectomy is not possible. The benefits of laparoscopic hysterectomy include decreased bleeding, shorter hospital stay, speedier return to normal activities, and lower infection risk than open procedures. However, it is also associated with longer operating time and more urinary tract injuries. 7 myomectomy is linked with less postoperative morbidity and pain than laparotomy. 8,9 No reports to date have compared LM and LH in women with symptomatic uterine leiomyomas presenting for surgery. In our centre, both approaches are commonly offered to such women. Therefore, we decided to evaluate and weigh the morbidity from both procedures. MATERIALS AND METHODS This prospective study compared the features of two laparoscopic surgery approaches, uterine myomectomy and hysterectomy. Women with symptomatic fibroids who were eligible for both procedures were included in the study. Eligibility was defined by the presence of any typical symptoms of uterine leiomyomas, such as excessive bleeding, pelvic pain, bladder or bowel compression, infertility, or ABBREVIATIONS Hb hemoglobin IQR LH LM VAS interquartile range laparoscopic hysterectomy laparoscopic myomectomy visual analogue scale recurrent miscarriages associated with uterine leiomyomas measuring 2 to 15 cm in mean diameter and confirmed by pelvic ultrasound. The exclusion criteria were age less than 18 years old, suspicion of malignancy, and inability to read the English language. All women were recruited after they had already decided on the type of surgery they would undergo. The choice between removal and preservation of the uterus was left to the patient, influenced by her desire for future pregnancy, personal convictions, and preferences. Patients were given information about the study, and participation was voluntary. If they decided to participate in this research investigation, a physician obtained their informed consent. All procedures were performed by a single surgeon with the assistance of fellows in minimally invasive surgery in two hospitals, the Stanford University Medical Center, Palo Alto, CA, and the El Camino Hospital, Mountain View, CA. Both techniques have been well-described elsewhere. 6,10 Subjects were recruited over a one-year period. The two groups of women were compared with respect to age, race, parity, BMI, past surgical history, hemoglobin concentration, uterus and fibroid sizes, and indications for surgery at study entry. Uterus and fibroid sizes were assessed on pelvic examination as being equivalent to a number of weeks of gestation and by ultrasound and/or MRI. Uterine volume was calculated, as proposed by Yaman et al., 11 from endovaginal ultrasound, pelvic ultrasound, or MRI. Surgical indications were divided into four groups according to the main complaint: abnormal uterine bleeding, fertility issues, pelvic pain, and compressive symptoms. Data related to surgery were collected, including length of operative time, use of robotics, perioperative bleeding (estimated blood loss and transfusion), and additional procedures performed during surgery (treatment of endometriosis, ovarian cystectomy, and salpingooophorectomy). The length of hospital stay and the weight of the pathology specimen (fibroid or uterus) were also recorded for study purposes. General health and wellbeing were evaluated by the standardized, validated 12-question SF-12v2 Health Survey (Quality Metric Inc., Lincoln, RI) at three specific times: patients were invited to complete the form preoperatively, then at seven and 28 days after surgery. The data were calibrated and transformed to a 0 to 100 scale, following the user s manual for norm-based scoring, in which the general United States population has a mean of 50 with a standard deviation of 10. A mental component summary and a physical component summary were obtained for each patient at each time. The differences in the scores (mental and physical) at one week and four weeks postoperatively compared with the baseline scores (preoperatively) were calculated and compared between the two study groups. 58 JANUARY JOGC JANVIER 2012

3 Comparison of Morbidity Associated With Myomectomy and Hysterectomy for the Treatment of Uterine Leiomyomas After hospital discharge, the participants completed a daily questionnaire about their pain, their return to normal activities, and any potential complications. A visual analogue scale from 0 to 10 was adopted for pain evaluation. Daily scores were added to provide a cumulative VAS score. Also, the need for non-narcotic analgesics (acetaminophen 500 mg or ibuprofen 800 mg) and narcotics (hydrocodone 5 mg, oxycodone 5 mg, hydromorphone 2 mg, morphine 7.5 mg, meperidine 50 mg, or codeine 30 mg) was recorded as medication duration, defined as the last day of narcotics use among the first 28 postoperative days. A complication was defined as any need for unplanned medical assistance, including telephone consultation, office consultation, hospitalization, or surgery in the first postoperative month. A significant portion of our patients, who were from other states, were asked to return the questionnaires by mail. They were reminded by telephone if the questionnaires were not received on time. The primary outcome was defined as the difference in the drop in the physical component summary of the SF-12v2 one week after surgery between women who underwent LM and those who had LH. According to power calculations from the user s manual, a sample size of 33 participants was required to detect a significant difference of 5 points between two non-related groups (alpha = 0.05, power = 80%, two-tailed test, intertemporal correlation = 0.70). Secondary outcomes included the drop in the physical component summary after four weeks, estimated blood loss, fall in hemoglobin concentration, need for transfusion, length of surgery, length of hospital stay, mean postoperative pain (calculated as the sum of every score on the daily VAS over one month), total narcotic and non-narcotic analgesia, the rate of patients going back to work at four weeks postoperatively, and the complication rate. Continuous variables were compared with the Mann Whitney U test, and dichotomic variables, with Pearson chi-square test. Statistical analyses were performed with SPSS 17.0 (IBM Corp., Armonk NY). Institutional review board approval was obtained at both surgery centres. A license for SF-12v2 Health Survey administration was obtained previously. RESULTS Sixty-one eligible women consented to participate during the study period (January 1 to December 31, 2008). Forty underwent myomectomy (21 with morcellator and 19 with morcellation through a mini-laparotomy 12 ), and 21 underwent hysterectomy (3 supracervical and 18 total). No conversion from one procedure to the other occurred. Women in the LH group were older and had higher parity and more complicated past surgical histories (Table 1). Otherwise, patients were comparable with respect to race, BMI, preoperative hemoglobin concentration, and uterus and fibroid size. All study patients were premenopausal, and none was on GnRH agonist therapy at the time of surgery. The mean diameter of the largest fibroid of each patient was over 6 cm. All women desiring future fertility chose myomectomy, but other indications for surgery were similar in both groups. The most common reasons for surgery were pelvic pain, abnormal bleeding, and compression symptoms. Details of the actual procedures are presented in Table 2. Hysterectomy was associated with significantly longer median operative time than myomectomy (223 vs. 188 minutes, P = 0.02). However, both surgeries resulted in comparable median operative blood loss and length of hospital stay. The da Vinci robot (Intuitive Surgical, Sunnyvale, CA) was deployed in eight myomectomies and nine hysterectomies. Myomectomy with robotic assistance had similar median operative time (244 minutes, IQR 149 to 256 vs. 185 minutes, IQR 156 to 226; P = 0.38) and lower estimated blood loss (125 ml, IQR 100 to 150 vs. 225 ml, IQR 156 to 400; P = 0.02) compared with standard laparoscopy. 13 In cases of hysterectomy, median operative time was longer when the robot was used (239 minutes, IQR 218 to 248 vs. 216 minutes, IQR 201 to 229; P = 0.02), but there was no difference in terms of estimated blood loss (300 ml, IQR 175 to 500 vs. 250 ml, IQR 100 to 550; P = 0.49). 14 Specimen weight was similar between groups and corresponded to preoperative evaluation. Primary outcome in the two groups is reported in Table 3. The rate of follow-up at one week was 54%, and several women did not complete or did not return the questionnaires at four weeks, even after the telephone call reminder. In both groups, the SF-12v2 Health Survey physical score seven days after surgery was 20 points lower than the preoperative score, which represented two standard deviations below average. Patients in both groups improved in the following three weeks, and the scores at 28 days after surgery were also similar. Mental score did not fluctuate during the recovery period. Data on pain and use of analgesia during the first 28 days after surgery are shown in Table 4. VAS score, rates of use or non-use of analgesics or narcotics, and complication rates were similar between the two groups. Seventeen women disclosed that they had requested some form of medical assistance. Among them, six in the myomectomy group and four in the hysterectomy group had had a medical complication requiring consultation with a physician. One patient, who had myomectomy with morcellation through a mini-laparotomy, was re-admitted for a large subcutaneous hematoma resulting in anemia, lipothymia, and vertigo requiring blood transfusion. In the hysterectomy group, one woman developed an abdominal abscess and was treated with intravenous antibiotics on postoperative day 13 for a total of JANUARY JOGC JANVIER

4 Table 1. Demographic and baseline characteristics of patients Myomectomy n = 40 Hysterectomy n = 21 P Age, years 41 (38 to 45) 47 (43 to 50) <0.01 Race, n (%) Black 4 (10) 1 (5) 0.53 White 27 (68) 17 (81) Other 9 (22) 3 (14) Parity 0 (0 to 1) 1 (0 to 2) <0.01 BMI, kg/m 2 23 (21 to 27) 25 (23 to 28) 0.08 Prior abdominal surgery, n (%) 16 (40) 14 (67) 0.05 Indications for surgery, n (%) Bleeding 25 (63) 16 (76) 0.03 Abdominal pain 27 (68) 17 (81) Compression symptoms 20 (50) 9 (43) Infertility 15 (38) 0 Hemoglobin, g/l 129 (120 to 134) 126 (116 to 139) 0.72 Estimated uterine size 14 (10 to 16) 14 (12 to 18)* 0.55 Largest fibroid diameter, mm 63 (47 to 80) 50 (35 to 80) 0.58 Uterine volume, ml 294 (139 to 484) 398 (136 to 677) 0.43 *n = 19 n = 27 n = 11 Table 2. Surgery characteristics Myomectomy n = 40 Hysterectomy n = 21 P Operative time, minutes 188 (154 to 238) 223 (214 to 241)* 0.02 Use of robot, n (%) 8 (20) 9 (43) 0.06 Perioperative bleeding Estimated blood loss, ml 200 (150 to 375) 300 (150 to 500) 0.36 Transfusion, n (%) 7 (18) 5 (24) 0.56 Additional procedures, n (%) Treatment of endometriosis 29 (73) 19 (90) 0.10 Ovarian cystectomy or oophorectomy 22 (55) 10 (48) 0.58 Hospital stay, hours 27 (24 to 33) 29 (27 to 51) 0.08 Specimen weight, grams 185 (36 to 390) 305 (107 to 674)* 0.11 *n = 20 n =19 n = JANUARY JOGC JANVIER 2012

5 Comparison of Morbidity Associated With Myomectomy and Hysterectomy for the Treatment of Uterine Leiomyomas Table 3. SF-12v2 Health Survey Results myomectomy n hysterectomy n P Physical component summary Preoperative 53 (50 to 57) (42 to 55) week postoperative 28 (21 to 40) (23 to 37) weeks postoperative 42 (35 to 50) (35 to 48) Difference at 1 week postoperative 20 ( 30 to 13) ( 30 to 10) Difference at 4 weeks postoperative 7 ( 11 to 1) 22 8 ( 2 to 5) Mental component summary Preoperative 46 (39 to 52) (36 to 56) week postoperative 49 (44 to 57) (38 to 55) weeks postoperative 47 (39 to 53) (44 to 58) Difference at 1 week postoperative 3 ( 5 to 13) 23 3 ( 6 to 13) Difference at 4 weeks postoperative 3 ( 7 to 6) 22 8 ( 5 to 14) Table 4. Morbidity after hospital discharge Myomectomy n = 24 Hysterectomy n = 8 P Pain evaluation and analgesia Cumulative VAS 51 (39 to 82) 46 (26 to 68) 0.42 Non-narcotic use, tablets 35 (16 to 49) 24 (9 to 56) 0.57 Narcotic use, tablets 4 (1 to 17) 3 (1 to 8) 0.62 Length of narcotic use, day 3 (2 to 8) 4 (1 to 9) 0.91 Return to activities First bowel movement, day 4 (3 to 5) 4 (3 to 4) 0.37 First day of normal diet, day 4 (3 to 6) 4 (4 to 8) 0.48 Back to work within 28 days, n (%) 13 (54) 5 (50) 0.82 Office consultation, n (%) 6 (25) 4 (40) 0.38 Hospitalization, n (%) 1 (5) 2 (20) 0.14 three days. Another patient developed fever and abdominal pain on postoperative day four (with negative results on investigation) which resolved with use of intravenous heparin and antibiotics for suspected septic pelvic thrombophlebitis. 15 DISCUSSION This observational, prospective study determined that laparoscopic myomectomy and laparoscopic hysterectomy have a similar impact on the immediate postoperative course and on women s health in the first month after surgery. While the procedural time for hysterectomy was 35 minutes longer, there was no difference in adverse outcomes, perioperative blood loss, and length of hospital stay. The results of our investigation are crucial because they give reassuring data about the two minimally invasive surgical options currently available to treat symptomatic leiomyomas. They prove that LM and LH are equally safe, and the decision to proceed with either surgery should be individualized according to the patient s desire for future fertility, as well as her personal convictions and preferences. The dramatic quality of life improvement deriving from the laparoscopic approach, as opposed to an open procedure, has been demonstrated repeatedly for both hysterectomy 7 and myomectomy. 8,9 The originality of the current work resides in the comparison of both procedures performed laparoscopically. The SF-12v2 Health Survey is an excellent JANUARY JOGC JANVIER

6 and well-validated tool to detect subtle differences and to evaluate the general impact on quality of life. 16,17 The effect of surgery on patients score was similar in both groups, and our patients were representative of the general United States population. Our study was powered to detect a clinically significant difference of 5 points in the score. The present investigation is limited by the absence of randomization and the potential biases associated with the choice of surgery. Moreover, the findings are potentially biased by the additional procedures performed with the LH or LM. Our findings are interesting but further studies are needed for definite conclusions. 18 However, in the absence of data from currently available literature, we considered a prospective, observational study to be a prudent and minimally invasive approach, and it would also allow us to respect the wishes of our patients. Moreover, the surgical procedures were performed by a single surgeon; therefore, differences or the absence of differences could not be directly related to the surgeon s experience and skill. In this case, the surgeon had extensive experience in performing both procedures. It is possible that a surgeon with different experience in either of the two techniques could obtain different results. The transfusion rate in this study (24% for LH and 18% for LM) is higher than usually reported, mainly because of cultural and personal factors in both patients and the surgical team. This cannot be transposed to the situation in Canada and is a limitation of this study. Also, the long-term surgical recurrence rate was not evaluated. Morbidity resulting from a subsequent myomectomy being required, or ultimately a hysterectomy, should be considered. Finally, our sample size and study power were limited by a relatively short time period for the recruitment of patients and by partial completeness of the follow-up data. We believe, however, that our study offers original data and reliable evidence to help counsel patients faced with the choice of these two procedures. CONCLUSION Women with symptomatic leiomyomas requiring surgical treatment can be informed that LM and LH are equivalent options in terms of short-term morbidity. Future randomized, controlled trials among women with no desire for fertility (to eliminate a selection bias) are necessary to confirm our findings. ACKNOWLEDGEMENTS The authors thank Dr Linda Nicoll, Dr Radmila Kazanegra, Dr James Watson, Dr Aroussen Laflamme, as well as Erin Amiri, Lori Arnone, Irene Patiño, Margo Perez, and Stephanie Renzi for their contributions to this study. REFERENCES 1. Buttram VC, Reiter RC. Uterine leiomyomata: etiology, symptomatology and management. Fertil Steril 1981;36: Pokras R, Hufnagel VG. Hysterectomies in the United States, Am J Public Health 1988;78: Iverson RE Jr, Chelmow D, Strohbehn K, Waldman L, Evantash EG. Relative morbidity of abdominal hysterectomy and myomectomy for management of uterine leiomyomas. Obstet Gynecol 1996;88: Sawin SW, Pilevsky ND, Berlin JA, Barnhart KT. Comparability of perioperative morbidity between abdominal myomectomy and hysterectomy for women with uterine leiomyomas. Am J Obstet Gynecol 2000;183: Rouzi AA, Al-Noury AI, Shobokshi AS, Jamal HS, Abduljabbar HS. Abdominal myomectomy versus abdominal hysterectomy for symptomatic and big uterine fibroids. Saudi Med J 2001;22: Nezhat C, Nezhat F, Nezhat CH. Nezhat s operative gynecologic laparoscopy and hysteroscopy. 3rd ed. New York: Cambridge University Press; Johnson N, Barlow D, Lethaby A, Tavender E, Curr E, Garry R. Surgical approach to hysterectomy for benign gynaecological disease. Cochrane Database Syst Rev 2005;25:CD Stringer NH, Walker JC, Meyer PM. Comparison of 49 laparoscopic myomectomies with 49 open myomectomies. J Am Assoc Gynecol Laparosc 1997;4: Holzer A, Jirecek ST, Illievich UM, Huber J, Wenzl RJ. versus open myomectomy: a double-blind study to evaluate postoperative pain. Anesth Analg 2006;102: Nezhat C, Nezhat F, Silfen SL, Schaffer N, Evans D. myomectomy. Int J Fertil 1991;36: Yaman C, Jesacher K, Pölz W. Accuracy of three-dimensional transvaginal ultrasound in uterus volume measurements: comparison with twodimensional ultrasound. Ultrasound Med Biol 2003;29: Nezhat C, Nezhat F, Bess O, Nezhat CH, Mashiach R. ally assisted myomectomy: a report of a new technique in 57 cases. Int J Fertil Menopausal Stud 1994;39: Nezhat C, Lavie O, Hsu S, Watson J, Barnett O, Lemyre M. Roboticassisted laparoscopic myomectomy compared with standard laparoscopic myomectomy a retrospective matched control study. Fertil Steril 2009;91: Nezhat C, Lavie O, Lemyre M, Gemer O, Bhagan L, Nezhat C. hysterectomy with and without a robot: Stanford experience. JSLS 2009;13: Nezhat C, Farhady P, Lemyre M. Septic pelvic thrombophlebitis following laparoscopic hysterectomy. JSLS 2009;13: Singh A, Gnanalingham K, Casey A, Crockard A. Quality of life assessment using the Short Form-12 (SF-12) questionnaire in patients with cervical spondylotic myelopathy: comparison with SF-36. Spine 2006;31: Ware J Jr, Kosinski M, Keller SD. A 12-item short-form health survey: construction of scales and preliminary tests of reliability and validity. Med Care 1996;34: Nezhat C, Lavie O, Lemyre M, Unal E, Nezhat CH, Nezhat F. Robotassisted laparoscopic surgery in gynecology: scientific dream or reality? Fertil Steril 2009;91: JANUARY JOGC JANVIER 2012