Annals of Surgical Oncology Cite Track
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
10.1245/s10434-006-9266-7
Annals of Surgical Oncology 14:1081-1091 (2007)
© 2007 Society of Surgical Oncology
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Henry, L. R.
Right arrow Articles by Hoffman, J. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Henry, L. R.
Right arrow Articles by Hoffman, J. P.

Original Article

Resection of Isolated Pelvic Recurrences After Colorectal Surgery: Long-Term Results and Predictors of Improved Clinical Outcome

Leonard R. Henry, MD1, Elin Sigurdson, MD, PhD1, Eric A. Ross, PhD2, John S. Lee, MD1, James C. Watson, MD1, Jonathan D. Cheng, MD3, Gary M. Freedman, MD4, Andre Konski, MD4 and John P. Hoffman, MD1

1 Department of Surgical Oncology, Fox Chase Cancer Center, Philadelphia, Pennsylvania 19111, USA
2 Department of Biostatistics, Fox Chase Cancer Center, Philadelphia, Pennsylvania 19111, USA
3 Department of Medical Oncology, Fox Chase Cancer Center, Philadelphia, Pennsylvania 19111, USA
4 Department of Radiation Oncology, Fox Chase Cancer Center, Philadelphia, Pennsylvania 19111, USA

Correspondence: Address correspondence and reprint requests to: John P. Hoffman, MD, E-mail: JP_Hoffman{at}fccc.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Background: Recurrence in the pelvis after resection of a rectal or rectosigmoid cancer presents a dilemma. Resection offers the only reasonable probability for cure, but at the cost of marked perioperative morbidity and potential mortality. Clinical decision making remains difficult.

Methods: Patients who underwent resection with curative intent for isolated pelvic recurrences after curative colorectal surgery from 1988 through 2003 were reviewed retrospectively. Clinical and pathological factors, salvage operations, and complications were recorded. The primary measured outcome was overall survival. Univariate and multivariate analyses were conducted to identify prognostic factors of improved outcome.

Results: Ninety patients underwent an attempt at curative resection of a pelvic recurrence; median follow-up was 31 months. Complications occurred in 53% of patients. Operative mortality occurred in 4 (4.4%) of 90 patients. Median overall survival was 38 months, and estimated 5-year survival was 40%. A total of 51 of 86 patients had known recurrences (15 local, 16 distant, 20 both). Multivariate analysis revealed that preoperative carcinoembryonic antigen level and final margin status were statistically significant predictors of outcome.

Conclusions: The resection of pelvic recurrences after colorectal surgery for cancer can be performed with low mortality and good long-term outcome; however, morbidity from such procedures is high. Low preoperative carcinoembryonic antigen and negative margin of resection predict improved survival.

Key Words: Colorectal cancer • Pelvis • Recurrence • Resection


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Colorectal cancer is the third leading cause of cancer mortality in the United States, with 150,000 new cases annually resulting in nearly 57,000 deaths. Rectal cancers accounted for approximately a third of this incidence and 8500 deaths in 2002.1,2 Surgical resection represents the cornerstone of effective therapy for primary disease.

The disease of 50% of patients will recur after definitive resection of a rectal cancer. Of disease that fails to respond to therapy, approximately 70% will be solitary and distributed equally among the lungs, liver, and pelvis.3 Surgical series report local recurrence after rectal cancer resection to occur in 4% to 38% of patients.4,5 If patients with local recurrence remain untreated, they will survive less than a year.6 Chemoradiation can prolong survival for to 12 to 15 months,7 with occasional long-term survival reported.8 Complete surgical resection (with or without additional modalities) represents the best chance for long-term survival and improved quality of life.9,10

Gunderson and Sosin6 found that among patients undergoing a second-look laparotomy after rectal cancer resection, nearly 50% of patients with recurrence had disease confined to the pelvis. Additional autopsy data suggest that in as many as 33% of patients dying with recurrent rectal cancer, the pelvis was the sole site of disease.11 These data suggest that depending on the timing of recurrent cancer diagnosis, one-third to one-half of persons with a local recurrence are at least potentially curable with complete surgical resection. However, the use of surgery in this setting is not universally accepted. Bozetti et al.12 reported that <10% of patients with resection of a pelvic recurrence benefited from operation, with only 19% experiencing 5-year survival in those undergoing R0 resection. They do not consider surgery as the primary modality in these patients.

Enthusiasm for attempting surgical resection of local recurrences in the pelvis must be tempered with the knowledge that some patients will develop metastatic disease and may not derive important benefit from their resection. Highly selected series of patients undergoing curative resections report that margin-negative (R0) resection is accomplished in 36% to 83% of patients8,13–20 with corresponding 5-year survivals in 18% to 50%.3,8,14,16,17,20–28

Balanced against the above rates of surgical success is the reality that attempted resections of pelvic recurrences are often extensive in nature, and they are conducted on heavily pretreated patients. Morbidity is frequent, and operative mortality risk can be as high as 9% of patients.13,22,24 Striking a favorable risk/benefit ratio in this group of patients requires optimal patient selection, exposing only those with a good probability of benefit to risks of extirpative surgery.

Several authors have reported experiences with this difficult group of patients and have identified predictors of both margin-negative resections and survival.8,13–30 Our center’s initial experience with 19 resections of pelvic recurrences was published in 1993.31 We now report our updated series of 90 patients over a 15-year period with the longer follow-up.


   METHODS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Operative logs, tumor registry, and coding data were queried to identify patients undergoing resection of a pelvic recurrence after colorectal surgery for rectal and sigmoid primary cancers from 1988 through 2003. Although we do accept relief of intractable pain as an indication for operation in these patients, palliation was not an end point in this study, and patients submitted to resection for this intent were excluded. Additionally excluded were patients with known preoperative unresected metastatic disease.

Retrospective analysis was performed. Collected data included demographic information, clinico-pathological data regarding the primary and recurrent tumors and their management, salvage operations, and operative complications. The primary measured end point was survival, which was estimated by the Kaplan-Meier method. Univariate and multivariate analyses of the effect of covariates on the end point were analyzed by the log rank test and the Cox proportional hazard model. A forward stepwise variable selection procedure was used to identify a parsimonious model. Differences between groups of patients were determined by the Fisher exact or Kruskall-Wallis tests. All statistical analyses were conducted with SAS software (SAS Institute, Cary, NC). A P value of .05 was considered significant.

Definitions
Patients who had undergone previous resection of a recurrence of disease or primary tumor chemotherapy or radiotherapy for a presumed unresectable recurrence outside our facility were considered to have had prior definitive treatment for their recurrence. Primary cancers arising ≥12 cm from the anal verge were considered sigmoid colon primaries. Pre-operative carcinoembryonic antigen (CEA) was recorded if documented within 2 months of initiation of definitive therapy. Disease-free interval was calculated from the time of initial resection to diagnosis of recurrence. In patients who experienced recurrence of disease and who underwent a curative attempt at resection outside our hospital, the disease-free interval was calculated from the time of the recurrent resection to the diagnosis of the second recurrence to be treated at our center. Tumors classified as having adverse histology included those that were of high grade, that were mucinous, or that had lymphatic, venous, or perineural invasion.

Recurrences were stratified by the initial cancer resection (restorative vs. nonrestorative) and the extent of resection required for recurrent tumor extirpation (limited vs. extended). Limited resections were defined as recurrent tumors confined to the rectal wall, or solitary excision of a recurrence not requiring excision of adjacent organs or tissues. Recurrent tumor size was obtained from preoperative imaging studies or from the pathological report.

The type of salvage operation for resection of recurrence (local, low anterior resection [LAR], abdominoperineal resection [APR], pelvic exenteration) was determined from the operative report. Both early and late postoperative complications were documented. Operative mortality was defined as death within 30 days of operation or during the initial hospitalization.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Patients and Primary Tumor Management
Ninety patients met the study’s inclusion criteria (56 men, 34 women) with a median age at recurrent resection of 62 years (range, 33–91 years). Median follow up for the entire group was 31 months (range, 0–162 months) and 47 months in those alive at last follow-up (range, 4–162 months). A total of 72 patients (80%) had their initial resection for colorectal cancer outside our facility. The primary tumor resection was a local excision in 11 (12%), anterior resection in 62 (69%), and APR in 17 (19%). The operation for the primary cancer was restorative in 64 (71%) cases and nonrestorative in 26 (29%). Twelve had margin-positive resections, nine of which were believed to be microscopically positive and three that left gross disease behind. The primary tumor location, stage, and management are summarized in Table 1Go.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Characteristics of 90 primary colorectal tumors
 
Previous Recurrent Management
Nineteen patients underwent operation for a pelvic recurrence before their resection at our facility. Of these, 13 underwent an exploratory operation without removal of the tumor, which consisted of tumor biopsy and/or stoma creation. Six operations were judged to have been conducted with curative intent (three LAR, three APR). An additional 12 patients had recurrence that was deemed unresectable (at outside facilities) and that was initially treated with chemotherapy or chemoradiotherapy. These 18 patients treated definitively for recurrence before referral to our center were separated for analysis and are referred to as the prior treatment group.

Clinical Presentation
Adequate information was available to assess the clinical presentation of recurrence in 84 patients. Of these, 25 patients (30%) were detected through surveillance, and 59 (70%) complained of symptoms at surgical evaluation (Table 2Go). Median disease-free interval for the entire group was 17.5 months (range, 3–170 months).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Recurrent presentation of 84 cases of colorectal cancer
 
Salvage Procedures
Five staff surgical oncologists performed salvage procedures with assistance from additional subspecialists as required. Three patients (3.3%) had extrapelvic disease identified at exploration (two liver, one serosal implant). All three underwent resection of the extrapelvic site and resection of the recurrent disease. The two with liver metastasis died at 22 months (dNED) and 34 months (died of disease). The patient with a serosal implant experienced recurrence of disease at 9 months and died of disease at 14 months.

The salvage operations performed are listed in Table 3Go. The four local resections for recurrences were composed of two perineal recurrences, one sacral resection, and one resection of an enlarged replaced internal iliac node. Even in cases of repeat LAR or APR, resection of additional organs or tissue was commonly required. Overall, resections were of a limited nature in 27 cases (30%) and extended in 63 (70%). In addition to rectum or tumor, the most common organs resected were bladder (n = 34), prostate (n = 27), uterus, fallopian tubes, or ovaries (n = 14), bowel (n = 13), vagina (n = 12), and sacrum (n = 2). There was no statistically significant difference in recurrent resection type (limited vs. extended) based on the primary cancer resection type (restorative vs. nonrestorative). Tissue expanders were placed in 27 (30%) of cases to protect bowel from planned radiation, a technique previously reported from our center.32,33 Flap (muscle or myocutaneous) closure of the operative defect or perineum was required in 12 resections (13.3%).


View this table:
[in this window]
[in a new window]

  		TABLE 3. Salvage procedures and additional therapy
 
Margins
A negative resection margin (R0) was obtained in 51 patients (56.7%). Twenty-two patients (24%) and 17 patients (19%) had R1 and R2 resections, respectively. The most common areas of positive margins were posterior (10 of 39, 25.6%), lateral (10 of 39, 25.6%), or a combination of both (16 of 39, 41.0%).

To identify associations between positive resection margins and other covariates, groups with (R+) and without (R0) positive margins were evaluated for differences. Margin negativity was obtained more frequently in patients with sigmoid primary tumor (82%) versus those in the prior treatment group (44%) (P = .01) and rectal primary tumor (38%) (P = .0006). Negative margins were obtained in 70% of patients after recurrence after local excision, but this was not statistically different from the other groups. The following factors were significantly associated with a positive margin: prior nonrestorative resection (R+, 69%) versus prior restorative resection (R+, 33%, P = .002), those undergoing extended resections (R+, 59%) versus those with limited resections (R+, 7%, P ≤ .0001), pain at presentation (R+, 58%) versus those without pain (R+, 32%, P = .04), and those with recurrent tumors ≥5 cm (R+, 69%) versus those with smaller tumors (R+, 24%, P = .0002). Interestingly, the recovery of more lymph nodes in the resected specimen was significantly associated with the margin-negative group (P = .0001). As would be expected, those with margin-positive resection were more likely to receive additional radiotherapy (84% vs. 46%, P = .0003).

Additional Treatment
Additional chemotherapy and radiotherapy for the recurrence are summarized in Table 3Go. The chemotherapy was nearly universally based on 5-fluorouracil with infrequent usage of other agents (irinotecan in five, oxaliplatin in two). Postoperative treatment was more common than induction therapy. Six patients did not receive additional chemotherapy because of postoperative complications. Four additional patients refused chemotherapy after surgery, and two were believed to be overly deconditioned after surgical recovery to tolerate chemotherapy.

Fifty-six (64%) of 88 patients received additional radiotherapy. Brachytherapy (24 cases) or intraoperative electron-beam radiotherapy (IORT) (8 cases) were used in 32 (36%) of all cases, and 23 (59%) of 39 patients with either microscopic or grossly positive margins. Some received a combination of brachytherapy (or IORT) in addition to external-beam therapy. Follow-up time of patients in the brachytherapy group (median 30 months) was similar to those receiving external-beam therapy alone (median 33 months). When we used local control as an end point, there was no difference found between those receiving brachytherapy (or IORT) versus those receiving external-beam therapy alone, stratified for margin-negative resections. In patients with positive resection margins (R1 and R2), local control was maintained in 6 (60%) of 10 receiving external-beam therapy alone versus 7 (30%) of 23 who received brachytherapy or IORT. This difference did not reach statistical significance (P = .14). Overall, 42.9% of patients received trimodal therapy (surgery, chemotherapy, and radiotherapy) for their recurrence.

Deaths and Complications
Four patients died as a result of complications from their procedure. Two were less than 30 days from operation: one died of a massive cardiovascular accident in the perioperative period, and one died of postoperative bowel infarction. Two additional operative mortalities occurred after 30 days. One experienced a postoperative myocardial infarction with subsequent congestive heart failure, and one patient had a long course of acute respiratory distress syndrome and eventually died of multisystem organ failure.

Eighty-seven records were sufficient to document postoperative complications. Of these, 46 patients (53%) had 67 complications (49 early, 18 late). The most frequent complication type was infectious (27.6%), followed by wound complication (excluding infection), cardiopulmonary (excluding pneumonia), and gastrointestinal. Anastomotic leaks or fistulas occurred in six patients (Table 4Go).


View this table:
[in this window]
[in a new window]

  		TABLE 4. Operative complicationsa
 
Survival
Follow-up was complete to death, of at least 5 years’ duration, or up to date within the past year in 86 patients (96%). At a median follow-up of 31 months for the entire series and 47 months in survivors, 40 patients (44%) had died from disease, 13 (14%) from other causes, and 6 (7%) from indeterminate causes (2 of whom were known to have had recurrence); and 6 (7%) were alive with disease. Twenty-five (28%) patients were alive and without disease. Median and 5-year survival and disease-free survival for the group was 38 and 19 months and 40% and 32%, respectively (Fig. 1Go).


Figure 1
View larger version (9K):
[in this window]
[in a new window]

  		FIG. 1. Kaplan-Meier estimation of overall survival in 90 patients submitted to attempted curative resection of pelvic recurrence of disease.

 
By univariate analysis of the entire group, a recurrence after prior local excision (compared with the group with prior treatment, Fig. 2Go), prior restorative resection, the absence of pain at recurrence, CEA level ≤20 ng/dL, tumor size of <5 cm, limited resection, and negative margin status were all statistically significant factors for improved overall survival (Table 5Go). Factors analyzed for impact on survival and found not to be significant were age, sex, stage of primary cancer, operative site of primary tumor resection (Fox Chase Cancer Center vs. outside), margins of primary tumor resection, adjuvant therapies for both the primary and recurrent cancer, disease-free interval, the presence or absence of symptoms at recurrence, recurrent tumor grade or adverse histology, or the number of modalities used in recurrence treatment. Multivariate analysis of factors identified as significant by univariate analysis revealed that only preoperative CEA ≤20 ng/mL, and margin-negative resection retained statistical significance. No patient with an R2 resection or preoperative CEA >20 mg/dL was alive at 5-year follow-up (Figs. 3Go, 4Go).


Figure 2
View larger version (14K):
[in this window]
[in a new window]

  		FIG. 2. Overall survival by treatment group. Local, recurrence after local excision; Prior Trt, those with recurrences treated with chemoradiation or attempted curative surgery before referral; Rectal, recurrences with primary tumor <12 cm from anal verge; Sigmoid, recurrences with primary ≥12 cm from anal verge.

 

View this table:
[in this window]
[in a new window]

  		TABLE 5. Significant covariates associated with survival and local control in univariate analysis
 

Figure 3
View larger version (11K):
[in this window]
[in a new window]

  		FIG. 3. Overall survival by preoperative carcinoembryonic antigen (CEA) level: <20 ng/dL versus ≥20 ng/dL.

 

Figure 4
View larger version (13K):
[in this window]
[in a new window]

  		FIG. 4. Overall survival by margin status of resected specimen. R0, negative for disease; R1, microscopically positive for disease; R2, gross residual disease.

 
Twenty-two (24%) of 90 patients did not survive for 18 months after resection. This group, when compared with those with longer survival, had a significantly higher percentage of the following: patients who had radiation with their primary cancer management, pain associated with their recurrence, a preoperative CEA >20 ng/dL, tumors >5 cm, extended resections, and positive margins at recurrent resection (Table 6Go).


View this table:
[in this window]
[in a new window]

  		TABLE 6. Significant differences between groups according to survival time
 
Failure Patterns
Fifty-one of 86 patients were known to have developed additional disease after resection of recurrent disease. Of these, treatment failed locally in 15 and the patient developed a second pelvic recurrence, 16 developed distant disease, and 20 experienced both. Thus, the local control rate was 59.3%. Median follow-up for those with and without local control were 72 and 30 months, respectively. Univariate analysis of factors found to significantly predict patients achieving local control after recurrent resection are listed in Table 5Go.

Eleven patients of the 35 whose disease either initially or eventually failed to respond locally underwent additional attempts at curative resection of their second recurrence. Combining these 11 patients with the 6 who underwent a prior resection for recurrence yielded 17 patients in this series who had an attempted curative resection of a second recurrence. Median survival for this group was 23 months, and 5-year estimated survival was 25% (Fig. 5Go). Of the patients who ultimately experienced systemic failure, five underwent lung resection for metastasis. One remained alive with disease at 9 months, and the remaining were all dead between 11 and 40 months.


Figure 5
View larger version (9K):
[in this window]
[in a new window]

  		FIG. 5. Overall survival in 17 patients undergoing attempted curative resection of a second recurrence in the pelvis.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Our selection of patients for attempted curative surgical resection of a pelvic recurrence is based on a belief that no extrapelvic disease is present and that the recurrent tumor can be removed with good oncologic outcome. In our series, three patients with unanticipated, small, solitary extrapelvic metastasis identified at laparotomy underwent combined resection of their pelvic recurrence in combination with the extrapelvic site. None survived 5 years. This small experience precludes making any conclusion regarding the use of proceeding with resections of the recurrence or metastasis in this setting. Ultimately, the surgeon’s judgment will be informed by factors such as metastatic location and tumor burden, the probability of a favorable oncologic outcome, and expected additional morbidity to be incurred, as well as any anticipated palliative benefit.

Preoperative staging consists of computed tomographic scanning of the chest, abdomen and pelvis. Additional imaging studies are obtained at the surgeon’s discretion. We have found magnetic resonance image useful when further anatomic detail is desired, particularly in evaluating involvement of the sacrum and sciatic nerves and in evaluating liver lesions in conjunction with the pelvic recurrence. Because of the time period, very few patients in this study underwent functional imaging with positron emission tomography (PET) before operation. PET scanning has been demonstrated to increase sensitivity in detecting metastasis in this setting27,34 which may be missed by more conventional imaging. Although we now often obtain preoperative PET scans in these patients, no conclusions regarding the usefulness of PET scans can be drawn from this series.

Assuming no marked extrapelvic disease is present, we view certain anatomic factors as contraindications to attempted curative resection. Diffuse sidewall involvement that would likely result in multiple areas of positive margin are unlikely, in our opinion, like that of others,23 to obtain good survival benefit from resection. In addition, sacral involvement above the S2 level contraindicates resection because of the functional sequelae of higher resections and the small likelihood of obtaining clear margins. Although we did not distinguish between patterns of pain in this study, it is our experience that pain in the sciatic nerve distribution is unlikely to be resectable with negative margins and unlikely to be palliated with surgery. We do not view hydronephrosis to preclude a curative resection.35

We use the above criteria to report favorable results with resection in this large, mature series of a select group of patients to include a 40% long-term survival, a curative resection rate of >55%, and a local control rate approaching 60%. These results compare reasonably to other series with regards to these success measures.8,13–18,20–24 Lopez-Kostner et al.28 recently reported a superior 5-year survival of 50%, but their series included a very high percentage of patients who experienced recurrence of disease after local excisions.

Our high salvage rate is partly due to patients having resections after recurrence after local excisions, but the percentage of these patients is relatively small in this series. We have also included patients with primary tumors in the sigmoid. These were also demonstrated here to do quite well when treated for local recurrence. The reasons for success in resecting recurrences from more proximal primary cancers are likely multifactorial and include limited prior radiation exposure, prior restorative resections, and disease that recurs in the soft tissues of the pelvis rather than the bony pelvis.

We obtained a high rate of margin-negative resections. We did not attempt to rigorously categorize these tumors with regard to location within the pelvis (e.g., posterior, lateral, central) because we were not confident that we could accurately do so. However, our data suggest that most of the recurrences were located centrally or anteriorly in the pelvis and therefore were amenable to margin negative resection. This interpretation is based on the high number of patients previously treated with rectal restoration, many of whom had recurrences located in the peria-nastamotic area (data not shown), and the high percentage of exenterations performed.

The ability to obtain a negative margin is of paramount importance in this group of patients. Our data show that 5-year survival can be anticipated in >60% of patients with an R0 resection, but <20% of those with even a microscopically positive margin. We demonstrated no 5-year survivors in patients with gross residual disease after resection. The ability to predict margins preoperatively (in addition to the assessment of imaging studies) would be of great value. In our study, multiple covariates were associated with a positive resection margin in >50% of cases (see Results). This information may prove useful for patient selection for surgery, operative planning, and coordination of care with radiation oncologists.

The prognostic value of the preoperative CEA level (and preoperative CEA doubling time) has been previously established.13,23,29 Our study validates its prognostic importance. The median survival in the increased CEA group of 16 months is even less than those with grossly positive margins. Further, it is striking that only 36% of patients with a preoperative CEA >20 ng/mL survived >18 months and totaled only 8% of all patients in the group with the longer survival. In addition, the preoperative CEA level was highly significant in univariate analysis and remained significant within the multivariate model. Because most of these patients were referred from outside facilities and the methods of follow-up were not clear, we cannot judge the value of serial CEA in identifying recurrences in their potentially curable stage. However, it is clear from our data (and that of others) that CEA determination has prognostic importance and may be useful to help judge curability and plan adjuvant or induction therapy.

Despite our encouraging overall survival reported here, nearly one-quarter of our patients died within ≤18 months after resection. The differences identified between this group and those with longer survival were similar to the results of the univariate survival analysis for the entire cohort, with the addition that those who received radiotherapy with the primary cancer were three times more likely to have died in <18 months after recurrent resection. The cause of this may be multifactorial, with higher stage primary tumors, a more aggressive tumor biology, or a limit in administration of adjuvant radiation after recurrent resection all being possible etiologies. When considering predictors that can be identified preoperatively, only a preoperative CEA >20 ng/mL predicted death within 18 months in most patients. However, because no single factor identified (including increased preoperative CEA level) precluded a longer survival, we would urge caution in considering any single prognostic factor as a contraindication to resection.

Complications of therapy in this study were common, but mortality was acceptable. We have included those events occurring after 30 days, which may represent complications related to multimodality therapy rather than surgery alone. Of importance is the observation that >20% of patients in whom adjuvant chemotherapy was recommended, it was never received for reasons that may be related to the resection. For this reason, when our multidisciplinary recommendation is for multimodality therapy, it is generally our bias to provide the treatment before surgery.

In conclusion, our series demonstrates that a high rate of salvage treatment can be obtained with limited mortality in patients with recurrent colorectal cancer in the pelvis. The morbidity of such resections is high and alters postoperative planned therapy in more than one-fifth of patients. Margins of resection and preoperative CEA level predict overall survival. We have additionally identified several associations with margin-negative resection and local control that may aid in treatment planning. Resection of second recurrences in the pelvis may be advantageous in otherwise favorable patients.


   ACKNOWLEDGMENTS
 
Presented in part at the Society of Surgical Oncology, 58th Annual Cancer Symposium. Atlanta, GA, March 3–6, 2005.

Received for publication March 3, 2005. Accepted for publication October 5, 2006.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. American Cancer Society. Cancer facts and figures, 2002. Available at: http://www.cancer.org.
  2. American Cancer Society. Cancer facts and figures, 2004. Available at: http://www.cancer.org.
  3. Tepper JE, O’Connell M, Hollis D, Niedzwiecki D, Cooke E, Mayer RJ. Analysis of surgical salvage after failure of primary therapy in rectal cancer: results from intergroup study 0114. J Clin Oncol 2003; 21:3623–8.[Abstract/Free Full Text]
  4. Karanjia ND, Schache DJ, North WR, Heald RJ. "Close shave" in anterior resection. Br J Surg 1990; 77:510–2.[Medline]
  5. Carlsson U, Lasson A, Ekelund G. Recurrence rates after curative surgery for rectal carcinoma, with special reference to their accuracy. Dis Colon Rectum 1987; 30:431–4.[CrossRef][Medline]
  6. Gunderson LL, Sosin H. Areas of failure found at reoperation following "curative surgery" for adenocarcinoma of the rectum. Cancer 1974; 34:1278–92.[CrossRef][Medline]
  7. Wanebo HJ, Gaker DL, Whitehill R, Morgan RF, Constable WC. Pelvic recurrence of rectal cancer. Options for curative resection. Ann Surg 1987; 205:482–95.[Medline]
  8. Garcia-Aguilar J, Cromwell JW, Claudio M, Lee SH, Madoff RD, Rothenberger DA. Treatment of locally recurrent rectal cancer. Dis Colon Rectum 2001; 44:1743–8.[CrossRef][Medline]
  9. Esnaola NF, Cantor ML, Johnson AT, et al. Pain and quality of life after treatment in patients with locally recurrent rectal cancer. J Clin Oncol 2002; 20:4361–7.[Abstract/Free Full Text]
  10. Miner TJ, Jaques DP, Paty PB, Guillem JG, Wong WD. Symptom control in patients with locally recurrent rectal cancer. Ann Surg Oncol 2003; 10:72–9.[Abstract/Free Full Text]
  11. Welch JP, Donaldson GA. The clinical correlation of an autopsy study of recurrent colorectal cancer. Ann Surg 1979; 189:496–502.[Medline]
  12. Bozetti F, Bertario L, Rossetti C, et al. Surgical treatment of locally recurrent rectal carcinoma. Dis Colon Rectum 1997; 40:1421–4.[CrossRef][Medline]
  13. Wanebo HJ, Koness RJ, Vezeridis MP, Cohen SI, Wrobleski DE. Pelvic resection of recurrent rectal cancer. Ann Surg 1994; 220:586–97.[CrossRef][Medline]
  14. Hahnloser D, Nelson H, Gunderson LL, et al. Curative potential of multimodality therapy for locally recurrent rectal cancer. Ann Surg 2003; 237:502–8.[CrossRef][Medline]
  15. Cunningham JD, Enker W, Cohen A. Salvage therapy for pelvic recurrence following curative rectal cancer resection. Dis Colon Rectum 1997; 40:393–400.[CrossRef][Medline]
  16. Shoup M, Guillem JG, Alektiar KM, et al. Predictors of survival in recurrent rectal cancer after resection and intraoperative radiotherapy. Dis Colon Rectum 2002; 45:585–92.[CrossRef][Medline]
  17. Law WL, Chu KW. Resection of local recurrence of rectal cancer: results. World J Surg 2000; 24:486–90.[CrossRef][Medline]
  18. Kakuda JT, Lamont JP, Chu DZJ, Paz IB. The role of pelvic exenteration in the management of recurrent rectal cancer. Am J Surg 2003; 186:660–4.[CrossRef][Medline]
  19. Ogunbiyi OA, McKenna K, Birnbaum EH, Fleshman JW, Kodner IJ. Aggressive surgical management of recurrent rectal cancer—is it worthwhile. Dis Colon Rectum 1997; 40:150–5.[CrossRef][Medline]
  20. Salo JC, Paty PB, Guillem J, Minsky BD, Harrison LB, Cohen AM. Surgical salvage of recurrent rectal carcinoma after curative resection: a 10-year experience. Ann Surg Oncol 1999; 6:171–7.[Abstract]
  21. Suzuki K, Dozois R, Devine RM, et al. Curative reoperations for locally recurrent rectal cancer. Dis Colon Rectum 1996; 39:730–6.[CrossRef][Medline]
  22. Hashiguchi Y, Sekine T, Sakamoto H, et al. Intraoperative irradiation after surgery for locally recurrent rectal cancer. Dis Colon Rectum 1999; 42:886–95.[CrossRef][Medline]
  23. Yamada K, Ishizawa T, Niwa K, Chuman Y, Akiba S, Aikou T. Patterns of pelvic invasion are prognostic in the treatment of locally recurrent rectal cancer. Br J Surg 2001; 88:988–93.[CrossRef][Medline]
  24. Saito N, Koda K, Takiguchi N, et al. Curative surgery for local pelvic recurrence of rectal cancer. Dig Surg 2003; 20:192–200.[CrossRef][Medline]
  25. Gagliardi G, Hawley PR, Hershman MJ, Arnott SJ. Prognostic factors in surgery for local recurrence of rectal cancer. Br J Surg 1995; 82:1401–5.[Medline]
  26. Delpero JR, Pol B, Le Treut YP, et al. Surgical resection of locally recurrent colorectal adenocarcinoma. Br J Surg 1998; 85:372–6.[CrossRef][Medline]
  27. Hugier M, Houry S, Barrier A. Local recurrence of cancer of the rectum. Am J Surg 2001; 182:437–9.[CrossRef][Medline]
  28. Lopez-Kostner F, Fazio VW, Vignali A, Rybicki LA, Lavery IC. Locally recurrent rectal cancer: predictors and success of salvage surgery. Dis Colon Rectum 2001; 44:173–8.[CrossRef][Medline]
  29. Maetani S, Onodera H, Nishikawa T, et al. Significance of local recurrence of rectal cancer as a local or disseminated disease. Br J Surg 1998; 85:521–5.[CrossRef][Medline]
  30. Cheng C, Rodriquez-Bigas MA, Petrelli N. Is there a role for curative surgery for pelvic recurrence from rectal carcinoma in the presence of hydronephrosis? Am J Surg 2001; 182:274–7.[CrossRef][Medline]
  31. Hoffman JP, Riley L, Carp NZ, Litwin S. Isolated locally recurrent rectal cancer: a review of incidence, presentation, and management. Semin Oncol 1993; 20:506–19.[Medline]
  32. Hoffman JP, Lanciano R, Carp NZ, et al. Morbidity after intraperitoneal insertion of saline-filled tissue expanders for small bowel exclusion from radiotherapy treatment fields:a prospective four year experience with 34 patients. Am Surg 1994; 60:473–82.[Medline]
  33. Hoffman JP, Sigurdson ER, Eisenberg BL. Use of saline-filled tissue expanders to protect the small bowel from radiation. Oncology 1998; 12:51–4.
  34. Staib L, Schirrmeister H, Reske SN, Beger HG. Is 18 F-fluorodeoxyglucose positron emission tomography in recurrent colorectal cancer a contribution to surgical decision making? Am J Surg 2000; 180:1–5.[CrossRef][Medline]
  35. Henry LR, Sigurdson E, Ross E, Hoffman JP. Hydronephrosis does not preclude curative resections of pelvic recurrences after colorectal surgery. Ann Surg Oncol 2005 (in press).




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Henry, L. R.
Right arrow Articles by Hoffman, J. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Henry, L. R.
Right arrow Articles by Hoffman, J. P.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS