Follow-up in colorectal cancer patients: A cost-benefit analysis

Annals of Surgical Oncology, 3(4):349-357 Published by Lippincott-Raven Publishers © 1996 The Society of Surgical Oncology, Inc.

Follow-up A

in Colorectal Cancer Patients" Cost-Benefit Analysis

Riccardo A. Audisio, MD, Paolo Setti-Carraro, MD, Marco Segala, MD, Deborah Capko, MD, Bruno Andreoni, MD, and Giorgio Tiberio, MD

Background: No conclusive evidence exists concerning the effectiveness of follow-up programs after curative surgery for colorectal cancer, and presently cost-benefit analyses have not indicated that follow-up strategies increase survival or quality of life.

Methods: Five hundred five patients who survived curative surgery for stage I-III colorectal adenocarcinoma were closely followed for at least 4 years.

Results: One hundred forty-one (28%) patients had recurrence. Of these, 32 underwent one or more surgical procedures for cure, whereas 109 could only benefit from palliation. Eighteen were cured. The mean survival of all recurrent cases was 44.4 months. Of those operated on with curative intent, the mean survival was 69.3 months compared with 37.1 months in those operated on with palliative intent. Of those 18 patients who were cured by reoperative surgery, the average survival was 81.4 months. The overall follow-up cost was $1,914,900 (U.S.) for the 505 patients; $13,580 (U.S.) for each recurrence, $59,841 (U.S.) for each case treated for cure, and $136,779 (U.S.) for those effectively cured.

Conclusions: Careful postoperative monitoring is expensive yet effective when one con- siders that one-quarter of the detected recurrences were suitable for potentially curative second surgery; however, only 3.6% of the original group were effectively cured. Follow- up programs should be tailored according to the stage and site of the primary to reduce costs.

Key Words: Colorectal neoplasm--Follow-up--Reoperation--Costs.

Follow-up protocols are aimed at early detection of tumor recurrence in patients who have previously had potentially curat ive surgery as well as detecting metachronous tumors and distant metastat ic disease.

Compar isons of re t rospect ive series with differing follow-up protocols have been unable to provide conclusive recommendat ions . A prospect ive study would be needed to clarify the efficacy of any follow- up program, although this would be subject to ethi- cal constraints.

Reliability of these studies also depends on the

Received March 24, 1995; accepted April 21, 1995. From the Istituto di Chirurgia d'Urgenza, Universiffa di Milano

(P.S.C., M.S., G.T.), General Surgery 2, EIO-European Institute of Oncology, Milan, Italy (R.A.A., D.C., B.A.).

Address correspondence and reprint requests to Dr. P. Setti- Carraro, Istituto di Chirurgia d'Urgenza, Ospedale Maggiore Poli- clinico, via F. Sforza 33, 20122 Milan, Italy.

Presented at the 48th Annual Cancer Symposium of The Soci- ety of Surgical Oncology, Boston Massachusetts, March 23-26, 1995.

experience of the investigators (surgeon, endoscop- ist, pathologist, and radiologist), and this will substantially influence the interpretat ion of results.

Finally, there is increasing concern about the costs of follow-up protocols . Little data are available because cos t -benef i t analyses have been lacking.

The present report takes into considerat ion a large series of patients t reated consecut ively for cure at a single institution. Follow-up results are assessed by cos t -benef i t analysis.

PATIENTS A N D METHODS

During the 11-year period f rom 1980 to 1990, 505 consecut ive patients survived potentially curat ive resectional surgery for histologically p roven colorectal cancer at one institution. Three hundred fifty- eight underwent elective operat ions and 147 had emergency surgery. One hundred forty-six had rectal neoplasms and 359 had colonic neoplasms. Pa- tients were stratified by sex, age, site of malignancy, type of surgery, Dukes ' stage, and adjuvant treat-

349

350 R. A. A UDISIO ET AL.

T A B L E 1. Pat i en t s ' s trat i f icat ion at f i r s t surgical t r e a t m e n t ( N = 505)

Sex (male/female) 236/269 Age (yrs) Mean 65.2; range 29-95;

SD 13.0 Site of malignancy

Right colon 150 (29.7%) Splenic flexure 19 (3.8%) Left colon 190 (37.6%) Rectum 146 (28.9%)

Type of surgery Right colectomy 127 Left colectomy 125 Anterior resection 123 Segmental resection 61 Abdominoperineal res. 33 Hartmann resection 8 Transanal resection 8 Total colectomy 5 Others 15

Dukes' stage A 34 (6.8%) B 321 (63.5%) C 150 (29.7%)

Adjuvant treatments Chemotherapy 34 Radiotherapy 42 Chemotherapy & radiotherapy 8

ments as detailed in Table 1. Stage IV (Dukes' D after Turnbull) patients were excluded from the analysis. All patients entered a rigid follow-up protocol (Table 2) and were prospectively assessed with a minimum follow-up of 4 years; 28 patients (5.5%) were lost to follow-up (25 without and 3 with recurrent disease). At each outpatient appointment, detailed questioning and a carefully conducted physical examination were performed. Ultrasound (US) was used to evaluate the liver, abdomen, and pelvis. All patients were examined by the same clinician (P.S.C.) and the same ultrasonographer.

For the purposes of definition, the right colon included the proximal two-thirds of the transverse colon with the ascending colon and cecum; the splenic flexure included the left one-third of the transverse colon, the angle, and the first portion of the descend-

ing colon; the left colon incorporated the remainder of the descending colon, sigmoid and rectosigmoid junction, and the rectum included the extraperito- neal component treated either by a sphincter-saving procedure or by an abdominoperineal resection (APR).

Local recurrence (LR) was defined as evidence of recurrence at the anastomosis, in the abdominal wall, in the perineum, or in the drain site, and metastatic disease (M+) as the evidence of hepatic, peritoneal, or extraabdominal recurrence of cancer. Sur- gery for cure was regarded as any procedure judged to be potentially curative both intraoperatively and at pathological examination of the specimen. Pallia- tive treatment was regarded as any therapy aimed to improve the quality or quantity of life but not to cure disease. Palliation was either surgical (bypass, diversion, debulking, endoscopic treatment), nonsurgical (palliative chemotherapy and/or radiation therapy or supportive treatment) or both.

Carcinoembryonic antigen (CEA) was measured in all patients at 3-month intervals for the first year after surgery and intermittently thereafter (n.y. 0-5 ng/ml). Levels exceeding the upper range limit by 25% were considered abnormal and those patients with an increased CEA level completed the remainder of the protocol (Table 2). Second-look surgery or radioimmunoguided surgery based on abnormal CEA levels was not performed during this study period. All costs are given in approximate US dollars.

STATISTICS

The X 2 test was used to evaluate large samples and Fisher's exact test was employed for small samples.The Kaplan-Meier life table method was used for survival analysis and the statistical differences between groups was calculated using the Mantel- Haenszel test (1). Differences were regarded as significant for a probability value <0.05.

TABLE 2. Fol low-up s c h e m e

Postoperative mo

3 6 9 12 16 20 24 30 36 48 60

History & physical + + + + + + + + + + + CEA + + + + + + + + + + + US + + + + + + + + + + + Colonoscopy + + + + + Chest radiograph + + + + +

CEA, carcinoembryonic antigen; US, ultrasound.

Ann Surg Oncol, Vol. 3, No. 4, 1996

COLORECTAL CANCER FOLLOW-UP 351

TABLE 3. Recurrent disease according to site of primary

Site of primary LR M+ M+ & LR

Right colon (150 pts) 6 Liver 9 1 Lung 3 3 Diffuse 3 2 Others 6 1

Total 6 (4.0%) 21 (14%) 7 (4.7%) Splenic flexure (19 pts) 1 Liver 2 1

Lung 1 Diffuse 1 1 Others - - 1

Total 1 (5.3%) 4 (21.0%) 3 (15.8%) Left colon (190 pts) 8 Liver 10 3

Lung 4 1 Diffuse 5 4 Others 8 3

Total 8 (4.2%) 27 (14.2%) 11 (5.8%) Rectum (146 pts) 22 Liver - - 4

Lung 6 5 Diffuse 5 5 Others 2 4

Total 22 (15%) 13 (8.9%) 18 (12.3%) Total (505 pts) 37 (7.3%) 65 (12.9%) 39 (7.7%)

M+, number of patients per site of metastatic disease; M+ and LR, number of patients with metastasis and local recurrence; LR, number of patients with local recurrence only; diffuse, always including liver involvement.

RESULTS

Recurrence One hundred forty-one patients (27.9%) devel-

oped recurrent disease at a mean interval of 23 months after surgery (median 18 months; range 2- 108 months). Of these, 37 (26%) had LR only, 65 (46%) had M+, and 39 (28%) had both LR and M+. The mean intervals to each recurrence Were 15, 28, and 23 months, respectively (medians 12, 21, 19; ranges 2-77, 7-108, and 5-62).

The pattern of distribution of recurrent disease was significantly different between rectal and colonic carcinoma, with rectal tumors showing a higher proportion of LR compared with the colonic cases (Table 3). Of recurrences in the rectal tumor group, 75.5% were local whereas 85.9% of patients with recurrence in the colonic tumor group had distant metastatic disease (p < 0.0001).

Recurrent disease was detected because of the occurrence of new symptoms and/or signs found at physical examination alone in 32.7% of the cases (Table 4). Colonoscopy alone diagnosed 2.7% of recurrences and 6.2% of patients had combined symptoms and an abnormal colonoscopy examination; 10.6% had symptoms and an elevated CEA with 4.4% having symptoms and abnormal radiology. CEA alone detected 4.4% of the recurrences; however, it was elevated in 16.8% of those recurrent

cases with abnormal US examinations. A total of 33.6% of recurrences had abnormal CEA levels. Chest radiography was diagnostic in 8.8% of the cases.

The incidence of recurrence according to Dukes' stage is shown in Table 5. There was a significantly higher proportion of recurrences among stage C tumors when compared with both stages A and B, but there was no difference comparing the incidence of recurrence between stages A and B (A versus B = ns; A versus C = p < 0.005, B versus C = p < 0.0001).

Treatment of recurrent disease Of the 141 patients with recurrent disease, 53

(37.6%) were considered suitable for a potentially curative procedure and underwent a second operation. Thirty-two recurrences were resected, whereas 21 were inoperable and had surgical palliation only. The remaining 88 (62.4%) patients had nonsurgical palliation.

Curative surgery Thirty-two patients (22.7% of all recurrences) had

at least one surgical procedure judged to be curative both intraoperatively and at pathological examination. This selected group included 10 with LR only, 13 with M+, and 9 with LR and M+ (Tables 6-8).

Ann Surg Oncol, Vol. 3. No. 4, 1996


TABLE 4. Number of patients (%) with positive tests among 113 patients at the recurrence of disease

History & Chest physical radiograph US CEA Colonoscopy

History & physical 37 5 5 12 7 (32.7%) (4.4%) (4.4%) (10.6%) (6.2%)

Chest radiograph 5 10 - - 2 - - (4.4%) (8.8%) (1.8%)

US 5 - - 8 19 - - (4.4%) (7.1%) (16.8%)

CEA 12 2 19 5 - - (10.6%) (1.8%) (16.8%) (4.4%)

Colonoscopy 7 - - - - - - 3 (6.2%) (2.7%)

Total 66 17 32 38 10 (58.4%) (15%) (28.3%) (33.6%) (8.8%)

US, ultrasound; CEA, carcinoembryonic antigen.

Those who had recurrence and were cured at a second or further resection were detected by positive history and physical examination alone in six cases, on abnormal US in six, and raised CEA levels in four. One recurrence was found by chest radiograph, and one by colonoscopy.

Four patients had additional surgical resections: two for further local recurrence and two for recurrent metastatic disease.

Palliative treatments One hundred nine patients (77%) had palliative

treatment only. This group included 27 with LR alone (73% of all LR), 52 with M+ disease (80% of all M+), and 30 with combined LR and M+ (77% of all LR and M+).

Survival The mean overall survival for the entire group was

59.5 months (median 50; range 1-159 months) (Table 9). The mean survival of the 141 patients who had recurrence was 44.4 months (median 33; range 5-147 months); and that of the 32 patients operated on for cure was 69.3 months (median 63.5; range 15-147 months) (Fig. 1). Of those operated on with curative intent, 18 are long-term survivors and disease free at a mean of 81.4 months (median 93.0; range 15-147

months). Patients receiving palliative treatment had a mean survival of 37.1 months (median 29.0, range 5-135 months) with a mean survival of 39.8 months (median 29.0; range 12-128 months) for the 21 patients who had surgical palliation, and 36.5 months (median 29.0; range 5-135 months) for the 88 who had nonsurgical palliation.

There was a significant difference in survival between the 141 patients who had recurrence and the 364 who did not (Fig. 2; p < 0.0001), but not among the subsets of the recurrent cases (Fig. 3).

Metachronous tumors Twenty patients (4%) developed a metachronous

colorectal tumor during annual colonoscopy during a minimum follow-up of 3 years.

Costs The average cost of a follow-up visit and physical

examination in a private institution in Italy is $90. CEA, US, colonoscopy, and chest radiography cost $30, $125, $440, and $100, respectively.

The cost of this follow-up protocol is $1,520 for the first year, $1,275 for the second year, $1,035 for the third year, and $785 yearly thereafter. The 5- year cost per patient averages $5,400. The overall cost for this study population, computed from the

T A B L E 5. Patients' distribution and time to recurrence according to pathological staging

No. with recurrence Mean time to Dukes' stage Entered f%) recurrence (mos) Range (mos)

A 34 4 (11.8) 41.2 10-89 B 321 73 (22.7) 24.8 2-105 C 150 64 (42.5) 20.3 2-108 Total 505 141 (27.9) 23.0 2-108

x 2 24.895; df = 2; p < 0.0001.



T A B L E 6. Outcome of 37 patients with local r e c u r r e n c e

Palliation 27 pts Surgery for cure 10 pts N E D 5 pts

REC 5 pts (palliation 3 pts; surgery for cure 2 pts)

NED, no evidence of disease; REC, recurrent disease.

survival curve, was $1,914,900. Therefore, each of the 32 cases requiring a second surgical procedure with the intention of cure costs $59,841, and the 18 patients actually cured cost $106,383 each.

DISCUSSION

The rationale for a follow-up protocol is based on the evidence that surgery is the only curative treatment for primary and recurrent disease (2-7). Early detection of metachronous (2,8,9) or recurrent disease may result in an increased rate of cure and in a less destroying surgical procedure; however, the effectiveness and efficacy of postoperative surveillance in reaching this target has often been ques- tioned (10-16). Follow-up programs include a wide spectrum of different variables, and consequently no conclusive data are available.

Polk et al. (8) reported that the presence or ab- sence of symptoms was not helpful in detecting recurrent tumor in a large cohort of patients, whereas Cochrane et al. (11) had only one asymptomatic patient in a series of 71 cases with documented recurrence. This contrasts with our experience and that of others (12,17-20). Our study found that 32.7% of cases with recurrent disease were detected at follow- up because of symptoms and/or physical signs alone.

Almost two-thirds of patients with recurrence had symptoms and/or clinical signs either alone or in association with other tests. Careful history taking and physical examination were important in identifying one-third of patients ultimately cured of recurrent disease, and the association of history, physical examination, US, and CEA identified 88% of these cases. The chance of surgical cure, however, is substantially

T A B L E 7. Outcome o f 65 patients with M+


REC 7 pts (palliation 5 pts; surgery for cure 2 pts)

NED, no evidence of d i s e a s e ; R E C , recurrent disease.

T A B L E 8. Outcome o f 39 patients with LR & M+


REC 6 pts

NED, no evidence of disease; REC, recurrent disease.

greater in the asymptomatic group (66%) when compared with symptomatic patients (33%) and this is in agreement with other reports (15,21,22).

CEA alone detected 4.4% of cases but elevations of CEA were frequently associated with abnormal clinical signs and symptoms (10.6%) or abnormal ultrasonographic findings (16.8%). Only 1.8% of patients with lung metastases had abnormal CEA levels. Because CEA alone has not been helpful in our experience, we do not share the enthusiasm of others who believe that follow-up that is CEA-based will lead to improved cure rates (21,23,24). Because an elevated CEA frequently occurred with symptoms and considering the low cost and the ease of perform- ing the assay, we believe that it should be included in the follow-up protocol.

The number of affected subjects detected by other diagnostic procedures is extremely limited (6), although US first detected asymptomatic recurrence in 28.3% of our patients. Abdominopelvic US is a noninvasive and harmless procedure, at least as accurate as computed tomography (CT) scans in detecting liver metastases, nodal retroperitoneal involvement, and ureteric obstruction; however, CT scans are more accurate in detecting abdominal masses after APR (25). Both techniques fail to iden- tify peritoneal seedings. IfCT was to be employed in the present study, this would have led to a threefold increase in the cost of abdominopelvic imaging. Chest radiography alone picked up a noticeable pro-

T A B L E 9. Survival according to groups

No. of Group patients Mean Median Range

Overall 505 59.5 50 1-159 No recurrence ~ 364 65.5 62.0 1-159 Recurrence a 141 44.4 33.0 5-147

Second surgery for cure b 32 69.3 63.5 15-147

Surgical palliation C 21 39.8 29.0 12-128 Nonsurgical

palliation c 88 36.5 29.5 5-135

a p < 0.0001. b Significantly different f rom palliated; p < 0.0001. c No significant difference be tween surgical and nonsurgical

palliation; p = ns.



0,~--

0.6

0 .4-

0.2 -

0 0

Probability

+ , ++ +

-- • ' i ! - -

i 6 12 24 36 48 60 72 84 96 108 120 132 144 56

Months from surgery

FIG. 1. Cumulat ive survival by type of treatment of recurrent disease. - V - , curat ive surgery (n = 32); ~ , palliation (n = 109). p -< .0001.

Probability 1 , ,

0 . 8 ~

0,6

0,4-

0,2 . . . . . . . . . . . . . . . . . . . . . . . . . ~ " ~ ~ i . . . . . ,---

o / / i b i O 6 12 24 36 48 60 72 84 96

Months from surgery

I 1 I 108 120 132 144 156

FIG. 2. Cumulat ive survival. - ~ - , overall (n = 505); - T - , no recurrence (n = 364): ~, recurrence (n = 141). Recurrence vs no recurrence, p < .0001.

0.4

0.2

0

Probability

0 6 12 24 36

. . . . . . . . . . . . . . . . i-

48 60 72 84 96 108 120 132 144

Months from surgery

FIG. 3. Cumulat ive survival by type of recurrence. - V - , M+ (n = 65); ~ LR (n = 3 7 ) ; - U - LR & M+ (n = 39). p = .53.

156



portion of recurrences that otherwise might have been missed, because most of them were not associated with other signs or symptoms.

It has been suggested that annual colonoscopy after resection plays an important role in detecting metachronous cancers (26-28), whereas it might rarely allow early detection of asymptomatic intra- luminal bowel recurrence (11,29). The place of colonoscopy is clear before the first operation in detecting synchronous lesions, and in the surveillance of patients at increased risk of metachronous adeno- mas and cancers, whereas its role in the diagnosis and cure of recurrent disease appears to be limited (28,30,31). We found only one recurrence with colonoscopy alone, because most were within reach of either digital rectal examination or rigid sigmoid- oscopy.

We believe that seniority of the investigators and continuity in follow-up practice reinforce patient confidence and may result in early reporting of new symptoms, improving the sensitivity and specificity of the office visit. This does not mean, however, that all symptomatic recurrences--in particular, suture- line recurrence--are incurable when discovered or that much asymptomatic recurrence is necessarily curable at presentation. Obvious symptoms can still lead to potentially curative surgery even outside of a follow-up program, particularly when patients are properly educated and encouraged to report their occurrence.

The approach to cancer surveillance is markedly different throughout the world (32). In Europe most physicians rely strongly on repeated clinical examinations with minimal emphasis placed on tests, whereas American centers supplement follow-up with repeated expensive biochemistry, radiology, and invasive investigations. It is believed that the next decade will witness an increasing concern in the cost-benefit analysis of medical practice (33), driving our efforts toward a cost-effective monitoring policy.

In this series, the prevalence of recurrence was low in Dukes' A tumors (11.8%), but this proportion increased to 22.7 and 42.5%, respectively, in Dukes' B and C patients. These differences were statisti- cally significant.

Our analysis confirmed that recurrence patterns differ between colonic and rectal tumors. More than 80% of recurrent colon cancer patients had metastatic spread, whereas a similar proportion of recurrent rectal tumors relapsed locally. Our anastomotic recurrence rate of 15% after anterior resection indi-

cared a three- to fourfold probability of LR after surgery for rectal cancer as compared with colon cancer, in accordance with previous reports (34-36). Ten of these patients underwent curative surgery and seven are alive and disease free after 6 years, stressing the importance of early detection and aggressive treatment, as suggested by other authors (37-39).

The highest proportion of local and/or metastatic recurrence in the colonic cancer cases occurred if the primary was at the splenic flexure. Metastatic liver involvement was frequent, occurring in 43.8% of these patients, and was four times more common in colonic cancer cases than in cases of rectal cancer.

Resection of colorectal metastatic disease of the liver or lung is currently accepted as a standard therapeutic approach, provided these are the only sites of relapse (40-43). In the present series, 65 patients (46%) with recurrence had metastatic recurrence only, and 13 had curative surgery, resulting in 9 long-term survivors. Patients presenting with synchronous metastatic and local relapse may be considered for potentially curative surgery, and this was performed in a further nine patients, with three long-term survivors (LR 27%, M+ 20%, and LR and M+ 23%).

Our results are in keeping with data from the litera- ture (14,16,44-53) suggesting that of the population entering a follow-up program, 5-30% of those with recurrent disease are candidates for curative resection and 3-9% will survive.

Tumor recurrence is a negative prognostic factor; however, a surgical approach aimed at cure may be attempted even in the presence of combined local and metastatic disease. In this series, 32 of 141 (22.7%) of the recurrent cases had potentially curative surgery resulting in 18 long-term survivors, including 4 patients who had repeated resections for second local or metastatic recurrence and who are alive and disease free >6 years later. Conversely, there is clinical evidence that palliative treatments, both surgical and nonsurgical, do not significantly improve survival. These findings should be considered when further treatment in cases of recurrent unresectable colorectal cancer is proposed, because the quality of life is not substantially improved.

In conclusion, 3.6% of patients with recurrent disease were detected during postoperative follow-up and subjected to resection, resulting in cure. The number of subjects who benefit from such a follow- up program is small. The results would justify the high cost of the protocol, which averaged $106,383



per patient cure. Similar results at a lower cost may be achieved by tailoring the program in accordance with the Dukes' stage and the site of the primary. Patients should be encouraged to recognize and report early symptoms. These, in combination with physical examination, CEA, and US, are most likely to detect early recurrences, possibly offering the greatest possibility of curative re-resection.

A c k n o w l e d g m e n t : We t h a n k Dr. A n d r e w Zba r for re- v iewing the manusc r ip t .

References

I. Mantel N. Chi square tests with one degree of freedom exten- sions of the Mantel-Haenszel procedures. J Am Stat Assoc 1963 ;58:620-5.

2. Devesa JM, Morales V, Enriquez JM, et al. Colorectal cancer. The bases for a comprehensive follow up. Dis Colon Rectum 1988;31:636-52.

3. Willett CG, Tepper JE, Cohen AM, et al. Failure patterns following curative resection of colonic carcinoma. Ann Sarg 1984;200:685-90.

4. Phillips RKS, Hittinger R, Blesovsky L, et al. Local recurrence following "curat ive" surgery for large bowel cancer. The overall picture. Br J Surg 1984:71:12-6.

5. Beal JM, Cornell GN. A study of the problem of recurrence of carcinoma at the anastomotic site following resection of the colon carcinoma. Ann Surg 1956;143:1-7.

6. Wright HK, Thomas WH, Cleveland JC. The low recurrence rate of colonic carcinoma in ileocolic anastomoses. Surg Gyn- ecol Obstet 1969;129:960-2.

7. Vernava AM, Longo WE, Virgo KS, et al. Current follow up strategies after resection of colon cancer. Dis Colon Rec- tum 1994;37:573-83.

8. Polk HC, Spratt JS Jr, Butcher HR Jr. Frequency of multiple primary malignant neoplasms associated with colorectal carcinoma. Am J Surg 1965;109:71-5.

9. Weir JA. Colorectal cancer: metachronous and other associated neoplasms. Dis Colon Rectum 1975;18:4-5.

10. BeartRW, O'ConnellMJ. Postoperative follow upofpatients with carcinoma of the colon. Mayo Clin Proc 1983 ;58:361-3.

II. Cochrane JPS, Williams JT, Faber RG, Slack WW. Value of outpatient follow up after curative surgery for carcinoma of the large bowel. Br Med J 1980;280:593-5.

12. Deveney KM, Way LW. Follow up of patients with colorectal cancer. Am J Surg 1984;148:717-22.

13. Isbister WH. The follow up of patients following surgery for colorectal cancer: a personal view. Ann Acad Med 1988; 17:66-71.

14. Sugarbaker PH, Gianola FJ, Dwyer A, Neuman NR. A sim- plified plan for follow up of patients with colon and rectal cancer supported by prospective studies of laboratory and radiologic test results. Surgery 1987;102:79-87.

15. TSrnquist A, Ekelund G, Leandoer L. The value of intensive follow up after curative resection for colorectal carcinoma. Br J Surg 1982;69:725-8.

16. B6hm B, Schwenk W, Hucke HP. Does methodic long term follow up affect survival after curative resection of colorectal carcinoma? Dis Colon Rectum 1993 ;36:280-6.

17. Welch JP, Donaldson GA. Detection and treatment of recurrent cancer of the colon and rectum. Am J SIII\e 1978;135: 505-11.

18. Beart RW Jr. Postoperative follow up of colorectal cancer. In: Beahrs OH, Higgins GA, Weinstein JJ, eds. Coloreetal tumors. Philadelphia: JB Lippincott, 1986:299-301.

19. Kelly CJ, Daly JM. Colorectal cancer--principles of postoperative follow up. Cancer 1992;70:1397-413.

20. Fantini GA, DeCosse JJ. Surveillance strategies after resection of carcinoma of the colon and rectum. Surgery 1990: 171:267-73.

21. Minton JP, Hoen JL, Gerber DM, et al. Results of a 400 patient carcinoembryonic antigen second look colorectal cancer study. Cancer 1985 ;55:1284-90.

22. Bahler H, Seefeld U, Deyhle P, et al. Endoscopic follow up after colorectal surgery. Cancer 1984;54:791-4.

23. Martin EW, Cooperman M, Carey LC, Minton JP. Sixty second-look laparotomies indicated by rise in serial carci- noembryonal antigen. J Surg Res 1980;28:389-94.

24. M~tkel~i J, Laitinen S, Kairaluoma MI. Early results of follow up after radical resection for colorectal cancer. Preliminary results of a prospective randomised trial. Surg Oncol 1992;1:157-61.

25. Mendez R J, Rodriguez R, Kovachevich T, et al. CT in local recurrence of rectal carcinoma. J Comput Assist Tomogr 1993;17:741-4.

26. Weber CA, Deveney KE. Routine colonoscopy in the management of colorectal carcinoma. Am J Surg 1986; 152:87-90.

27. Kronborg O, Hage E, Deichgraeber E. The remaining colon after radical surgery for colorectal cancer: the first three years of a prospective study. Dis Colon Rectum 1983;26:172-6.

28. Juhl G, Larson GM, Mullins R, et al. Six year results of annual colonoscopy after resection of colorectal cancer. World J Surg 1990;14:255-61.

29. Patchett SE, Mulcahy HE, O'Donoghue DP. Colonoscopic surveillance after curative resection for colorectal cancer. Br J Surg 1993;80:1330-2,

30. Mecklin JP, J~irvinen H. Treatment and follow up strategies in hereditary nonpolyposis colorectal carcinoma. Dis Colon Rectum 1993;36:927-9.

31. Cali RL, Pitsch RM, Thorson AG, et al. Cumulative incidence of metachronous colorectal cancer. Dis Colon Rectum 1993; 36:388-93.

32. Moffat FL. Letter to the Editor. Am J Gastroenterol 1992; 3:404.

33. Moffat FL. Comprehensive surveillance after curative surgery for adenocarcinoma of the colon. Am J Gastroenterol 1991 ;86:939-40.

34. Welch JP, Donaldson GA. Detection and treatment of recurrent cancer of the colon and rectum. Arn J Surg 1978; 135:505.

35. Winchester DP, McBride CM. Cancer of the colon; a compar- ison of survival factors. South Med J 1974;67:1029.

36. Gunderson J, Sosin H. Areas of failure found at reoperation (second or asymptomatic look) following curative surgery for adenocarcinoma of the rectum. Cancer 1974 ;34:1278-92.

37. Wanebo HJ, Gaker DL, Whitehill R, et al. Pelvic recurrence of rectal cancer: options for curative resection. Ann Surg 1987 ;205:482-95.

38. Pealman NW, Donohue RE, Stiegmann GV, et al. Pelvic and sacropelvic exenteration for locally advanced or recurrent anorectal cancer. Arch Surg 1987;122:537-41.

39. Benotti P, Steele G Jr. Patterns of recurrent colorectal cancer and recovery surgery. Cancer 1992;70:1409-13.

40. Hemming AW, Langer B. Repeat resection of recurrent hepatic colorectal metastases. Br J Surg 1994;81:1553-4.

41. 13ozzetti F, Bignami P, Morabito A, Doci R, Gennari L. Patterns of failure following surgical resection of colorectal liver metastases. Ann Surg 1987;205:264-70.

42. McCormack PM, Burt ME, Bains MS, Martini N, Rusch VW, Ginsberg R. Lung resection for colorectal metastases; 10 year results. Arch Sarg 1992;127:1403-6.

43. Wilking N, Petrelli NJ, Herrera L, Regal AM, Mittelman A. Surgical resection of pulmonary metastases from colorectal adenocarcinoma. Dis Colon Rectum 1985;28:562-4.

44. Metzger U, Bronz K, Bilhler H, et al. Die prospektive


C O L O R E C T A L C A N C E R F O L L O W - U P 357

nachsorgestudie radikal operierter kolorektaler karzinome. Schweiz Med Wochenschr 1985;115:1001-4.

45. Mentges B. Kontrversen in der nachsorge des kolonkarzi- noms. Zentralbl Chit 1988;113:1087-91.

46. Schiessel R, Wunderlich M, Herbst F. Local recurrence of colorectal cancer: effect of early detection and aggressive surgery. Br J Surg 1986;73:342-4.

47. Ovaska JT, Jg.rvinen H, Mecklin JP. The value of a follow up programme after radical surgery for colorectal carcinoma. Scand J Gastroenterol 1989;24:416-22.

48. Ovaska JT, J~irvinen H, Kujari H, Perttilfi I, Mecklin JP. Follow up of patients operated on for colorectal carcinoma. Am J Surg 1990;159:593-6.

49. Schiessel R, Wenzl E. Was bringt die nachsorge beim kolore- ktalen karzinom? Wien Med Wochenschr 1988;11:263-6.

50. Schildberg FW. Probleme des lokalrezidivs nach colorecta- lem carcinom. Langenbecks Arch Chir 1987;372:505-11.

51. Mentges B, Mentges W, Grfissner R, BrOckner R. Art und prognose des lokoregiongtren rezidivs beim rektumkarzinom. Dtsch Med Wochenschr 1988;113:806-10.

52. Mentges B, Briickner R. Prognostische kriterien und risiko- gruppen beim kolorektal karzinom. Wien Med Wochenschr 1988;13:317-22.

53. Camufias J, Enriquez JM, Devesa JM, et al. Value of follow up in the management of recurrent colorectal cancer. Eur J Surg Oncol 1991;17:530-5.


Documents

Follow-up in colorectal cancer patients: A cost-benefit analysis