Better Survival in Patients with Esophageal Cancer After Surgical Treatment in University Hospitals: A Plea for Performance by Surgical Oncologists

Gastrointestinal

Better Survival in Patients with Esophageal Cancer AfterSurgical Treatment in University Hospitals: A Plea for

Performance by Surgical Oncologists

Christian Verhoef, MD,1,2 Rens van de Weyer, MD,2,3 Michael Schaapveld, PhD,2

Esther Bastiaannet, MSc,2 and John Th. M. Plukker, MD, PhD1

1Department of Surgical Oncology, Groningen University Medical Center, University of Groningen, P.O. Box 30 001, 9700 RBGroningen, The Netherlands

2Comprehensive Cancer Center North-Netherlands, P.O. Box 330, 9700 AH, Groningen, The Netherlands3Department of Internal Medicine, Maxima Medical Center, P.O. Box 90 052, 5600 PW, Eindhoven, The Netherlands

Background: In primary esophageal cancer, studies have frequently focused on surgicalpatients in an effort to link outcome to hospital- or surgeon-related experience, with operativemortality used as the main outcome measure. Many studies have found an inverse relationshipbetween operative mortality and hospital volume and surgical expertise. This study aims toassess the inuence of surgeon-related expertise and hospital volume on the relative survival ofoperated esophageal cancer patients.Methods: From January 1994 to January 2002, a total of 1149 consecutive patients with

primary esophageal cancer were diagnosed in the region of the Comprehensive Cancer CenterNorth-Netherlands. As a proxy for surgeon-related expertise, hospitals in this region werecategorized into three types: university, teaching nonuniversity, and nonteaching hospitals.The inuence of hospital type on the relative survival of operated patients was studied by amultivariate Poisson regression model.Results: Of the 1149 patients, 18.5% underwent surgery. There was no evidence of selective

referral for surgery between the three hospital types with regard to age, tumor stage, andlocation. For operated patients, the 5-year relative survival was 49.2% for the universityhospital versus 32.6% and 27.3% for teaching nonuniversity and nonteaching hospitals,respectively (P = .0039). When adjusted for age, tumor stage, hospital volume and referralfrequency, the relative excess risk of death for the university hospital was considerably lowerat .57 (95% condence interval, .291.12) compared with nonteaching hospitals and .43 (95%confidence interval, .24.76) compared with teaching nonuniversity hospitals (P = .0126).Conclusions: In our region, patients with esophageal cancer who underwent esophagectomy

in the university hospital had a markedly better relative survival compared with those whounderwent surgery at teaching nonuniversity and nonteaching hospitals, emphasizing the needfor referral of esophageal surgery to centers with a greater experience.Key Words: Survival benetReferralExpertiseEsophageal carcinoma.

With an incidence of 6.3 per 100,000 (Europeanstandardized rate) in the period 19941998, esopha-

geal cancer ranked 13th among malignancies in menand 12th in women in the Netherlands. Incidenceincreased in both male and female patients, with anestimated annual percentage change of 3.1% and2.0%, respectively. The mortality for men and womenwas, respectively, 9.4 per 100,000 and 3.1 per 100,000in 2004, indicating that the prognosis of patients withesophageal cancer remained poor in this period

Received April 12, 2006; accepted November 3, 2006; publishedonline: February 9, 2007.Address correspondence and reprint requests to: John Th. M.

Plukker, MD, PhD; E-mail: [email protected]

Published by Springer Science+Business Media, Inc. 2007 The Society ofSurgical Oncology, Inc.

Annals of Surgical Oncology 14(5):1678)1687DOI: 10.1245/s10434-006-9333-0

1678

(http://www.kankerregistratie.nl). The only curativeoption for esophageal cancer is surgery, which im-plies that improving the outcome of surgery is thebest means of reducing mortality.1

Esophageal cancer is one of the most challengingpathological conditions confronting the surgeon. Ittherefore seems reasonable to assume that concen-tration of esophageal surgery could improve out-come. Several studies have shown that variouscharacteristics, including surgeon subspecialty certi-cation, hospital setting, and the number of proce-dures performed, are associated with practicevariation, complication rates, and even outcome.24

Evidence of improved outcome associated with spe-cialist care exists for breast cancer,2,4 ovarian can-cer,3,5 and malignant teratoma.6 However, there is noevidence of comparable quality for esophageal can-cers. Several studies on esophageal cancer have fo-cused on surgical patients in an effort to link outcometo experience, either on the part of the institution orthe surgeon, by using postoperative mortality as themain outcome measure.79 A study on esophagecto-mies performed in England during the late 1980sfound no independent association between operativemortality and hospital surgical volume,10 whereastwo North American studies did demonstrate a lowermortality rate for esophagectomy when high-volumeand low-volume hospitals were compared.11,12 Vari-ous smaller studies also found lower operative mor-tality rates among high-volume surgeons.13,14

Few studies, if any, have attempted to relateesophageal cancer patient survival to surgeon exper-tise or hospital volume. To remedy this, in this study,we assessed the eect of surgeon-related expertise andhospital volume on the relative survival of operatedesophageal cancer patients. We compared data fromuniversity, teaching nonuniversity, and nonteachinghospitals.

PATIENTS AND METHODS

Patients

All patients diagnosed with a primary invasiveesophageal cancer in the region of the ComprehensiveCancer Centre North-Netherlands (CCCN) betweenJanuary 1994 and January 2002 were eligible for en-try onto the study. Patients with a history of cancerother than nonmelanoma skin cancer were excluded.The patients were selected through the population-based Regional Cancer Registry of the CCCN, whichcovers the northern part of the Netherlands, a mainly

rural area with a population of approximately 2.1million. The area is served by 17 community hospi-tals, 3 of which are teaching hospitals and 1 of whichis a university hospital; the hospitals include fourradiotherapy departments and seven pathology lab-oratories.

Data Collection by the Regional Cancer Registry

PALGA, a Dutch nationwide network and reg-istry of histopathology and cytopathology, regularlysubmits reports of all diagnosed malignancies to thecancer registry. The national hospital discharge databank, which receives discharge diagnoses of admit-ted patients from all hospitals, completes caseascertainment. The cancer registry has no access todeath certicates. After notication, trained registrypersonnel collect data on diagnosis and stagingfrom the medical records, including pathology andradiology reports, in the hospitals. The cancer reg-istry collected all data regarding the diagnosis andstaging, but collected no data on specic surgicaltreatment in this patient population before 2002.The data collection occurs at least 4 months afterdiagnosis to comprehensively document all aspectsof preoperative staging. All patients are stagedaccording to the tumor, node, metastasis system(TNM) system for esophageal cancer in use duringthat period.15,16

In the Netherlands, the population registries of themunicipality contain information on the vital statusof their inhabitants. Vital status was established ei-ther through information derived from the patientsmedical records or through linkage of cancer registrydata with information from the population registriesof the municipality within the registry areas orthrough linkage with the national death registry ofthe Central Bureau of Genealogy. The regional can-cer registry of the CCCN checked vital status by ac-tive record linkage with municipal populationregistries in 20022003 and 2005 and with the na-tional death registry of the Central Bureau of Gene-alogy in 2004.

Guidelines for Staging and Treatment

By establishing multidisciplinary teams and cancernetworks, the CCCN strives to improve the quality ofcancer care. Within the CCCN area tumor workinggroups, comprising delegated specialists representingall regional hospitals, that have been developing andrevising guidelines on diagnosis and treatment. Theregional guidelines for esophageal cancer were based

BETTER SURVIVAL IN ESOPHAGEAL CANCER 1679

Ann. Surg. Oncol. Vol. 14, No. 5, 2007

on the international TNM classication according tothe International Union Against Cancer in use at thattime.

Statistics and Denitions of Variables

The v2 test was used to compare the distributionover the patient population for categorical variables.For continuous variables, analysis of variance wasused. Relative survival analysis was performed toestimate the effect of university, teaching nonuni-versity, and nonteaching hospitals on the prognosisof operated patients with esophageal cancer. Thea priori hypothesis was that patient volume andhospital expertise would increase from nonteachingto university teaching hospitals. Survival time wascalculated from the date of diagnosis and ended atthe date of death, including perioperative death, orthe date of most recent linkage with the municipalpopulation registries and/or national death registry.The overall survival probability was estimated by theKaplan-Meier method.The expected survival probability was calculated

by using age-, sex-, and period-matched mortalityrates that were based on life expectancy tables in theNetherlands (http://www.statline.cbs.nl/StatWeb/)and was estimated by the Ederer 2 method.17 Thecumulative relative survival (the ratio of the overallsurvival to the expected survival) was estimated byStata (version 8.0) software and the strs function. Therelative survival, which estimates the net esophagealcancer survival in the hypothetical situation thatesophageal cancer is the only possible cause of death,has been shown to be a good estimator of disease-specic survival in the absence of information on thecause of death or in case information on the cause ofdeath is inaccurate.The excess mortality rate was calculated by sub-

tracting the expected number of deaths, estimatedfrom the expected survival probability, from the ob-served number of deaths in a subgroup or stratumand dividing the resulting excess number of deaths bythe number of accumulated person-years, takingcensoring into account. The relative excess risks(RER) of death were estimated as the ratio of excessmortality rates. RERs were estimated by a multi-variate generalized linear model with a Poisson errorstructure, which was based on collapsed relativesurvival data, by using exact survival times.18 By useof this model, the effect of the type of hospital (uni-versity, teaching nonuniversity, and nonteaching) wasstudied, adjusting for the effect of various covariableson the excess mortality experienced by our cohort.

Variables included in the nal model were age(70), stage based on collapsedTNM data (stage 1,2a, 2b, 3/4, unknown), hospitalvolume (33.3%],low referral [33.3%]), and time since diagnosis(1-year intervals). The pathological stage was usedwhenever possible; in the absence of informationabout the pathological stage, the clinical stage wasused.

RESULTS

In the period 19942002, a total of 1149 patientswere diagnosed with esophageal cancer, comprising796 men and 353 (69.3%) women (30.7%). Themedian age was 68 years (range, 17103 years). Pa-tient characteristics are described in Table 1. A largeproportion of the patients was diagnosed at an ad-vanced stage of disease; 45.5% were stage III orhigher. A further 28.8% were insufciently staged. Ofthe 1149 patients, 85 patients (7.4%) were initiallydiagnosed in the university hospital, 428 patients(37.2%) in teaching nonuniversity hospitals. Theremaining 636 patients (55.4%) were diagnosed in thenonteaching hospitals. Patients who were referredfrom the hospital of initial diagnosis for treatmentmay have subsequently undergone further diagnostictesting, and previously performed tests determined tobe insufcient were repeated. The 5-year relativesurvival rate for men with esophageal cancer was12.8% vs. 9.8% for women (P = .496). The 5-yearrelative survival markedly decreased as the stage ad-vanced. The 5-year relative survival was 71.5% instage I, 26.5% in stage IIA, 13.3% in stage IIB, 9.2%in stage III, 1.3% in stage IV, and 6.7% for patientswith unknown stage (P < .0001).As Table 1 shows, only 213 patients (18.5%)

underwent surgery. Older age (P< .001), advanced orunknown stage (P < .001), and proximal tumorlocation (P< .001) resulted in a lower probability oftumor resection. Squamous cell carcinoma was alsoassociated with less surgery, but it was highly corre-lated with the tumor location. In all, 21.4% of the pa-tients diagnosed in nonteaching hospitals underwentsurgery, compared with 15.9% and 10.6% for teachingnonuniversity and university hospitals, respectively(P = .011). Adjusted for age, stage, and tumor loca-tion, the odds of operation was 1.89 (95% condenceinterval [95% CI], 1.262.82) for patients diagnosedin a nonteaching hospital compared with patientsdiagnosed in a teaching nonuniversity hospital.

C. VERHOEF ET AL.1680


Of all operated patients 45.1% were referred forsurgery after diagnosis in the hospital of presentation(Table 2). The nonteaching hospitals referred 57.4%of patients diagnosed in their hospitals for an oper-ation elsewhere. Of the 14 nonteaching hospitals in

which esophageal cancer patients were diagnosed, 12referred nearly all (75%100%) patients, and tworarely (.0%6.9%) referred patients. The teachingnonuniversity hospitals referred 26.5% of the patientsdiagnosed in their hospitals to a larger institution for

TABLE 1. Characteristics of operated and nonoperated patients with esophageal cancer diagnosed 19942002

Characteristic

Total Operated Nonoperated

P valueN % n % n %

Sex .465Male 796 69.1 152 71.4 644 68.6Female 353 30.9 61 28.6 292 31.2Histology

therapy, with one hospital referring 63.2% of theirpatients.Table 3 compares characteristics of the operated

patients in the three hospital types. Of the 213 oper-ated patients, 95 underwent surgery in the universityhospital; 86 of these patients were referrals. The threeteaching nonuniversity hospitals provided surgery to60 patients, including 10 referrals: one hospital per-formed surgery on 7 patients, with the two otherhospitals operating on more than 20. The remaining58 patients underwent surgery at one of the 14 non-teaching hospital after all were initially diagnosed inthe same hospital; two low-referral hospitals operatedon more than 10 patients, eight hospitals operated on5 or fewer patients, and the remaining four high-referral hospitals did not perform any esophagealcancer surgery. There were no statistically signicantdifferences in the distribution of age (P= .230), stage(P= .299), or tumor location (P= .130) between thehospital types, again showing little evidence ofselective referral. However, of all operated patientsper hospital, a slightly larger proportion of stage III/IV tumors were operated on in the nonteaching(46.6%) and university hospitals (41.1%) comparedwith the teaching nonuniversity hospitals (26.7%),and the portion of stage IIA tumors was somewhathigher in nonteaching (31.0%) and teaching nonuni-versity hospitals (38.3%) compared with the univer-sity hospital (25.3%).

The cumulative relative survival for university,teaching nonuniversity, and nonteaching hospitals isshown in Fig. 1. Surprisingly, relative survival wasmarkedly better in the university hospital comparedwith teaching nonuniversity and nonteaching hospi-tals. The 5-year relative survival was 49.2% for theuniversity hospital versus 32.6% and 27.3% forteaching nonuniversity and nonteaching hospitals,respectively (P = .0039, Table 4). In univariateanalysis, the RER for university and teaching non-university was, respectively, .48 (95% CI, .30.74) and

TABLE 3. Characteristics for operated esophageal cancer patients diagnosed 19942002, according to hospital of surgery

Characteristic

TotalTeaching,

nonuniversity University Nonteaching

P valueN % n % n % n %

StageStage 1 32 15.5 12 20.0 14 14.7 6 10.3 .299a

Stage 2A 65 26.8 18 38.3 23 25.3 24 31.0Stage 2B 26 12.2 8 13.3 12 12.6 6 10.3Stage 3 + 4 82 38.5 16 26.7 39 41.1 27 46.6Unknown 8 7.0 1 1.7 6 6.3 1 1.7Age at diagnosis (y)

.88 (95% CI, .551.42) compared with nonteachinghospitals. The proportion of operated patients whodied within 3 months after diagnosis differed betweenuniversity, teaching nonuniversity and nonteachinghospitals (4.2%, 13.3% and 19.0% respectively, P =.013). Excluding patients who died within the rst 3months, the RER for university and teaching non-university was, respectively, .59 (95% CI, .351.00)and .97 (95% CI, .561.69) compared with non-teaching hospitals.In a multivariate analysis to adjust for the prog-

nostic eect of patient age, tumor stage, tumorlocation, hospital volume, frequency of referral, andtime since diagnosis, we found that stage, age, hos-pital type, and time since diagnosis were indepen-dently associated with the RER (Table 4). The RERincreased with more advanced stage. Patients aged

with survival, but it is unclear whether this reects theoperated patients or the population as a whole.10

Our study is one of the rst studies to research theeect of a marker for hospital/surgeon experience onpatient survival. Most studies to date have evaluated30-day postoperative or in-hospital mortality, andseveral demonstrated an inverse relationship betweenvolume or surgical experience and operative mortal-ity after esophageal resection.8,11,19 A Dutch studyfound that hospitals performing 1 to 10 operationsper year for esophageal cancer and cancer of thegastroesophageal junction had an operative mortalityof 12.2%, compared with 4.9% for hospitals per-forming >50 procedures per year.19 An Americanstudy had a similar result, with a mortality of 3.0%among high-volume hospitals and 12.2% among low-volume hospitals for both distal and proximalesophageal cancer,12 but this study used a cutoffpoint of ve procedures per year to differentiate be-tween high- and low-volume hospitals. What thresh-old distinguishes high-volume from low-volumehospitals remains matter of discussion. The results ofthese studies do suggest that centralization ofesophageal surgery, so that only a few hospitals perregion operate on esophageal cancer patients, mayimprove survival. The results of our study supportthe recommendation for referral of esophageal cancerpatients to a center where there is a specic focus onesophageal cancer treatment. Combined with the factthat the surgical literature is increasingly advocatingthe need for centralization, we think that furtherresearch into the advantages of centralization ofesophageal cancer treatment is warranted.One of the possible pitfalls in our study remains

selective referral. We showed that the likelihood ofbeing operated on was 47% higher for patients diag-nosed in a nonteaching hospital than for those diag-nosed in teaching nonuniversity hospitals. Adjustedfor age, stage, and tumor location, this was evenhigher, 73%, indicating that nonteaching hospitalsconsidered patients with a worse prognosis possible

candidates for surgery, or at least for referral toevaluate resectability. However, adding hospitalreferral frequency to our multivariate analysesshowed no inuence of referral on overall and relativesurvival or RER. The referral pattern showed littleevidence for selective referral. Those nonteachinghospitals that referred patients referred almost all,and the teaching nonuniversity hospitals referredonly few patients for surgery (Table 2). The referralpattern of the nonteaching hospitals implies that theuniversity hospital operated on a priori prognosti-cally worse patients. This is a likely explanation forthe higher number of stage III patients in the uni-versity hospital.To minimize the eventual eect of any residual

selection referral, the relative survival rate was ad-justed for case mix, despite there being no statisticallysignicant dierences in the distribution of age, stage,and sex between the dierent types of hospital.One may suggest that patients who underwent

esophagectomy in the university hospital were mostlyreferred for treatment, thus adding a delay beforesurgery. A consequence could be that these patientshad thus a slightly longer preoperative survival time,estimated to be between 2 and 4 weeks in our study.This short delay could mean that prognosticallyworse patients scheduled for surgery eventually fallout of the surgery category through disease progres-sion during the delay period. However, little is knownin the literature about the eect of longer preopera-tive delays on surgical outcome or eligibility inesophageal cancer. Although patients with advanceddisease may miss surgery through stage progression,patients who do end up having surgery also progress,meaning the university hospital operates on patientswith more advanced disease. This should negativelyinuence the survival outcome and would not explainthe better performance by the university hospital.Furthermore, even if we were very conservative andexcluded all patients who died in the rst 3 months ofour study, the university hospital still performs far

TABLE 5. Adjuvant therapies for patients operated for esophageal cancer diagnosed 19942002, according to hospital oftreatment

Treatment

NonteachingTeaching,

nonuniversity University Total

n % n % N % N %

Surgery 57 98.3 58 96.7 79 83.2 194 91.1Surgery + radiotherapy 1 1.7 5 5.3 6 2.8Surgery + chemotherapy 2 3.3 11 11.6 13 6.1Total 58 100.0 60 100.0 95 100.0 213 100.0



better than teaching nonuniversity and nonteachinghospitals. So although we cannot discount earlymortality as a factor in survival, we think that it isunlikely that the dierence in performance can befully explained by this.We had no information about the operative pro-

cedure that had been performed. Treatment guide-lines indicated a curative surgical approach fortumors encompassing 5 cm of the length of theesophagus, as based on ultrasonographic or radio-logical examination. Surgical resection could be at-tempted for tumors 58 cm in length. For this lastgroup, neoadjuvant chemotherapy, which was pro-vided in the university hospital after proof of locallyadvanced disease, could be attempted to improveresectability. For adenocarcinomas, a transthoraco-abdominal approach with two-field lymphadenecto-my was advised, combining a midline laparotomyand a right-sided thoracotomy. Alternatively a tran-shiatal blind esophagectomy could be performed witha cervical esophagogastrostomy. For distal adeno-carcinomas without Barrett dysplasia, a left-sidedthoracotomy with intrathoracic anastomosis was analternative approach.In general, patients in our population who were

operated on in the university hospital underwent atransthoracic esophagectomy, with the exception ofsupercial T1 tumors, whereas patients treated inregional hospitals frequently underwent surgery witha transhiatal approach. Therefore, the two mainoperative strategies we encountered were transhiatalresection and transthoracoabdominal resection witha two-eld lymph node dissection. There is no evi-dence in literature that the outcome diers for thesetwo procedures,14,20 except for a tendency toward animproved long-term survival in the extended trans-thoracic group in the study of Hulscher et al.21 So theclinical outcome in our population is likely uninu-enced by differences in surgical procedure. However,there is possibly a stage migration effect between themore thorough pathological staging in operationswith lymphadenectomy, as in the university hospital,and understaging in patients undergoing a transhiatalesophageal resection.A few of the operated patients received neoadju-

vant therapy in our study, mostly in the universitysetting, which might account for a small part of thebetter survival in the university hospital. Separateanalysis, however, showed that patients in our study,who received neoadjuvant therapy with surgical re-moval of the tumor, did not perform better thanpatients who underwent surgical removal alone in theuniversity hospital. In other studies, the preoperative

eect of cisplatin-based chemotherapy on both ade-nocarcinoma and squamous cell carcinoma showedno increase in overall survival.2225 Preoperativechemotherapy or radiotherapy can result in down-staging and thus lead to a better resectability, but noclear downstaging was seen in our study when com-paring pre- and postoperative clinical and patholog-ical stage. In several randomized, controlled studies,postoperative radiotherapy demonstrated either noincrease26,27 or a decrease28 in survival comparedwith resection alone. Postoperative chemotherapyhas also been compared with surgical managementalone in several randomized controlled trials, withoutdemonstrating an improvement in survival.29

According to these results, the effect of perioperativetreatment is not likely to inuence our data.We found a tentative relationship between higher

volume and a better relative survival. However, thisissue still is a debatable problem in determiningtreatment guidelines. Therefore, we suggest thatguidelines concerning specic referral of esophagealcancer patients should be based on hospital out-comes, preferably in experienced centers, rather thanon annual numbers of procedures as long as thefactor that is determining patient survival is still un-known.The individual surgeon could be an important

parameter in determining the hospital outcome. Al-though the implications of the assertion that somesurgeons have better outcomes than others makeclinicians uncomfortable, there should be little doubtthat it is true. Variation in performance has beenshown to be related to surgeon characteristics,including surgical volume, subspecialization, and thehospital setting in which they operate.7,30 Individualsurgical experience has been associated with thepostoperative mortality of esophageal cancer. Suttonet al.9 showed a reduction in mortality from 6% to 3%after 150 procedures. Miller et al.14 published resultsin one center demonstrating an operative mortality of22% among esophageal resections performed by asurgeon who performed fewer than six procedures ayear. These results are widely quoted in the surgicalliterature as proving that surgeons without the nec-essary expertise should not perform esophagealresections.There is some evidence that subspecialization

improves outcomes. Herr et al.31 found that pa-tients who underwent radical cystectomy by urologyoncologists had substantially lower rates of localtumor recurrence than those who were operated onby general urologists (6% vs. 23% (P = .006).Dueck et al.32 reported that patients who under-



went surgery for a ruptured abdominal aorticaneurysm had markedly better outcomes when thesurgery was performed by a vascular surgeon ratherthan a general surgeon. The effect of subspecial-ization of the surgeon on the outcome of esopha-geal cancer has not yet been studied, but it may bea promising factor for decisions with regard tocentralization.It has been suggested in previous reports that the

skill of the anesthesia and nursing sta aects mor-bidity and relative survival of esophagectomy pa-tients and that it confounds the surgeons personaloutcome.7,9 Better critical-care experience of thesupport staff may explain a higher relative survival inuniversity hospitals; staff may be more adept at car-ing for esophagectomy patients. Some authors havesuggested that the expertise of the anesthesia andnurses in a hospital is directly correlated to the hos-pital and surgical load in that hospital.9 However,expertise can be acquired elsewhere, and expertiseonly develops through effective feedback, not only bynumber of patients.Although referral to dedicated centers possibly

results in improved relative survival, the focus entailssome disadvantages, which should be considered.Referral to centers means that many patients have totravel to distant sites, which can create hardship forthe patient and his or her family. In-hospital familysupport and postoperative follow-up are more di-cult when the hospital is farther away. Finlaysonet al.33 demonstrated that 45% of the patients preferto stay in their local area even if the projected oper-ative mortality is doubled. However, that study rep-resents the American situation, and it is questionablewhether distance is perceived to be a problem in theDutch situation. In our region, which has a relativelyhigh density of hospitals, a recent patient surveyshowed that traveling distance was not considered acritical issue.In conclusion, we demonstrated that in our region,

the relative survival for patients operated on foresophageal cancer is better in the university hospitalcompared with teaching nonuniversity and non-teaching hospitals, emphasizing the need for referralto centers focused on the treatment of esophagealcancer. The underlying parameter for the observeddierence remains unclear. We suggest that centers atleast periodically review the morbidity and mortalityrates of esophageal resections to assess their outcomeand the possibility of referral. Eligibility for centersfocused on esophageal cancer treatment shouldtherefore be based on patient outcomes rather thanon patient numbers.

REFERENCES

1. Mody RP. Carcinoma oesophagusoverview, update of lit-erature. J Indian Med Assoc 2002; 100:56972.

2. Gillis CR, Hole DJ. Survival outcome of care by specialistsurgeons in breast cancer: a study of 3786 patients in the westof Scotland. BMJ 1996; 312:1458.

3. Kehoe S, Powell J, Wilson S, Woodman C. The inuence of theoperating surgeons specialisation on patient survival in ovar-ian carcinoma. Br J Cancer 1994; 70:10147.

4. Sainsbury R, Haward B, Rider L, Johnston C, Round C.Inuence of clinician workload and patterns of treatment onsurvival from breast cancer. Lancet 1995; 345:126570.

5. Junor EJ, Hole DJ, Gillis CR. Management of ovarian cancer:referral to a multidisciplinary team matters. Br J Cancer 1994;70:36370.

6. Harding MJ, Paul J, Gillis CR, Kaye SB. Management ofmalignant teratoma: does referral to a specialist unit matter?.Lancet 1993; 341:9991002.

7. Birkmeyer JD, Stukel TA, Siewers AE, Goodney PP, Wenn-berg DE, Lucas FL. Surgeon volume and operative mortalityin the United States. N Engl J Med 2003; 349:211727.

8. Gillison EW, Powell J, McConkey CC, Spychal RT. Surgicalworkload and outcome after resection for carcinoma of theoesophagus and cardia. Br J Surg 2002; 89:3448.

9. Sutton DN, Wayman J, Grifn SM. Learning curve foroesophageal cancer surgery. Br J Surg 1998; 85:1399402.

10. Bachmann MO, Alderson D, Edwards D, et al. Cohort studyin South andWest England of the inuence of specialization onthe management and outcome of patients with oesophagealand gastric cancers. Br J Surg 2002; 89:91422.

11. Dudley RA, Johansen KL, Brand R, Rennie DJ, Milstein A.Selective referral to high-volume hospitals: estimating poten-tially avoidable deaths. JAMA 2000; 283:115966.

12. Swisher SG, Deford L, Merriman KW, et al. Effect of opera-tive volume on morbidity, mortality, and hospital use afteresophagectomy for cancer. J Thorac Cardiovasc Surg 2000;119:112632.

13. Matthews HR, Powell DJ, McConkey CC. Effect of surgicalexperience on the results of resection for oesophageal carci-noma. Br J Surg 1986; 73:6213.

14. Miller JD, Jain MK, de Gara CJ, Morgan D, Urschel JD.Effect of surgical experience on results of esophagectomy foresophageal carcinoma. J Surg Oncol 1997; 65:201.

15. International Union Against Cancer (UICC). TNM Classi-cation of Malignant Tumors Berlin: Springer-Verlag; 1992.

16. International Union Against Cancer (UICC) TNM Classi-cation of Malignant Tumors New York: Wiley-Liss; 1997.

17. Ederer F, Heise H Instructions to IBM 650 Programmers inProcessing Survival Computations: Methodological Note No.10. Bethesda, MD: National Cancer Institute, 1959.

18. Dickman PW, Sloggett A, Hills M, Hakulinen T. Regressionmodels for relative survival. Stat Med 2004; 23:5164.

19. van Lanschot JJ, Hulscher JB, Buskens CJ, Tilanus HW, tenKate FJ, Obertop H. Hospital volume and hospital mortalityfor esophagectomy. Cancer 2001; 91:15748.

20. Sabanathan S, Shah R, Mearns AJ, Richardson J, Goulden C,Shakir T. Results of surgical treatment of oesophageal cancer.J R Coll Surg Edinb 1996; 41:295301.

21. Hulscher JB, van Sandick JW, de Boer AG, et al. Extendedtransthoracic resection compared with limited transhiatalresection for adenocarcinoma of the esophagus. N Engl J Med2002; 21:16629.

22. Kelsen DP, Ginsberg R, Pajak, et al. Chemotherapy followedby surgery compared with surgery alone for localized esopha-geal cancer. N Engl J Med 1998;339:197984.

23. Lam KY, Law S, Ma LT, Ong SK, Wong J. Pre-operativechemotherapy for squamous cell carcinoma of the oesophagus:do histological assessment and p53 overexpression predictchemo-responsiveness? Eur J Cancer 1997; 33:12215.



24. Roth JA, Pass HI, Flanagan MM, Graeber GM, RosenbergJC, Steinberg S. Randomized clinical trial of preoperative andpostoperative adjuvant chemotherapy with cisplatin, vindesine,bleomycin for carcinoma of the esophagus. J Thorac Cardio-vasc Surg 1988; 96:2428.

25. Schlag PM. Randomized trial of preoperative chemotherapyfor squamous cell cancer of the esophagus. The ChirurgischeArbeitsgemeinschaft Fuer Onkologie der Deutschen Gesell-schaft Fuer Chirurgie Study Group. Arch Surg 1992; 127:144650.

26. Teniere P, Hay JM, Fingerhut A, Fagniez PL. Postoperativeradiation therapy does not increase survival after curativeresection for squamous cell carcinoma of the middle, loweresophagus as shown by a multicenter controlled trial. FrenchUniversity Association for Surgical Research. Surg GynecolObstet 1991; 173:12330.

27. Zieren HU, Muller JM, Jacobi CA, Pichlmaier H, Muller RP,Staar S. Adjuvant postoperative radiation therapy after cura-tive resection of squamous cell carcinoma of the thoracicesophagus: a prospective randomized study. World J Surg1995; 19:4449.

28. Fok M, Sham JS, Choy D, Cheng SW, Wong J. Postoperativeradiotherapy for carcinoma of the esophagus: a prospective,randomized controlled study. Surgery 1993; 113:13847.

29. Ando N, Iizuka T, Ide H, et al. Surgery plus chemotherapycompared with surgery alone for localized squamous cellcarcinoma of the thoracic esophagus: a Japan ClinicalOncology Group StudyJCOG9204. J Clin Oncol 2003;21:45926.

30. Porter GA, Soskolne CL, Yakimets WW, Newman SC. Sur-geon-related factors and outcome in rectal cancer. Ann Surg1998; 227:15767.

31. Herr HW, Faulkner JR, Grossman HB, et al. Surgical factorsinuence bladder cancer outcomes: a cooperative group report.J Clin Oncol 2004; 22:27819.

32. Dueck AD, Kucey DS, Johnston KW, Alter D, Laupacis A.Survival after ruptured abdominal aortic aneurysm: effect ofpatient, surgeon, and hospital factors. J Vasc Surg 2004;39:125360.

33. Finlayson TL, Moyer CA, Sonnad SS. Assessing symptoms,disease severity, and quality of life in the clinical context: atheoretical framework. Am J Manag Care 2004; 10:33644.



Better Survival in Patients with Esophageal Cancer After Surgical Treatment in University Hospitals: A Plea for Performance by Surgical OncologistsAbstractPATIENTS AND METHODSPatientsData Collection by the Regional Cancer RegistryGuidelines for Staging and TreatmentStatistics and Definitions of VariablesRESULTSTab1Tab2Tab3Fig1DISCUSSIONTab4Tab5ReferencesCR1CR2CR3CR4CR5CR6CR7CR8CR9CR10CR11CR12CR13CR14CR15CR16CR17CR18CR19CR20CR21CR22CR23CR24CR25CR26CR27CR28CR29CR30CR31CR32CR33

/ColorImageDict > /JPEG2000ColorACSImageDict > /JPEG2000ColorImageDict > /AntiAliasGrayImages false /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 150 /GrayImageDepth -1 /GrayImageDownsampleThreshold 1.50000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict > /GrayImageDict > /JPEG2000GrayACSImageDict > /JPEG2000GrayImageDict > /AntiAliasMonoImages false /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 600 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict > /AllowPSXObjects false /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox true /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile (None) /PDFXOutputCondition () /PDFXRegistryName (http://www.color.org?) /PDFXTrapped /False

/Description >>> setdistillerparams> setpagedevice

Documents

Better Survival in Patients with Esophageal Cancer After Surgical Treatment in University Hospitals: A Plea for Performance by Surgical Oncologists