NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Holzheimer RG, Mannick JA, editors. Surgical Treatment: Evidence-Based and Problem-Oriented. Munich: Zuckschwerdt; 2001.

Cover of Surgical Treatment

Surgical Treatment: Evidence-Based and Problem-Oriented.

Show details

Locoregional recurrence of rectal cancer

, M.D. and , M.D.

Author Information

The literature on locoregional recurrence (LR) after a potentially “curative” resection of a rectal cancer is difficult to penetrate, particularly because many authors mix colonic and rectal cancer and primary rectal cancer with recurrent disease. LR after a rectal resection causes profound morbidity and impairs quality of life severely, usually leading to death of the patient. The incidence of LR after surgery alone varies considerably from median 19% (range 3–50%) in 52 individual series (McCall & Wattchow 1997) to 23–41% in the “surgery only” arm in RCTs on adjuvant therapy (radiation +/- chemotherapy; 16 studies) and to 28–31% in national audits (Norway). Generally, the LR rate can approximately be halved by adjuvant radiation therapy (pre- or postoperatively with or without supplementary chemotherapy) or by optimization of the surgical technique (Total Mesorectal Excision). LR rate is related to the surgeon's skill and experience.

Definition, reporting and extent of disease

It is well documented that LR rates depend on the definition employed and the subgroups studied (Marsh et al 1995) and that LR rates are probably underestimated in the abscence of routine autopsy.

Local recurrence is defined as evidence of recurrent disease within the pelvis after a surgical resection, including recurrence at the site of anastomosis and perineal wound.

LR rates must be related to resection type (R-classification) of the primary operation and should be reported in these categories:

  • LR occurring in isolation (including “true” anastomotic recurrence)
  • Distant recurrence (metastases) alone
  • Local plus distant recurrence
  • Total LR rate

If adjuvant radiotherapy is given, separate LR rates should be given for within and outside the radiation field.

Natural history of LR

Generally, LR of rectal cancer represents a local manifestation of a disseminated disease. However, more than 50% (53% in a national audit in Norway) of LR of rectal cancer occur as isolated deposits, and many patients (up to 30–50%) are reported to die without evidence of distant metastases. LR in the bowel anastomosis alone is a rare occurrence (approximately 10%). In around 10% of patients the LR develop distally to the bowel anastomosis demonstrating that metastases to the distal mesorectum cause recurrence.

Typically 55–80% present during the first 2 years after operation, rarely it may develop after 5 years. There appears to be a clear correlation between LR rates and survival (two studies); i.e. high LR rates reduce survival.

The majority of data on LR originates from referral hospitals and the problem of selection bias is apparent. There is a paucity of data in an unselected patient population as to the natural history of rectal cancer patients with isolated LR. The true proportion of patients with resectable LR is therefore unknown, in smaller series (where the potential for selection bias is apparent) the resectability rate is 50–70%.

Detection and preoperative assessment

Although autopsy studies have suggested that patients may die from disseminated disease with asymptomatic recurrence, all LR will eventually cause symptoms such as pelvic and/or perineal pain or rectal bleeding. In some patients serial CEA may precede symptomatic evidence of LR, but the role of CEA in follow up of such patients is still debated. Other serum markers offer no additional useful information.

Three research questions must be addressed; what is the role and potential benefit of routine follow-up (including serial CEA) after a curative (R0) resection in patients with rectal cancer, and if LR is suspected, which diagnostic modality gives the best diagnostic accuracy, and thirdly how best to stage the LR with the aim of surgical resection?


Most will agree that auditing of patients after a R0-resection for rectal cancer is required as a quality control instrument. The potential benefit (clinical effectiveness) of routine follow up incorporating laboratory tests (including CEA), imaging and endoscopy is more controversial. Six RCTs comparing a (more or less intense) follow up program with no follow up have been published in patients with colorectal cancer. A definite conclusion can as yet not be drawn, particularly because no study have addressed patients with rectal cancer only. (From a research perspective, patients with rectal cancer should be followed and assessed differently from colonic cancer). Even with close surveillance more than half of patients have diagnoses of LR after onset of symptoms, mainly perineal, low back or pelvic pain. Most pelvic recurrences can be reached by the exploring finger (after AR).

A balanced conclusion so far is that routine follow-up may identify recurrences at an earlier stage and that probably more operations (with a curative intent?) can be undertaken for recurrent disease. It is unlikely that survival in the follow up group is much better compared to that of the no follow up group but a small benefit from an “intensive” follow up program cannot be ruled out.

Diagnostic imaging

There are several apparent methodological problems in assessing the value of different imaging modalities in patients with LR:

  • The additional benefit of an imaging modality in the clinical setting is rarely assessed, i.e. were the investigation carried out in patients with symptoms (i.e. suspected LR) or in asymptomatic patients with high risk for recurrence? Management decisions are often not incorporated.
  • Most studies are small, largely retrospective in nature, and case selection and case mix (both primary and recurrent tumors) is often unclear.
  • A general lack of comparative studies and of “blinded” comparisons between methods.
  • A focus on the sensitivity for individual investigations. A high specificity (i.e. a low false positive rate) is in a clinical context more important.
  • There are many reviews, often non-systematic, relative to the number of primary research.
  • The continuing technical development of equipment and an alleged improved imaging quality render objective assessment of results from different time periods difficult.

Endoscopic and/or endoluminal ultrasound (US) is only relevant after an anterior resection (AR) and in women following abdominoperineal resection (APR).

There are several common features which limit the value of any imaging method:

  • Postoperative inflammatory scarring may persist for months, particularly so after APR.
  • Postirradiation changes (after adjuvant radiation therapy) is frequently seen in the presacral space and in the muscles and may lead to an erroneous diagnosis of LR.
  • None of the investigative modalities (US, CT, MR) can reliably differentiate between cancer, fibrosis and inflammation.
  • All modalities are inaccurate in assessing microscopic invasion, in detection of metastatic foci in small lymph nodes, and CT and MR in assessing the rectal anastomosis after AR.
  • Imaging is often assessed before radiotherapy is given and the actual importance of infiltration on imaging is not known.

In general terms, endoluminal US is more accurate than digital examination and endoscopy. CT and MR are equally accurate (accuracy rates of 65–90%), but MR is somewhat superior to CT in assessing invasion of nerves, muscle and bone.

Endorectal MR coil imaging, preoperative monoclonal antibody radioimmuno detection and of PET scanning must be considered as investigative tools. Their resolution is down to 1 cm but their true value in patients with LR cannot be assessed as yet. Radioimmuno detection has varying sensitivity, is not specific and its clinical effectiveness unproven. Research addressing the use of radioimmuno detection should focus on the potential for upstaging of LR, i.e. identification of extra tumor sites.

PET provides a functional image and is of interest for improved tissue characterization, but its anatomical resolution is so far poor. The Health Care Financing Administration (HCFA) in USA has in March 1999 agreed to cover the use of PET for detecting and localizing recurrence in colorectal cancer patients with rising CEA. The International Network of Agencies for Health Technology Assessment (INAHTA) in a joint project on PET concluded that the evidence of clinical utility is not established.

Fine needle aspiration cytology (FNAC) or fine needle core biopsy (FNB) can give a tissue verified diagnosis. The sensitivity is in expert hands high, but specificity (and predictive value of a negative test) low.

Preoperative staging

Few studies using strictly defined criteria exist. Fixety of the LR is assessed by digital eximination if appropriate. Extrapelvic disease, sciatica and lower extremity lymphedema, encasement of external or common iliac vessels, ureteric obstruction above the pelvic brim and sacral involvement above S3 (although some describe resection of S2) all denote irresectability. Generally all imaging modalities understage the recurrence in the order of 30%.


Most studies addressing potentially curative treatment are small hospital cohort or case series analyzed retrospectively and in which the selection of patients is unclear. Better studies are urgently needed, preferably cohort studies with prospective data collection. It is highly unlikely that a RCT, comparing surgery with other treatment modalities, ever will be carried out. The studies on palliative treatment are far better with several good RCTs and systematic reviews (SR) being published.

Nearly all studies focus on survival and very few have addressed quality of life issues using disease specific or generic instruments. Many authors have investigated isolated symptoms using, in most cases, unresearched and unvalidated questionnaires (Camilleri-Brennan & Steele 1998).

Potentially curative treatment

Many authors conclude that there are now improved rates of resectability and survival but the scientific basis for this is unclear particularly so on a population basis. It appears that the few patients “cured” for their LR have had resective surgery (although some authors use the phrase “potentially curative radiotherapy”). The research questions which must be addressed are whether adjuvant treatment (pre-, intraoperative (IORT) and/or postoperative radiation and chemotherapy) increases survival and/or improves quality of life. Secondly, newer surgical strategies in the form of radioimmuno guided surgery (RIGS) must be discussed.


Principally three options exist and can be combined; APR or a Hartmann's procedure (in patients with LR after a previous AR), pelvic exenteration or abdominosacral resection. Pelvic exenteration achieves adequate margins if the recurrence involves anterior structures only (vagina, prostate, bladder). For a posterior pelvic recurrence a sacral resection is often required. The type of operation must be tailored to the actual patient, and no general rule can be given.

There appears to be a consensus that involved tissue must be removed in continuity with a resection margin of 2 cm. Although sensible, no solid documentation for this 2 cm rule can be found. If this rule is strictly adhered to this means that resection of sacrum and/or piriformis muscle with part of the neural plexus must be carried out in the majority of patients.

The operation is difficult because of an absence of planes for dissection around the LR. This is particularly so after a TME operation. A “guidance document” outlining an evidence based service framework for colorectal cancer published by NHS/Department of Health in England recommend that such surgery should only be attempted by a specialist with experience of this type of work.

In radioimmuno guided surgery (RIGS) antibodies against tumor specific antigens is injected preoperatively and intraoperative scanning using a hand-held gamma detection probe carried out. There is anectodal evidence that upstaging by detection of more tumors sites can occur. The true value of RIGS is still not documented and good studies are awaited.

Radiation therapy

Based on the well documented and beneficial adjuvant effects of radiation therapy in patients with locally advanced primary rectal cancer there appears to be consensus that radiation therapy should be added to all patients having surgical resection for LR. There is no good evidence, however, suggesting that preoperative adjuvant radiation is better than postoperative, or that both pre- and postoperative radiation (“sandwich” technique) confer any benefit.

Given preoperatively it may occasionally downstage a tumor (i.e. reduce size and infiltration) so that resection can be carried out. Intraoperative radiation therapy (IORT), introduced in 1964, can be added to surgery, either in isolation or in combination with external radiation. Several small non-controlled studies have shown that IORT gives good local tumor control and may improve survival. IORTs potentially positive effects must be balanced against the negative local radiation effects, in particular nerve and bladder injury. A randomized study on IORT is urgently needed.

The popularity of adjuvant radiation therapy (pre- or postoperative) for primary rectal cancer may have a negative consequence in that the radiation dose which can be added for a LR must be reduced to a level where a tumoricidal effect can not be expected.


There is no evidence based on specifically designed or analyzed studies that chemotherapy adds any benefit in the patient with isolated LR operated on for cure. Many add chemotherapy (5-FU based) based on the circumstantial evidence that chemoradiation might be effective in reducing LR rates after primary rectal cancer resection.

The complication rates after surgery are high (26–100%) and postoperative mortality ranges from 0–10%. Five year survival rates in this highly selected group varies considerably, from around 15% to more than 50%. A few reports suggest that the survival curve shows a progressive decline even after 5 years leaving very few 10-year survivors.

Palliative treatment

Nearly all studies address patients with advanced metastatic disease. There is a paucity of data on the potentially positive effects (a survival benefit, symptom (pain) prevention or relief) after palliative treatment in patients with irresectable LR without apparent distant spread.


There is anecdotal evidence that a R1-R2 resection with supplementary radiotherapy (external or IORT) may occasionally give long term survivors and improve quality of life. Most reviews conclude that a palliative surgical resection (i.e. partial resection) leaving positive resection margins confer no positive effect. The occasional resectable LR in combination with a localized distant metastasis (liver or lung) represents a difficult management decision.

Radiation therapy

There is good evidence that radiation therapy reduces pelvic or perineal pain in 50– 90% of patients. A recent systematic review concludes, however, that the optimal dose fractionation schedule for the palliation of pelvic recurrence remains undefined (Wong et al. 1998). Most authors agree that symptomatic relief by irradiation is generally of good quality but short-lived; few patients experience long-term relief of pain (> 6–8 months). 4–7% develop bone metastases for which radiotherapy is effective. In isolated patients radiotherapy appears to prolong life, mainly in patients with LR who have not already received pre- or postoperative radiotherapy. Radiotherapy reduces rectal bleeding, whereas its effect on other symptoms (mucous discharge, etc) is more uncertain.


Chemotherapy increases survival time in advanced colorectal cancer when compared with supportive care alone. Chemotherapy given early in the course of the metastatic disease produces better outcome than chemotherapy given after symptoms have become severe, increasing survival by 3–6 months without significant adverse effects on quality of life. No reliable estimates of the cost-effectiveness of chemotherapy in advanced or recurrent disease are available. Regional (pelvic arterial) chemotherapy or isolated pelvic perfusion chemotherapy (the latter procedure requires an operative approach) has been claimed to be effective in palliation and, in some patients, shrinking the tumor but its palliative potential is unproven.

Other treatment alternatives

There is no evidence that immunotherapy, using interferon or other immune modulators, improves outcomes. The effects of radiosensitizers, brachytherapy (IORT or interstitial radiation placement) and hyperthermia are unproven and they should not be given outside protocols. Intraluminal tumor debulking using laser, cryotherapy, electrocoagulation and urologic resectoscope has been put forward as alternatives to diverting colostomy in patients with obstruction after LR following AR, but their role is still unclear.

The median life expectancy after locoregional recurrence treated for palliation is around 7–13 months and with only anecdotal patients surviving up to 5 years.


Most patients with LR will have advanced local and/or disseminated disease at the time of diagnosis. All imaging modalities understage the recurrence. A number of studies suggest that some LR can be resected with a curative intent and that long term survival can be achieved in a few patients. Additional radiation therapy should be given, whereas the use of adjuvant chemoradiation is based on circumstantial evidence. The minimum resection margin is not well defined. It is highly unlikely that a palliative resection (i.e. R1-R2) of the LR confer any benefit for the majority of patients. All patients with a irresectable LR should receive radiation therapy; the symptomatic effect is good but short lived. Chemotherapy should be given to those with advanced (i.e. metastatic disease).


Avradopoulos K A, Vezeridis M P, Wanebo H J. Pelvic exenteration for recurrent rectal cancer. Adv Surg. (1996);29:215–233. [PubMed: 8720005]
Blend M J, Abdel-Nabi H. New methods for the staging of colorectal cancer using noninvasive techniques. Semin Surg Oncol. (1996);12:252–263. [PubMed: 8829284]
Cunningham J D, Enker W, Cohen A. Salvage therapy for pelvic recurrence following curative rectal cancer resection. Dis Colon Rectum. (1997);40:393–400. [PubMed: 9106686]
Farouk R, Nelson H, Radice E, Mercill S, Gunderson L. Accuracy of computed tomography in determining resectability for locally advanced or recurrent colorectal cancer. Am J Surg. (1998);175:283–287. [PubMed: 9568652]
NHS Executive (1997) Guidance on commissioning cancer services. Improving outcomes in colorectal cancer. The manual. Department of Health (UK), Leeds .
Mehta S, Johnson R J, Schofield P F. Staging of colorectal cancer. Clin Radiol. (1994);49:515–523. [PubMed: 7955861]
Mitchell E P. Role of carcinoembryonic antigen in the management of advanced colorectal cancer. Semin Oncol. (1998);25:12–20. [PubMed: 9786312]
Moffat F L, Falk R E. Radical surgery for extensive rectal cancer: is it worthwhile? Recent Results Cancer Res. (1998);146:71–83. [PubMed: 9670251]
Stomper P C, D'Souza D J, Bakshi S P, Rodriguez-Bigas M, Bruke P A, Petrelli N J. Detection of pelvic recurrence of colorectal carcinoma: prospective, blinded comparison of TC-99m-IMMU-4 monoclonal antibody scanning and CT. Radiology. (1995);197:688–692. [PubMed: 7480740]
Thoeni R F. Colorectal cancer. Radiologic staging. Radiol Clin N Am. (1997);35:457–485. [PubMed: 9087214]
Turk P S, Wanebo H J. Results of surgical treatment of nonhepatic recurrence of colorectal carcinoma. Cancer. (1993);71:4267–4277. [PubMed: 8508389]
Tveit K M, Wiig J N, Olsen D R, Storaas A, Poulsen J P, Giercksky K E. Combined modality treatment including intraoperative radiotherapy in locally advanced and recurrent cancer. Radiother Oncol. (1997);44:277–282. [PubMed: 9380828]
Wiig J N, Wolff P A, Tveit K M, Giercksky K E. Location of pelvic recurrence after “curative” low anterior resection for rectal cancer. Eur J Surg Oncol. (1999);25:510–594. [PubMed: 10556005]
Wolff B G, Bolton J, Baum R, Chetanneau A, Pecking A, Serafini A N, Fischman A J, Hoover H C Jr, Klein J L, Wynant G E, Subramanian R, Goroff D K, Hanna M G Jr. Radioim-munoscintigraphy of recurrent, metastatic, or occult colorectal cancer with technetium Tc 99m 88BV59H21-2V67-68 (HumaSPECT-Tc), a totally human monoclonal antibody. Patient management benefit from a phase III multicenter study. Dis Colon Rectum. (1998);41:953–962. [PubMed: 9715149]
Copyright © 2001, W. Zuckschwerdt Verlag GmbH.
Bookshelf ID: NBK6972


  • PubReader
  • Print View
  • Cite this Page

Related information

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...