Materials and Methods
A retrospective review of all patients treated for invasive carcinoma of the cervix between 1986 and 1998, following approval from the Human Subjects Committee, was carried out at Washington University School of Medicine, Barnes/Jewish Hospital. Patients with stage I–IIa disease who were explored with the intent for radical hysterectomy and pelvic (with or without paraaortic) lymphadenectomy were eligible for review. Patient records were abstracted, and data regarding patient characteristics (including height, weight, age, and comorbidity) were collected. Patients with a body mass index (BMI) over 30 kg/m2 and an absolute weight over 85 kg (190 lbs) comprised the study group. All patients underwent surgery by an attending gynecologic oncologist. The decision to explore a patient for radical hysterectomy and pelvic lymph node dissection rather than treat with primary irradiation was at the discretion of her attending gynecologic oncologist and the patient's preference. No absolute weight limit or tumor size was used. Other factors considered in assigning a primary treatment modality were the patient's desire to preserve sexual function, the potential duration of continued ovarian hormonal function, and medical comorbidities that would influence a patient's candidacy for surgery.
Medical comorbidity was classified as mild, moderate, or severe.8 Anesthesia records were reviewed, and the American Society of Anesthesiologists physical status classification, estimated blood loss, and operating room time were recorded. Operative reports were abstracted, and data on operative findings and complications were also recorded. All patients underwent a class III radical hysterectomy and pelvic lymphadenectomy.9 Para-aortic lymphadenectomy was performed at the discretion of the attending gynecologic oncologist. Enlarged lymph nodes were generally evaluated by frozen section, and the decision to complete the radical hysterectomy and pelvic lymph node dissection was at the discretion of the attending gynecologic oncologist. During the study interval, radical hysterectomy was performed in a similar fashion by all gynecologic oncologists. During the later years of the study, closed-suction drains were not routinely used, and were placed at the discretion of the attending gynecologic oncologist. Pathology reports were reviewed, and the number and status of the pelvic and para-aortic lymph nodes were noted. Patient length of stay, time to recovery of bowel function, and perioperative complications were recorded from hospital records, as well as patient follow-up including any postoperative complications and disease status at the date of last contact. Recurrence was recorded as local or distant. Overall and disease-free survival curves were constructed using Kaplan-Meier life-table estimates. During the same study interval, records of all patients with stage Ib cervical cancer treated with primary irradiation were reviewed. A consecutive series of patients with a BMI over 30 kg/m2 and an absolute weight over 85 kg were identified. The Student t test was used to compare the mean weight and BMI of this group with that of patients undergoing radical hysterectomy and pelvic lymphadenectomy.
Forty-eight women with a BMI greater than 30 kg/m2 (all with absolute weight over 85 kg) were identified who were explored for the possibility of radical hysterectomy and pelvic lymphadenectomy during the study interval. The median BMI of this study population was 36 kg/m2 (range 31–46 kg/m2), the median weight was 95 kg (range 86–121 kg), and the median age was 40 years (range 24–66 years). Sixty percent of the patients were white, 35% black, and 5% Hispanic. Eight women had stage Ia lesions, 35 had stage Ib1 lesions, and five had stage Ib2 disease. Preoperatively, cervical tumor diameter was determined to be microscopic in 34 (71%) patients, less than 4 cm in nine (19%), and over 4 cm in five (10%). Tumor histology was squamous in 37 (77%) patients, adenocarcinoma in nine (19%), and adenosquamous in two (4%). Thirteen of the 48 patients (28%) had a medical comorbidity. Ten (22%) demonstrated mild comorbid disease: seven with hypertension, and one each with coronary heart disease, asthma, and seizure disorder. Three (6%) had moderate comorbid disease: one each with insulin-dependent diabetes mellitus, chronic renal failure, and coronary heart disease. There were no patients with severe comorbid disease. This was consistent with the median American Society of Anesthesiologists' physical status classification of P2 (scale P1–P6).
Radical hysterectomy and pelvic lymphadenectomy were completed in 46 of 48 (96%) patients, and abandoned in two patients with stage Ib2 disease for findings of grossly metastatic disease in the pelvic lymph nodes. None were abandoned because of technical complications. For patients who completed radical hysterectomy and pelvic lymphadenectomy, median blood loss was 800 mL (range 200–4500 mL). The median length of completed surgery for the 46 patients was 265 minutes (range 155–455 minutes). Intraoperative complications occurred in five (11%) patients. Four patients had blood loss in excess of 1500 mL, and one experienced a transient obturator nerve palsy that resolved spontaneously in the perioperative period. None experienced intraoperative urinary, bowel, or vascular injuries.
Surgicopathologic factors are given in Table 1. Importantly, over half of all patients had residual tumor in their hysterectomy specimen, and almost all had margins free of disease. The median lymph node yield was not limited by patient obesity. Postoperatively, patients remained in the hospital for a median of 5 days (range 4–15 days), and were able to tolerate a diet within a median of 2 days (range 1–5 days) following the procedure. Of the 46 patients in whom radical hysterectomy and pelvic lymphadenectomy were completed, acute surgical morbidity was acceptable, and experienced by nine patients (20%), and was mostly related to wound complications. Six cases (13%) of mild postoperative ileus, and no cases of severe ileus or bowel obstruction were noted. There were four cases (9%) of wound infection with or without wound separation, and three cases (7%) of wound separation without infection. One patient had a postoperative pelvic abscess. Chronic surgical morbidity was rare. Two patients experienced postoperative lymphedema (one mild and one severe, with one following adjuvant radiation) and one patient had prolonged (5 months) voiding dysfunction. There were no cases of urinary fistulae reported.
Adjuvant radiation therapy was administered to seven patients (15%) following radical hysterectomy and pelvic lymphadenectomy, including all six patients with lymph node metastasis, and one patient with extensive lymphovascular space involvement and deep stromal invasion of tumor. Four patients (9%) experienced disease recurrence (at 8, 26, 31, and 37 months), all of which were local. Two of these four patients received adjuvant radiation for metastatic disease in pelvic lymph nodes. A median follow-up of 36 months (range 2–119 months) resulted in the following: 39 patients (85%) have no evidence of disease, one (2%) is dead of disease, one (2%) is alive with disease at 14 months after radical hysterectomy and pelvic lymphadenectomy, and two patients (4%) are without disease following therapy for their local recurrence (16 months following pelvic exenteration and 78 months following radiation and chemotherapy). Three patients (7%) were lost to follow-up. The 5-year overall survival of this group is 84% (95% CI 70.9, 97.5), and the 5-year disease-free survival is 80% (95% CI 65.2, 93.8) (Figure 1).
During the study interval, 90 patients with stage Ib cervical cancer with a BMI over 30 kg/m2 and weight over 85 kg were treated with primary irradiation at Washington University Barnes-Jewish Hospital, St. Louis, Missouri. When 45 consecutive patients treated with radiation were compared with patients undergoing radical hysterectomy and pelvic lymphadenectomy, we found that patients in the radiation group were significantly more obese (mean BMI 41.0 kg/m2 compared with 36.8 kg/m2, 95% CI 1.3, 7.1, P = .006) and heavier (mean weight 109.4 kg compared with 97.5 kg, 95% CI 5.9, 18.0, P = .001) than those women undergoing surgery.
The prevalence of obesity in the United States has continued to increase in recent years. The Centers for Disease Control and Prevention recently reported the results of its Third National Health and Nutrition Examination Survey,10 which states that 36% of adult women are currently obese, a 25% increase in prevalence since 1980. In 1997, the World Health Organization11 attempted to standardize the definition of obesity as a BMI over 30 kg/m2, thereby facilitating the interpretation, comparison, and communication of data regarding obesity. Unfortunately, studies on the impact of obesity in abdominal surgery have used a variety of definitions of obesity, including percent of ideal body weight and absolute weight.5–7,12,13 These studies have demonstrated an increase in the risk of complications, including increased operative time and blood loss, as well as an increased risk of pulmonary compromise, venous thrombosis, and wound infection and separation. Pitkin12 reviewed more than 200 women weighing over 300 lbs who underwent abdominal hysterectomy, and found increased operative time and blood loss, as well as an exceedingly high rate of wound complications (29%) when compared with nonobese women undergoing the same procedure. Levrant et al5 reported a series of 54 obese patients with stage I cancer of the cervix. This group, compared with 69 nonobese patients who also underwent radical hysterectomy and pelvic lymphadenectomy with para-aortic lymph node sampling, had an increased operative time and blood loss without an increase in long-term complications or compromised survival. The definition of obesity used was 20% above ideal body weight, and their data cannot be converted to reflect the recent changes in the definition of obesity. Soisson et al6 described a series of 43 obese women (over 25% above ideal body weight and absolute weight over 80 kg) who underwent radical hysterectomy and pelvic lymphadenectomy. Compared with 277 nonobese patients, the obese group had a significantly higher blood loss and operative time, without an increase in serious complications or compromised survival. Approximately 50% of patients in this study were less than the absolute threshold of 86 kg employed in the current investigation. Finan et al7 stratified women with early-stage cervical cancer who underwent radical hysterectomy and pelvic lymphadenectomy into groups based on BMI. In the 42 women with a BMI over 30 kg/m2, a longer operative time was noted. Specific data regarding blood loss and surgicopathologic factors were not compared among these groups. In this investigation, we used the updated definition of obesity of BMI over 30 kg/m2, so as to eliminate nonobese patients from our study population. By using this strict definition, we believe we can present a more accurate representation of surgical outcomes in obese women.
Given that radical hysterectomy and radiation have been demonstrated to have equal efficacy in the treatment of invasive cervical cancer,2–4 the decision to pursue one treatment modality over another is generally based on an assessment of comparative morbidity and patient preference. Although no absolute selection criteria exist in the counseling of patients regarding the primary treatment for early cervical cancer, we generally offer either surgery or radiation to patients who are not opposed to surgery, whose body habitus and lesion size will not preclude the completion of radical hysterectomy, and who are without absolute medical contraindications to surgery. Absolute contraindications to surgery include bleeding disorders, recent (less than 6 months prior to diagnosis) myocardial infarction or thromboembolism, and end-stage renal or cardiopulmonary disease. Likewise, patients who are unwilling to receive blood or blood-product transfusions are not considered as candidates for surgical treatment. The majority of our patients did not have microinvasive tumors, and tumor size did not appear to be a major deterrent to offering radical surgery to obese women at our institution. We believe that the use of a Bookwalter self-retaining retractor, instruments of adequate length, and an incision that affords optimal exposure (vertical midline, Cherney, or Maylard) achieves the most effective surgical outcomes in obese women explored for radical hysterectomy and pelvic lymphadenectomy.
The minimal comorbidity in our obese population undergoing exploration supports our usage of careful selection criteria in offering radical surgery to obese women with cervical cancer. In reviewing the records from 45 consecutive women with a BMI over 30 kg/m2 and an absolute weight over 85 kg treated with radiation for their early cervical cancer, we identified six patients who were not offered surgery as a result of medical comorbidity. These comorbid conditions included one patient with a massive myocardial infarction 2 months prior to diagnosis, two with recent venous thromboembolism, one with a pulmonary embolism on warfarin anticoagulation, and two with severe pulmonary disease. Two patients were counseled against surgery because of refusal of blood transfusions. The limited number of patients apparently assigned to primary irradiation because of complicated medical issues suggests that severe comorbidities is only one factor among many, used in counseling obese women about cervical cancer treatment. Further evidence of selection biases are noted when comparing obese and nonobese women undergoing radical hysterectomy and pelvic lymphadenectomy, through comparison of BMI and weight of obese patients with early cervical cancer treated with radiation, with those treated with surgery. Our findings suggest that patients in the radiated group were significantly more obese than patients undergoing radical hysterectomy and pelvic lymphadenectomy. This supports the observation that not all obese patients with early-stage cervical cancer are offered the option of radical surgery. Therefore, comparing obese and nonobese groups who underwent radical hysterectomy and pelvic lymphadenectomy may lead to inappropriate conclusions about relative surgical morbidity, surgical resectability, and patient survival.
We have demonstrated that in this carefully selected, specifically defined group of obese women with early cervical cancer, radical hysterectomy with pelvic lymphadenectomy is feasible, safe, and efficacious. Radical hysterectomy was performed with excellent pathologic results, including adequate margins and a high lymph node yield. Postoperatively, our patients generally convalesced at rates consistent with that of nonobese women, and had mild complications, mostly related to wound infection or breakdown. The excellent overall and disease-free survival supports the adequacy of radical surgery in this group. Radical surgery is therefore an option in carefully selected obese women with early-stage cancer of the uterine cervix.
1. Greenlee RT, Murray T, Bolden S, Wingo PA. Cancer statistics, 2000. CA Cancer J Clin 2000;50:7–33.
2. Roddick JW Jr, Greenlaw RH. Treatment of cervical cancer. A randomized study of operation and radiation. Am J Obstet Gynecol 1971;109:754–64.
3. Newton M. Radical hysterectomy or radiotherapy for stage I cervical cancer. A prospective comparison with 5 and 10 years follow-up. Am J Obstet Gynecol 1975;123:535–42.
4. Hopkins MP, Morley GW. Radical hysterectomy versus radiation therapy for stage Ib squamous cell cancer of the cervix. Cancer 1991;68:272–7.
5. Levrant SG, Fruchter RG, Maiman M. Radical hysterectomy for cervical cancer: Morbidity and survival in relation to weight and age. Gynecol Oncol 1992;45:317–22.
6. Soisson AP, Soper JT, Berchuck A, Dodge R, Clarke-Pearson D. Radical hysterectomy in obese women. Obstet Gynecol 1992;80:940–3.
7. Finan MA, Hoffman MS, Chambers R, Fiorica JV, DeCesare S, Kline RC, et al. Body mass predicts the survival of patients with new International Federation of Gynecology and Obstetrics Stage Ib1 and Ib2 cervical carcinoma treated with radical hysterectomy. Cancer 1998;83:98–102.
8. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. J Chronic Dis 1987;40:373–83.
9. Piver MS, Rutledge F, Smith JP. Five classes of extended hysterectomy for women with cervical cancer. Obstet Gynecol 1974;44:265–72.
10. Centers for Disease Control and Prevention. Update: Prevalence of overweight among children, adolescents, and adults—United States, 1988–1994. MMWR Morb Mortal Wkly Rep 1997;46:199–202.
11. World Health Organization. Obesity: Preventing and managing the global epidemic—Report of a WHO consultation on obesity, June 3–5, 1997. Geneva, Switzerland: World Health Organization Press, 1998.
12. Pitkin RM. Abdominal hysterectomy in obese women. Surg Gynecol Obstet 1976;142:532–6.
13. Foley K, Lee RB. Surgical complications of obese patients with endometrial carcinoma. Gynecol Oncol 1990;39:171–4.