Key messages

• Not removing pelvic lymph nodes (which are part of the immune system and in the lower tummy) probably does affect survival compared with removing all pelvic lymph nodes in women without obvious disease spread and may reduce the risk of disease coming back.

• Not removing pelvic lymph nodes, or removing only the first draining lymph nodes (called sentinels), probably greatly reduces the risk of developing swelling in the legs after surgery compared with removing all pelvic nodes, or pelvic/para-aortic (which are higher in the tummy) nodes.

What is endometrial cancer and how is it treated?

Endometrial cancer affects the lining of the womb (uterus), known as the endometrium. It is the sixth most common cancer in females worldwide. In 2022, there were 420,242 cases of womb cancer, causing 97,704 deaths worldwide.

Most women with endometrial cancer have early-stage disease at diagnosis (only in the womb). Treatment involves hysterectomy (surgical removal of the womb and neck of womb) and removal of the fallopian tubes (which transport the egg from the ovaries to the womb) and ovaries (which produce eggs). Removal of lymph nodes (lymphadenectomy) in the pelvis (lower tummy) or para-aortic areas (area around the main blood vessels in the upper tummy) (or both), is performed to see if further treatment with anticancer medicine (chemotherapy) or high-dose x-rays (radiotherapy) is needed to reduce the risk of recurrence.

Why is this important for women with endometrial cancer?

The rate of lymph node involvement in early-stage disease is low and may be predicted by examining the womb under a microscope and running further tests following hysterectomy. Lymphadenectomy carries a risk of long-lasting lymphoedema (swelling in parts of the lower body). New techniques for detecting and removing just the first draining lymph nodes from each side of the womb (sentinel lymph node biopsy) can replace lymphadenectomy in detecting involved nodes accurately, but we do not know if sentinel lymph node biopsy is beneficial to women, despite it being widely used.

What did we want to find out?

We wanted to know if removing all pelvic, just sentinel or no apparently normal lymph nodes from pelvic/para-aortic areas was beneficial to women with presumed early-stage endometrial cancer and if there were any unwanted effects.

What did we do?

We searched for studies comparing removal of all pelvic, just sentinel or no apparently normal nodes from pelvic/para-aortic areas in women with presumed early-stage endometrial cancer. We compared and summarised the results, and rated our confidence in the evidence, based on factors including study methods and the number of women in the studies.

What did we find?

We found five studies (one is continuing) with 2074 women with endometrial cancer conducted in the UK, South Africa, Poland, New Zealand, Chile, Italy, Egypt and Brazil, and published between 2008 and 2023.

Pelvic lymphadenectomy versus no lymphadenectomy

• Women with no lymphadenectomy are probably as likely to survive up to three years (146/1000 deaths with pelvic lymphadenectomy versus 126/1000 deaths with no lymphadenectomy).

• Women with no lymphadenectomy are less likely to have recurrent disease by three years (205/1000 women with pelvic lymphadenectomy versus 164/1000 women with lymphadenectomy).

• Women with no lymphadenectomy may have fewer complications from surgical injury (38/1000 women with pelvic lymphadenectomy versus 26/1000 women with no lymphadenectomy) and probably have fewer complications due to surgery (e.g. infection/blood clots; 13/1000 women with pelvic lymphadenectomy versus 4/1000 women with no lymphadenectomy).

• Women with no lymphadenectomy probably have a much lower rate of lymphoedema by three years (61/1000 women with pelvic lymphadenectomy versus 7/1000 women with no lymphadenectomy) and likely have fewer lymph collections (cysts; 14/1000 women with pelvic lymphadenectomy versus 3/1000 women with no lymphadenectomy).

Sentinel lymph node biopsy versus pelvic/para-aortic lymphadenectomy

• Preliminary data from one study are very uncertain about deaths, surgical complications and quality of life.

• Sentinel lymph node biopsy probably reduces lymphoedema rates (92/1000 women with sentinel lymph node biopsy versus 306/1000 women with pelvic/para-aortic lymphadenectomy).

No lymphadenectomy versus pelvic/para-aortic lymphadenectomy

• We are very uncertain about deaths, surgical complications and quality of life.

• Pelvic/para-aortic lymphadenectomy may reduce lymphoedema (18/1000 women with no lymphadenectomy versus 250/1000 women with pelvic/para-aortic lymphadenectomy).

What are the limitations of the evidence?

Some studies were published before the modern classification of endometrial tumours, and more of these women would now be offered chemotherapy/radiotherapy, even without confirmed lymph node involvement. This may have affected the results.

We hoped to rank the effectiveness of these treatments, but there were not enough studies or results to do this. Further data from 10 ongoing studies are awaited.

How up to date is the evidence?

The review is current to 22 March 2024.