Hysterectomy is the most common major gynecologic surgery performed in women in the United States, with approximately 600,000 women undergoing this procedure each year.1 Since the introduction of the laparoscopic hysterectomy in 1989,2 this procedure has been associated with shorter hospitalizations, faster recovery, and fewer postoperative infections compared with abdominal hysterectomy.3,4 However, laparoscopic hysterectomy requires specialized training and potentially longer operating time.3
Previous studies have reported different trends in the rates of laparoscopic hysterectomy. National data from the Healthcare Cost and Utilization Project reported an increase in the percentage of laparoscopic hysterectomy from 0.3% in 1990 to 9.9% in 1997.5 In contrast, data from Kaiser Permanente Northern California showed a decrease in this percentage from 13% in 1995 to 3.9% in 2003.6 Although laparoscopic hysterectomy rates are declining in the Northern California Kaiser population, this trend may not be representative of the United States.
Our primary objective was to estimate national hysterectomy rates by type of hysterectomy. Our secondary aims were to compare age, length of stay, and regional variation in type of hysterectomy performed for benign indications.
MATERIALS AND METHODS
We conducted a cross-sectional study using data from the 2003 Nationwide Inpatient Sample, which contains a 20% stratified, random sample of discharge data from approximately 1,000 community hospitals in the United States. Community hospitals are defined as all nonfederal, general, and specialty short-term hospitals, which include both public and academic medical centers. The 2003 Nationwide Inpatient Sample contains all-payer data on hospital inpatient stays from the 37 states that participated in the Healthcare Cost and Utilization Project, which is maintained by the Agency for Healthcare Research and Quality. This study was exempt from institutional review board review because we used publicly available data.
Each discharge contains up to 15 procedure and 15 diagnosis codes. Procedures and diagnoses were based on International Classification of Diseases, 9th Revision, Clinical Modification codes, as well as the Agency for Healthcare Research and Quality Clinical Classification Software.7 The Clinical Classification Software is a tool that clusters either diagnosis or procedure codes into clinically meaningful categories. We included any discharge with a procedure code for hysterectomy, which was defined by the Clinical Classification Software procedure code 124. The type of hysterectomy was defined by International Classification of Diseases, 9th Revision, Clinical Modification procedure codes 68.4 for total abdominal hysterectomy, 68.3 and 68.39 for subtotal abdominal hysterectomy, 68.5 and 68.59 for vaginal hysterectomy, 68.31 and 68.51 for laparoscopic hysterectomy, 68.6 and 68.7 for radical hysterectomy, and 68.9 for other unspecified hysterectomy.
We abstracted data regarding age, length of stay, and hospital characteristics. Subjects aged younger than 16 years were excluded, and hysterectomies related to disorders of pregnancy were also excluded. Approximately 30% of the observations had missing data on racial characteristics because many states do not collect this information. Given the amount of missing data, we performed a subset analysis for observations with data on race. For diagnoses, we used the following Clinical Classification Software diagnosis codes: benign neoplasm of the uterus (46), inflammatory disease of female pelvic organs (168), endometriosis (169), prolapse of female genital organs (170), menstrual disorders (171), and other, which included ovarian cyst (172), menopausal disorders (173), and other female genital disorders (175).7 For hysterectomies performed for benign conditions, we excluded any discharge associated with a diagnosis of any gynecologic, gastrointestinal, or urologic malignancy.
Given the sampling design of the Nationwide Inpatient Sample, we used Stata 9.1 (StataCorp, College Station, TX) to account for the sampling weights, strata, and clusters. In this database, there were 60 strata and 986 clusters, which were based on hospital identification number. The sampling weights were provided in the data set. To calculate the total hysterectomy rate, we estimated the total number of hysterectomies performed for both benign and malignant conditions in 2003. The rate was estimated by dividing the total number of hysterectomies by the number of women aged 16 years or older, which was based on the 2000 National Census.8 In a similar manner, we estimated the benign hysterectomy rate and then calculated individual rates for abdominal, vaginal, and laparoscopic hysterectomies for benign disease. We assessed the mean age and mean length of stay for each type of hysterectomy, as well as differences by race and region of the United States.
Using weighted data analysis and cluster sampling, we used analysis of variance to compare age and length of stay among abdominal, vaginal, and laparoscopic hysterectomies. For pairwise comparisons, we used t tests that were calculated based on the sampling design.
In the United States, 602,457 hysterectomies were performed in 2003. There were 112,014,898 women aged 16 years or older, which corresponded to an overall hysterectomy rate of 5.38 per 1,000 women-years (Table 1). Of these, 538,722 hysterectomies were performed for benign disease, for a rate of 4.81 per 1,000 women-years. Abdominal hysterectomy (66.1%) was the most common route, followed by vaginal (21.8%) and then laparoscopic (11.8%) hysterectomy.
For benign disease, we focused on the three primary hysterectomy types: abdominal, which included total and subtotal, vaginal, and laparoscopic. There was a statistically significant difference in the mean age of patients undergoing these procedures (abdominal 44.5±0.1 years, vaginal 48.2±0.2 years, and laparoscopic 43.6±0.3 years, P<.001) (Table 2). All pairwise comparisons were also statistically significant (P<.001).
For the subset of subjects with data on race, the population was 48.1% white, 11.3% African American, 7.3% Hispanic, and 3.6% Asian or other. Among white women, the route of hysterectomy was 62% abdominal, 24% vaginal, and 14% laparoscopic. For African-American women, the distribution was 83% abdominal, 10% vaginal, and 6% laparoscopic, and for Hispanic women, the distribution was 69% abdominal, 22% vaginal, and 8% laparoscopic.
By region of the country (Northeast, Midwest, South, and West), the South had the highest hysterectomy rate and the Northeast had the lowest (Table 2). The percentage of laparoscopic procedures was similar across all regions (P=.33). Regarding length of stay, hospitalization was shortest for laparoscopic hysterectomy (1.7±0.03 days), compared with vaginal (2.0±0.03 days) and abdominal hysterectomies (3.0±0.03 days) (P<.001). All pairwise comparisons were also statistically significant (P<.001) (Table 2).
For all benign hysterectomies, the most common indications were leiomyomata (37%), menstrual disorders (19%), prolapse (13%), and endometriosis (12%) (Fig. 1). The most common diagnosis for each type of hysterectomy varied. Fibroid uterus (46%) was the most common diagnosis for abdominal hysterectomy, and prolapse (44%) was the most common diagnosis for vaginal hysterectomy. For laparoscopic procedures, the distribution of diagnoses was more evenly divided among fibroid uterus (27%) and the other diagnoses. Diagnoses also varied by race. Of hysterectomies performed in white women, the diagnoses were fibroid uterus (33%), menstrual disorders (21%), prolapse (16%), and endometriosis (14%). The indications for hysterectomies in African-American women were fibroid uterus (70%), menstrual disorders (12%), endometriosis (6%), and prolapse (4%), and among Hispanic women, the diagnoses were fibroids (46%), prolapse (17%), menstrual disorders (14%), and endometriosis (11%).
In 2003, the overall hysterectomy rate, 5.38 per 1,000 women-years, was similar to rates from the previous decade, which were 5.1–5.8 per 1,000 women-years.1,5 Although the total hysterectomy rate has remained fairly stable, the distribution of the route of surgery has changed. The proportion of laparoscopic hysterectomy has increased from 0.3% in 19905 to 11.8% in 2003. This increase in laparoscopic hysterectomy has been reflected in a decline in both abdominal and vaginal hysterectomies, which accounted for 73.6% and 24.4%, respectively, of hysterectomies in 1990. Despite this increase, the rates of laparoscopic (11.8%) and vaginal hysterectomy (21.8%) remain significantly lower than that of abdominal hysterectomy (66.1%), even though these laparoscopic and vaginal procedures were associated with shorter lengths of stay. However, these changes may not be reflected uniformly across all races because the proportion of hysterectomies performed laparoscopically was higher in white women than in African-American or Hispanic women. Abdominal hysterectomy was more common in African-American women; this may be secondary to the higher percentage of fibroid diagnoses in this population compared with white and Hispanic women.
Our results are in contrast to those of Jacobson et al,6 who reported a decrease in laparoscopic hysterectomy from 13.0% in 1995 to 3.9% in 2003. That study was based on the Northern California Kaiser Permanente population; thus, the results may not represent national trends. The discrepancy in laparoscopic hysterectomy rates may be due to differences in patient populations or physician practice patterns. Because our results were generated from the Nationwide Inpatient Sample, our findings may better reflect national practices. Furthermore, although the total hysterectomy rate was highest in the South and lowest in the Northeast, the percentages of laparoscopic hysterectomy were similar across all regions of the United States, even in the West, which includes the Northern California Kaiser Permanente population.
It is unclear what proportion of hysterectomies should be performed by each route. The route of surgery must be individualized to each patient and physician. Numerous factors must be considered in this decision-making process, including the indication for surgery, patient characteristics (eg, body mass index and prior surgeries), concomitant procedures, risk for complications, length of stay, recovery, and cost-effectiveness. Overall, a total vaginal hysterectomy is considered the surgery of choice if it is feasible because the vaginal route is associated with less morbidity than any other method.3,4,9,10 When vaginal and laparoscopic routes are compared, studies have shown that laparoscopy does not provide any benefit but is associated with higher costs.9,11,12
When a vaginal hysterectomy is not possible, a decision must be made between an abdominal or laparoscopic approach. One advantage of laparoscopic over abdominal hysterectomy is shorter hospitalization, as evidenced by a mean length of stay of 1.7 days, compared with 3.0 days in this study. In addition, laparoscopy is associated with lower intraoperative blood loss,13 less postoperative pain,13,14 fewer infections,3 and faster recovery.14 These advantages must be weighed against the disadvantages of longer operative time3,14 and higher complication rates, especially urinary tract injuries.3,4,11 These disadvantages may explain why the rate of laparoscopic hysterectomy remains relatively low.
Another possible explanation for the low laparoscopic hysterectomy rate is that this procedure requires specialized training. Many obstetrics and gynecology residency programs have implemented formal laparoscopy training sessions,15 but a survey of senior obstetrics and gynecology residents in the United State reported that 67% of respondents thought that the emphasis on laparoscopy training should be increased.16 With adequate training, one study found that the laparoscopic hysterectomy can be successfully integrated into a residency training program or health maintenance organization.17
Cost and cost-effectiveness are additional factors to consider. One study found that the cost of laparoscopically assisted vaginal hysterectomy was higher than that of vaginal or abdominal hysterectomy, secondary to disposable supplies and longer operating room times.18 In a comprehensive cost-effectiveness analysis, with effectiveness measured in quality-adjusted life-years, vaginal hysterectomy was cost-effective compared with laparoscopic hysterectomy.11,12 Between abdominal and laparoscopic hysterectomy, cost-effectiveness was dependent on the cost of each procedure because there was little difference in mean quality-adjusted life-years. Cost varied based on whether reusable or disposable equipment was used. This cost-effectiveness analysis incorporated data from the United Kingdom. Thus, a cost-effectiveness analysis in the United States might yield different results, given the differences in our health care systems.
Because our study was based on the Nationwide Inpatient Sample, there were several potential sources of bias. Limitations include inaccurate coding of procedures and diagnoses, as well as bias from nonresponse or missing data. One example is the missing data on racial characteristics. We were also unable to evaluate the effect of patient characteristics, such as body mass index or parity, on route of hysterectomy because these variables were not included in the Nationwide Inpatient Sample. Another limitation was that we calculated hysterectomy rates using a denominator that was based on all women in the United States who were aged 16 years or older. This denominator included women who had previously had a hysterectomy, which suggests that we underestimated the true hysterectomy rate. However, this issue applies to previous studies on hysterectomy rates, and thus our results remain comparable with existing data.
Regarding length of stay, we reported on mean hospitalization associated with each type of hysterectomy; it is possible that concomitant procedures influenced length of stay. With this national administrative discharge database, we cannot ascertain the effect of each individual procedure on length of stay. However, we believe that physicians who perform a hysterectomy will chose to perform additional surgeries from a similar route. Therefore, additional procedures may not significantly affect length of stay.
We reported that the overall total hysterectomy rate has remained relatively unchanged but that the proportion of laparoscopic hysterectomy has increased to 11.8%. This is an important topic to study because hysterectomy is the most common gynecologic procedure performed in women in the United States. In addition, the route of hysterectomy can affect morbidity, hospital stay, and recovery time. A decision regarding route of hysterectomy involves numerous factors, and the advantages and disadvantages of each type of hysterectomy must be considered. With advances in technology and more emphasis on laparoscopy training in residency programs, it will be interesting to assess whether the rate of laparoscopic hysterectomy will continue to increase in the future.
1. Keshavarz H, Hillis SD, Kieke BA, Marchbanks PA. Hysterectomy surveillance–United States, 1994–1999. MMWR CDC Surveill Summ 2002;51(SS-5):1–8.
2. Reich H, Decaprio J, McGlynn F. Laparoscopic hysterectomy. J Gynecol Surg 1989;5:213–7.
3. Johnson N, Barlow D, Lethaby A, Tavender E, Curr E, Garry R. Surgical approach to hysterectomy for benign gynaecological disease (Cochrane Review). In: The Cochrane Library, Issue 2, 2006. Oxford: Update Software.
4. Meikle SF, Nugent EW, Orleans M. Complications and recovery from laparoscopy-assisted vaginal hysterectomy compared with abdominal and vaginal hysterectomy. Obstet Gynecol 1997;89:304–11.
5. Farquhar CM, Steiner CA. Hysterectomy rates in the United States 1990–1997. Obstet Gynecol 2002;99:229–34.
6. Jacobson GF, Shaber RE, Armstrong MA, Hung YY. Hysterectomy rates for benign indications. Obstet Gynecol 2006;107:1278–83.
7. Agency for Healthcare Research and Quality. Healthcare Cost and Utilization Project. Clinical classifications software (CCS) for ICD-9-CM. Available at: http://www.hcup-us.ahrq.gov/toolssoftware/ccs/ccs.jsp
. Retrieved August 14, 2007.
8. U.S. Census Bureau. United States Census 2000. Available at: http://www.census.gov/main/www/cen2000.html
. Retrieved August 14, 2007.
9. Summitt RL Jr, Stovall TG, Lipscomb GH, Ling FW. Randomized comparison of laparoscopy-assisted vaginal hysterectomy with standard vaginal hysterectomy in an outpatient setting. Obstet Gynecol 1992;80:895–901.
10. Appropriate use of laparoscopically assisted vaginal hysterectomy. ACOG Committee Opinion No. 311. American College of Obstetricians and Gynecologists. Obstet Gynecol 2005;105:929–30.
11. Garry R, Fountain J, Brown J, Manca A, Mason S, Sculpher M, et al. EVALUATE hysterectomy trial: a multicentre randomised trial comparing abdominal, vaginal and laparoscopic methods of hysterectomy. Health Technol Assess 2004;8:1–154.
12. Sculpher M, Manca A, Abbott J, Fountain J, Mason S, Garry R. Cost effectiveness analysis of laparoscopic hysterectomy compared with standard hysterectomy: results from a randomised trial. BMJ 2004;328:134.
13. Marana R, Busacca M, Zupi E, Garcea N, Paparella P, Catalano GF. Laparoscopically assisted vaginal hysterectomy versus total abdominal hysterectomy: a prospective, randomized, multicenter study. Am J Obstet Gynecol 1999;180:270–5.
14. Falcone T, Paraiso MF, Mascha E. Prospective randomized clinical trial of laparoscopically assisted vaginal hysterectomy versus total abdominal hysterectomy. Am J Obstet Gynecol 1999;180:955–62.
15. Stovall DW, Fernandez AS, Cohen SA. Laparoscopy training in United States obstetric and gynecology residency programs. JSLS 2006;10:11–5.
16. Einarsson JI, Young A, Tsien L, Sangi-Haghpeykar H. Perceived proficiency in endoscopic techniques among senior obstetrics and gynecology residents. J Am Assoc Gynecol Laparosc 2002;9:158–64.
17. Hoffman CP, Kennedy J, Borschel L, Burchette R, Kidd A. Laparoscopic hysterectomy: the Kaiser Permanente San Diego experience. J Minim Invasive Gynecol 2005;12:16–24.
18. Dorsey JH, Holtz PM, Griffiths RI, McGrath MM, Steinberg EP. Costs and charges associated with three alternative techniques of hysterectomy. N Engl J Med 1996;335:476–82.