| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Division of Gynecology, Department of Obstetrics and Gynecology, University of Mississippi Medical Center, Jackson, Mississippi; Indian Health Service, Chinle, New Mexico; Division of Urogynecology, Department of Obstetrics and Gynecology, The Cleveland Clinic Hospitals, Cleveland, Ohio; and Division of Clinical and Epidemiological Research, Department of Obstetrics and Gynecology, Duke University Medical Center, Durham, North Carolina.
Address reprint requests to: Ted M. Roth, MD, Division of Gynecology, Department of Obstetrics and Gynecology, University of Mississippi Medical Center, 2500 North State Street, Jackson, MS 39211; E-mail: timbukted{at}yahoo.com.
 |
ABSTRACT |
|---|
TOP
 ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: A total of 239 abdominal myomectomies were performed at Duke University Medical Center from July 1992 through June 1998. Charts were abstracted using standardized forms. We assessed patient characteristics, surgical indications, preoperative hematocrit, and operative findings. Outcomes were defined as any complication, including transfusion.
RESULTS: The population (n = 225) was 53% black and 47% white. The mean body mass index was 26. Fourteen percent had comorbidities. Twenty percent required transfusion. Black women were found to be more likely to have uteri with more than four leiomyomata and less likely to have only one leiomyoma (P = .001). Black women were 2.48 times more likely to have a complication (P < .006). Race was no longer a significant predictor for complications (odds ratio [OR] 1.36, 95% confidence interval [CI] 0.56, 3.15) after adjustment for uterine size (OR 1.86, 95% CI 1.3, 2.67), number of leiomyomata (OR 1.83, 95% CI 1.1, 3.14), and comorbidities (OR 2.77, 95% CI 1.1, 7.69). A similar pattern was seen for blood transfusion.
CONCLUSION: Black women undergoing myomectomy are more than twice as likely to have in-hospital complication or blood transfusion than white women. This is largely attributable to differences in uterine size and leiomyoma number. Research is needed to explore why black women are more likely to have larger and more numerous leiomyomata at the time of presentation for surgery.
The incidence of uterine leiomyomata is 25–40% of women in their childbearing years.1 Black women are at particularly high risk for leiomyomata compared with women of other races. In the Nurse’s Health Study, the incidence among black women was approximately three times that among white women, with diagnosis at an earlier age.2 Black women undergoing hysterectomy for leiomyomata are younger, have more numerous leiomyomata, and larger uterine size compared with white women.3,4 Higher hysterectomy rates and higher in-hospital mortality rates have been reported for black women in comparison with white women.5
In one study, 30% of black women who underwent hysterectomy had uteri greater than 500 g versus only 15% of white women.3 The study by Hillis et al demonstrated that women undergoing hysterectomy with a uterus greater than 500 g were 1.6 times (95% confidence interval [CI] 1.0, 4.0) more likely to develop operative or postoperative complications, specifically cuff cellulitis and transfusion.4 These observations suggest that much of the excess risk of complications in black women may be attributed to increased uterine size, but we are unaware of any studies that have analyzed complication rates by both uterine size and race.
There are no studies of risk of complications for hysterectomies performed specifically for leiomyomata, and information on surgical procedures in minority women historically has been inadequate. There is even less information available on racial differences in complications of myomectomy. We examined the potential relationships between race, uterine anatomy, and complications of myomectomy by using logistic regression to determine if complication rates for black and white women differed after statistical adjustment for uterine anatomy and other comorbidities.
|
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Principal diagnoses were obtained from the admission history and preoperative diagnosis section of the operative report. The following categories were established: menstrual disorders (eg, dysmenorrhea, menorrhagia), abdominal pain, dyspareunia, infertility, and other somatic complaints including back pain and urinary symptoms. These diagnoses were not mutually exclusive.
Specific categories of comorbidities were identified based on their ability to alter the risk of morbidity or mortality. These included hypertension, coronary artery disease, diabetes mellitus, asthma, a history of pelvic inflammatory disease, endometriosis, and thyroid disease. The percentage of black and white women with comorbidities was calculated. We did not calculate a total comorbidities score or weigh more serious diagnoses because of the low morbidity in this data set.
Other categoric variables included preoperative hematocrit, a history of abdominal surgery, age, and body mass index (BMI), which was calculated by dividing the patient weight in kilograms by height in meters squared (kg/m2). Women were grouped with BMIs less than 20, 20–25, and greater than 25. Women were age grouped into those younger than 25, 25–30, 30–35, 35–40, and older than 40 years. Operative time, obtained from the anesthesia record, was defined as the time from skin incision to wound closure. Other operative factors analyzed were the presence of adhesions and the number of leiomyomata removed. Uterine size was obtained primarily from the operative notes and in part from the history and physical examination. The study by Reiter et al demonstrates an excellent correlation between subjective uterine size and uterine weight.6
Complications included any peri- or postoperative blood transfusion, estimated blood loss greater than 1000 mL, repair of a perforated viscus or major blood vessel performed either intra- or postoperatively, postoperative ileus, readmission to the hospital between the time of discharge and the follow-up visit (6 weeks), and documented postoperative infection. We specifically did not include unexplained febrile morbidity because this is a frequent occurrence after myomectomy,7 it is difficult to compare with prior studies, and the clinical significance is uncertain. Estimated blood loss was obtained from the operative notes. Transfusion was also assessed as a separate outcome. Complications were defined based on the ICD-9-CM diagnostic codes found in the discharge diagnoses. The ICD-9-CM codes were matched as closely as possible to those used in previous studies of complications in hysterectomy.8
Patient demographic and clinical factors were evaluated by univariate analysis. Variables that may have influenced the risk of complications and transfusion among black and white women were assessed with bivariate analyses using the 
2 test. Logistic regression models were developed for each variable with a P value of 0.15 or less to calculate adjusted odds ratios (OR) for the probability of having a transfusion or one or more complications for black women in comparison with white women. All analyses were performed with STATA 6.0 (Stata Corp., College Station, TX).
|
RESULTS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
presents patient charactertistics by race. Fifty-three percent of women undergoing myomectomy were black, and 47% were white. At the time of presentation for surgery, black women tended to be younger than white women (P = .021), and were more likely to be obese (BMI greater than 25, P = .001).
|
Racial differences for preoperative clinical estimation of uterine size and number of leiomyomata found at the time of surgery are also presented in Table 1
. Black women were somewhat more likely to have large uteri than white women, although this trend did not reach statistical significance (P = .12). Black women were significantly more likely to have uteri with more than four leiomyomata and less likely to have only one leiomyoma (P = .001).
There were no deaths in this series. However, 29% of the population studied had at least one complication (Table 2). When only racial groups were compared, black women were 2.48 times more likely to have any complication (95% CI 1.29, 4.75). After adjustment for uterine size, number of leiomyomata, and comorbidities, the OR for complications for black women was substantially smaller at 1.36 (95% CI 0.56, 3.15) and not significant. Uterine size (OR 1.86 for each 4-week increment in estimated size, 95% CI 1.3, 2.67), number of leiomyomata (OR 1.83, 95% CI 1.1, 3.14), and presence of comorbidities (OR 2.77, 95% CI 1.1, 7.69) were significant predictors for complications after regression analysis.
|
|
DISCUSSION |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Kjerulff et al commented that it was not clear why black women having hysterectomy in their study were at higher risk than white women for complications.5 Data from other studies suggest that part of this risk may be related to differences in uterine size at the time of surgery, data which were not available to Kjerulff et al.5 For example, in another study by Kjerullf et al, 30% of black women who underwent hysterectomy had uteri greater than 500 g versus only 15% of white women.3 Hillis et al found that women undergoing hysterectomy with a uterus greater than 500 g were 1.6 times (95% CI 1.0, 4.0) more likely to develop operative or postoperative complications.4 However, no previous study has explicitly examined whether differences in uterine size account for observed racial differences in complication rates. In the present study, black women were 2.48 times more likely to have a complication (P < .006) and 2.28 times more likely to require a transfusion (P < .04) than white women before adjustment for uterine size and number of leiomyomata. These risks decreased substantially and were no longer significant in multivariate analyses that included uterine size and number of leiomyomata; this change in the risk estimate after adjustment is evidence of confounding—the excess risk of complications seen in black women is largely explained by differences in number of leiomyomata at the time of surgery.10
We did not correlate the estimated blood loss and use of transfusion, and we did not distinguish between autologous units and donor blood. The rationale for transfusion (ie, significant anemia, hypotension, decreased urine output) was also not abstracted as part of the study because it was not always uniformly recorded in the chart. There are obviously no standard parameters by which patients are transfused, and the decision to transfuse may be influenced by other nonclinical factors. Autologous blood may also be "given back" to a patient more readily than donor blood in the absence of strong indications for transfusion. Additionally, as at most institutions, there is no uniform method of preventing blood loss at the time of myomectomy at Duke University Medical Center. Some physicians use pitressin, others use a tourniquet, whereas others use a cell saver to collect blood lost at the time of surgery. The relatively small sample size precluded analysis of the potential impact of these surgical parameters on transfusion risk.
Our results are also limited by an inability to adjust for surgical experience and skill on the part of both attending and resident physicians. Our group of primary surgeons included university-based reproductive endocrinologists and gynecologic surgeons in general obstetrics and gynecology practice. House staff participated in all surgeries. A difficulty with retrospective studies in teaching hospitals is that the potential effects of surgical experience cannot readily be determined, and it is often impossible to discern from an operative report the contribution a physician had during the surgery.
Despite these limitations, and the relatively small sample size, our study is the only one we are aware of which demonstrates racial differences in short-term adverse outcomes from abdominal myomectomy. The observed racial difference in complication rates appears to be largely attributed to differences in uterine anatomy, ie, a larger number of leiomyomata and larger overall uterine size. Further research is needed to assess the determinants of these differences—genetic, environmental, dietary, differences in access to care, differences in thresholds for seeking care, etc. They also reinforce the need to develop easily codified methods for staging or scoring the degree of difficulty of benign cases to more appropriately compare results between procedures, hospitals, and surgeons.
|
Footnotes |
|---|
doi:10.1016/S0029-7844(03)00015-2
Received June 19, 2002. Received in revised form November 4, 2002. Accepted November 27, 2002.
|
REFERENCES |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2. Marshall LM, Spiegelman D, Barbieri RL, Goldman MB, Manson JE, Colditz GA, et al. Variation in the incidence of uterine leiomyoma among premenopausal women by age and race. Obstet Gynecol 1997;90:967–73.[Abstract]
3. Kjerulff KH, Langenberg P, Seidman JD, Stolley PD, Guzinski GM. Uterine leiomyoma: Racial differences in severity, symptoms and age at diagnosis. J Reprod Med 1996;41:483–90.[Medline]
4. Hillis SD, Marchbanks PA, Peterson HB. Uterine size and risk of complications among women undergoing abdominal hysterectomy for leiomyomas. Obstet Gynecol 1996; 87:539–43.[Abstract]
5. Kjerulff KH, Guzinski GM, Langenberg P, Stolley PD, Moye NE, Kazandjian VA. Hysterectomy and race. Obstet Gynecol 1993;82:757–64.
6. Reiter RC, Wagner PL, Gambone JC. Routine hysterectomy for large asymptomatic uterine leiomyomata: A reappraisal. Obstet Gynecol 1992;79:481–4.
7. Iverson RE, Chelmow D, Strohbehn K, Waldman L, Evantash EG. Relative morbidity of abdominal hysterectomy and myomectomy for management of uterine leiomyomas. Obstet Gynecol 1996;88:415–9.[Abstract]
8. Myers ER, Steege JF. Risk adjustment for complications of hysterectomy: Limitations of routinely collected data. Am J Obstet Gynecol 1999;181:567–75.[Medline]
9. New York State Department of Health Information Systems and Health Statistics Group. Hysterectomies in NY state—A statistical profile. Albany, New York: New York State Department of Health, 1988.
10. Rothman KJ, Greenland S. Precision and validity in epidemiologic studies. In: Rothman KJ, Greenland S, eds. Modern epidemiology. 2nd ed. Philadelphia, Pennsylvania: Lippincott, 1998.
This article has been cited by other articles:
|
|
 |
|
  O. C. Smart, J. T. Hindley, L. Regan, and W. G. Gedroyc Gonadotrophin-Releasing Hormone and Magnetic-Resonance-Guided Ultrasound Surgery for Uterine Leiomyomata. Obstet. Gynecol., July 1, 2006; 108(1): 49 - 54. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. R. Myers, S. Goodwin, W. Landow, M. Mauro, E. Peterson, G. Pron, J. B. Spies, R. Worthington-Kirsch, and for the FIBROID Investigators Prospective Data Collection of a New Procedure by a Specialty Society: The FIBROID Registry Obstet. Gynecol., July 1, 2005; 106(1): 44 - 51. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Worthington-Kirsch, J. B. Spies, E. R. Myers, J. Mulgund, M. Mauro, G. Pron, E. D. Peterson, S. Goodwin, and for the FIBROID Investigators The Fibroid Registry for Outcomes Data (FIBROID) for Uterine Embolization: Short-Term Outcomes Obstet. Gynecol., July 1, 2005; 106(1): 52 - 59. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.L. Marino, B. Eskenazi, M. Warner, S. Samuels, P. Vercellini, N. Gavoni, and D. Olive Uterine leiomyoma and menstrual cycle characteristics in a population-based cohort study Hum. Reprod., October 1, 2004; 19(10): 2350 - 2355. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
