Objective
Although endometrial cancer commonly occurs in elderly women, little is known about the perioperative outcomes of the oldest women (>80 years of age) who are treated surgically.
Study Design
We performed an analysis of women ≥65 years of age with endometrial cancer who underwent hysterectomy from 1998–2007 and who were registered in the Nationwide Inpatient Sample.
Results
A total of 25,698 women were identified. Compared with women who were 65-69 years old, women who were ≥85 years old were more likely to have perioperative surgical complications (12% vs 17%), postoperative medical complications (24% vs 34%), and a longer length of stay (3 vs 5 days) and to require a transfusion (6% vs 10%; P < .05 for all). The perioperative mortality rate was 0.4% in women who were 65-69 years old compared with 1.6% in women who were ≥85 years old ( P < .0001).
Conclusion
The morbidity that is associated with surgery for endometrial cancer is significantly higher in women who are >80 years old, even after medical comorbidities have been considered.
Gynecologic cancers disproportionately affect elderly women. Currently, the elderly population is the fastest growing segment of the US population; the estimated number of persons ≥65 years old is projected to reach 70 million by 2030. With the aging of the elderly population, the oldest old, those ≥80 years old, is expected to see the greatest growth over the next 2 decades. The growing size of the oldest demographic and the burden of cancer that this age group faces pose important clinical and public health challenges. Despite the fact that elderly patients are more likely to die of their tumors, these patients often receive less aggressive treatment than their younger counterparts, even when the treatments are well tolerated.
Like other solid tumors of the abdomen and pelvis, surgery is the mainstay of treatment for most women with endometrial cancer. An abundance of evidence has reported that advanced age is a risk factor for adverse outcomes and death for patients who undergo surgery. Although medical comorbidities influence outcome, even after adjustment for coexisting illnesses, advanced age remains an independent risk factor for perioperative morbidity. The effect of age on outcome has been demonstrated for a number of high-risk cardiac and oncologic procedures as well as for lower-risk operations. For elderly patients, the benefits of surgery must be weighed carefully with the potential complications.
The ideal treatment of endometrial cancer in the oldest population is unknown. Despite the fact that endometrial cancer is predominately a disease of postmenopausal women, data that describe the morbidity of primary surgical treatment in elderly women is based primarily on small institutional series. The objective of our study was to perform a population-based analysis to determine the perioperative outcomes for the oldest (80-85 and >85 years) women with endometrial cancer who have been treated surgically. We examined the interaction between age and comorbidity in influencing outcome to determine the optimal primary treatment of the oldest women with endometrial cancer.
Materials and Methods
Data from the Nationwide Inpatient Sample (NIS) was used. NIS is a national database that is maintained by the Agency for Healthcare Research and Quality and contains a random sample of approximately 20% of discharges from hospitals within the United States. The sampling frame for NIS includes nonfederal, general and specialty-specific hospitals throughout the United States. Sampled hospitals include both academic and community facilities. The sampling scheme represents approximately 90% of hospitals in the United States. NIS is the largest all-payer inpatient care database, which in 2007 included 8 million hospital stays from 40 states. Institutional review board exemption was obtained from Columbia University.
Data from 1998-2007 were analyzed. We included all women who were ≥65 years old and who underwent total abdominal hysterectomy for cancer of the uterine corpus. Patients were classified on the basis of age into the following categories for analysis: 65-69 years, 70-74 years, 75-79 years, 80-84 years, and ≥85 years of age. Patients who underwent vaginal, laparoscopic, or supracervical (open or laparoscopic) were excluded. The performance of lymphadenectomy was noted for each subject.
The primary outcomes of the study were perioperative morbidity and death. Perioperative morbidity, which was based on ICD-9 coding, was classified into the following 3 categories as previously reported: (1) intraoperative complications (bladder injury, ureteral injury, intestinal injury, vascular injury, and other operative injury), (2) perioperative surgical complications (reoperation, postoperative hemorrhage, wound complication, and venous thromboembolism), and (3) medical complications (cardiovascular, pulmonary, gastrointestinal, renal, infectious, and neurologic). We calculated the proportion of patients who required blood transfusion. Length of stay was calculated by subtracting the admission day from the date of discharge. Perioperative death was defined as death during the hospitalization in which the patient underwent hysterectomy.
Predictor variables included demographic and clinical variables. Race was categorized as white, black, and other. Each patient’s household income was noted. Year of diagnosis was dichotomized into 2 roughly equal periods from 1998-2002 and 2003-2007. The hospitals in which patients were treated were characterized on the basis of location (urban, rural), region of the country (northeast, midwest, west, south), size (small, medium, large), and teaching status (teaching, nonteaching). Risk adjustment for the cohort was performed with the Charlson comorbidity index, which is a validated tool that uses medical comorbidities to calculate an overall score that estimates the risk of death. The Charlson index has been correlated with hospital deaths and outcome. The ICD-9 coding to define the Charlson index as reported by Deyo et al was used.
Frequency distributions between categoric variables were compared by the chi-square test; continuous variables were compared with 1-way analysis of variance. Logistic regression models were used to determine independent predictors of morbidity and death. We developed models for each of the complication classes and death, as described earlier. Length of stay was estimated with linear regression models with the use of ordinary least squares. All analyses were performed with SAS software (version 9.2; SAS Institute Inc, Cary, NC).
Results
A total of 25,698 women (7657 women [30.0%], 65-69 years old; 6634 women [25.8%], 70-74 years old; 5594 women [21.7%], 75-79 years old; 3789 women [14.7%], 80-84 years old; 2024 women [7.9%], ≥85 years old) were identified. Table 1 displays the clinical and demographic characteristics of the cohort. Compared with younger women, those women who were >85 years old more often were white, were diagnosed more frequently in the later years of study, resided more commonly in the Midwest, were treated more frequently at rural hospitals and nonteaching facilities, had more medical comorbidities, and were less likely to undergo lymphadenectomy ( P < .05 for all).
| Characteristic | Age | P value | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| 65-69 y (n = 1651; 30%) | 70-74 y (n = 6634; 25.8%) | 75-79 y (n = 5594; 21.7%) | 80-84 y (n = 3789; 14.7%) | ≥85 y (n = 2024; 7.9%) | |||||||
| n | % | n | % | n | % | n | % | n | % | ||
| Race | < .0001 | ||||||||||
| White | 4608 | 60.2 | 4163 | 62.8 | 2648 | 65.2 | 2493 | 65.8 | 1333 | 65.9 | |
| Black | 562 | 7.3 | 379 | 5.7 | 274 | 4.9 | 134 | 3.5 | 67 | 3.3 | |
| Other | 532 | 7.0 | 419 | 6.3 | 278 | 5.0 | 195 | 5.1 | 70 | 3.6 | |
| Unknown | 1955 | 25.5 | 1673 | 25.2 | 1394 | 24.9 | 967 | 25.5 | 554 | 27.4 | |
| Year of diagnosis | < .0001 | ||||||||||
| 1998-2002 | 3840 | 50.2 | 3604 | 54.3 | 3093 | 55.3 | 1975 | 52.1 | 1071 | 52.9 | |
| 2003-2007 | 3817 | 49.9 | 3030 | 45.7 | 2501 | 44.7 | 1814 | 47.9 | 953 | 47.1 | |
| Household income | .01 | ||||||||||
| Low | 1024 | 13.4 | 853 | 12.9 | 645 | 11.5 | 453 | 12.0 | 241 | 11.9 | |
| Medium | 1837 | 24.0 | 1573 | 23.7 | 1344 | 24.0 | 985 | 26.0 | 519 | 25.6 | |
| High | 2013 | 26.3 | 1848 | 27.9 | 1519 | 27.2 | 1047 | 27.6 | 567 | 28.0 | |
| Highest | 2633 | 34.4 | 2234 | 33.7 | 1983 | 35.5 | 1242 | 32.8 | 663 | 32.8 | |
| Unknown | 150 | 2.0 | 126 | 1.9 | 103 | 1.8 | 62 | 1.6 | 34 | 1.7 | |
| Region | .0006 | ||||||||||
| Northeast (1) | 1822 | 23.8 | 1578 | 23.8 | 1362 | 24.4 | 871 | 23.0 | 453 | 22.4 | |
| Midwest (2) | 1840 | 24.0 | 1672 | 25.2 | 1412 | 25.2 | 990 | 26.1 | 584 | 28.9 | |
| South (3) | 2574 | 33.6 | 2211 | 33.3 | 1767 | 31.6 | 1191 | 31.4 | 610 | 30.1 | |
| West (4) | 1421 | 18.6 | 1173 | 17.7 | 1053 | 18.8 | 737 | 19.5 | 377 | 18.6 | |
| Hospital location | .0002 | ||||||||||
| Rural | 665 | 8.7 | 551 | 8.3 | 519 | 9.3 | 331 | 8.7 | 245 | 12.1 | |
| Urban | 6988 | 91.3 | 6081 | 91.7 | 5072 | 90.7 | 3456 | 91.2 | 1778 | 87.9 | |
| Unknown | 4 | 0.1 | 2 | 0.0 | 3 | 0.1 | 2 | 0.1 | 1 | 0.1 | |
| Hospital type | .07 | ||||||||||
| Nonteaching (0) | 3076 | 40.2 | 2783 | 42.0 | 2348 | 42.0 | 1611 | 42.5 | 895 | 44.2 | |
| Teaching (1) | 4577 | 59.8 | 3849 | 58.0 | 3243 | 58.0 | 2176 | 57.4 | 1128 | 55.7 | |
| Unknown (9) | 4 | 0.1 | 2 | 0.0 | 3 | 0.1 | 2 | 0.1 | 1 | 0.1 | |
| Hospital size | .14 | ||||||||||
| Small | 789 | 10.3 | 727 | 11.0 | 553 | 9.9 | 430 | 11.4 | 257 | 12.7 | |
| Medium | 1606 | 21.0 | 1400 | 21.1 | 1205 | 21.5 | 811 | 21.4 | 408 | 20.2 | |
| Large | 5258 | 68.7 | 4505 | 67.9 | 3833 | 68.5 | 2546 | 67.2 | 1358 | 67.1 | |
| Unknown | 4 | 0.1 | 2 | 0.0 | 3 | 0.1 | 2 | 0.1 | 1 | 0.1 | |
| Charlson index | < .0001 | ||||||||||
| 1 | 3981 | 52.0 | 3274 | 49.4 | 2669 | 47.7 | 1738 | 45.9 | 891 | 44.0 | |
| 2 | 2248 | 29.4 | 1963 | 29.6 | 1654 | 29.6 | 1180 | 31.1 | 643 | 31.8 | |
| 3 | 1428 | 18.7 | 1397 | 21.1 | 1271 | 22.7 | 871 | 23.0 | 490 | 24.2 | |
| Lymph node dissection | < .0001 | ||||||||||
| No | 3349 | 43.7 | 2907 | 43.8 | 2521 | 45.1 | 1807 | 47.7 | 1155 | 57.1 | |
| Yes | 4308 | 56.3 | 3727 | 56.2 | 3073 | 56.2 | 1982 | 52.3 | 869 | 42.9 | |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
