Hospital Coding of Postoperative Ileus: A Prospective Study.

Background Postoperative ileus (POI) is among the most common complications affecting patients who undergo major abdominal surgery. Because of the high volume of major surgery and the high incidence of postoperative ileus, failure to code for this complication may have a significant impact on hospital reimbursement and quality measures. Objectives This paper investigates the magnitude of the difference between the prevalence of POI as coded in administrative data versus the prevalence based upon a prospectively applied operational definition of POI in patients undergoing intestinal resection surgery. Methods Data was collected during the course of a prospective study at the University of Iowa Hospitals & Clinics on an investigational digital health device for predicting operationally defined POI. Following the first 24 hours post-surgery, a patient was identified as experiencing POI as operationally defined by the occurrence of vomiting, reversal of diet and/or placement of a nasogastric tube. For all subjects, billing data was also collected. Results A total of 203 adult patients undergoing intestinal resection surgery consented to participate. Of patients who developed POI based on the operational definition, 35% were not coded accordingly to capture appropriate risk adjustment and reimbursement. Conclusions Patients who experienced indicators of POI but who were not coded experienced over two days of additional time in the hospital compared to patients who did not experience POI, representing significant unreimbursed costs. Timing and duration of POI indicators appear to impact coding discrepancies and may suggest means for improving caregiver identification of POI in a patient's medical record.

The "Classification Pendulum" of Stage I Colorectal Cancer: A National Level Analysis of the Survival Difference Between T1 and T2 Colorectal Cancer.

BACKGROUND: The eighth edition of the American Joint Committee on Cancer classifies nonmetastatic, node-negative colorectal cancers invading the submucosa (T1) and muscularis propria (T2) as stage I tumors without additional subclassification. OBJECTIVE: The aim of the study was to compare survival of T1N0M0 versus T2N0M0 colorectal cancers and to investigate factors associated with decreased survival. DESIGN: This was an analysis of 2 large population-based data sets. SETTINGS: The study was conducted analyzing data from the Surveillance Epidemiology and End Result program and the National Cancer Database. PATIENTS: Adult patients undergoing major resection without additional therapy for stage I colorectal cancer were included. MAIN OUTCOME MEASURES: Overall and disease-specific survival for T1 versus T2 cancers were measured. Subgroup analyses by tumor location (colon versus rectum) were performed. RESULTS: A total of 30,228 (36.4% T1 and 63.6% T2) and 41,670 (41.1% T1 and 58.9% T2) patients were identified in the Surveillance Epidemiology and End Result database and the National Cancer Database. The 5-year overall survival rates were 87.1% and 86.2% for patients with T1 versus 82.7% and 80.7% for patients with T2 (p < 0.001) in the Surveillance Epidemiology and End Result database and the National Cancer Database. The 10-year overall survival rates were 71.3% and 66.3% for patients with T1 versus 62.2% and 57.2% for patients with T2 tumors (p < 0.001) in the Surveillance Epidemiology and End Result database and the National Cancer Database. The 5- and 10-year disease-specific survival for colorectal cancer in the Surveillance Epidemiology and End Result database was 97.0% (T1) versus 95.2% (T2) and 94.1% (T1) versus 90.3% (T2). Black race (HR = 1.26 and 1.65 for overall survival and disease-specific survival in the Surveillance Epidemiology and End Result database; HR = 1.20 for overall survival in the National Cancer Database) was associated with worse survival. LIMITATIONS: The study was limited by intrinsic biases related to large administrative data sets. CONCLUSIONS: Within stage I colorectal cancer, T2 tumors have decreased overall survival and disease-specific survival as compared with T1 cancers. This survival difference may justify revising the American Joint Committee on Cancer staging system to include the subclassification of stage Ia (T1N0M0) and stage Ib (T2N0M0). See Video Abstract at http://links.lww.com/DCR/B659. LA CLASIFICACIN PNDULO PARA EL CNCER COLORRECTAL EN ESTADIO I UN ANLISIS A NIVEL NACIONAL DE LA DIFERENCIA DE SOBREVIDA ENTRE EL CNCER COLORRECTAL T Y T: ANTECEDENTES:La octava edición del American Joint Committee on Cancer, clasifica los cánceres colorrectales no metastásicos con ganglios negativos, que invaden la submucosa (T1) y la muscularis propia (T2) como tumores en estadio I sin subclasificación adicional.OBJETIVO:El objetivo del estudio fue comparar la sobrevida de los cánceres colorrectales T1N0M0 versus T2N0M0 e investigar los factores asociados con la disminución de la sobrevida.DISEÑO:Análisis de dos grandes conjuntos de datos poblacionales.MARCO:El estudio se realizó analizando datos del Programa de Epidemiología de Vigilancia y Resultados Finales (SEER) y la Base de Datos Nacional del Cáncer.PACIENTES:Pacientes adultos en los cuales se realizó una resección mayor sin terapia adicional por cáncer colorrectal en estadio I.PRINCIPALES VARIABLES ANALIZADAS:Sobrevida global y específica de la enfermedad para los cánceres T1 versus T2. Se realizó un análisis de subgrupos según la ubicación del tumor (colon versus recto).RESULTADOS:Se incluyeron un total de 30.228 (36,4% T1 y 63,6% T2) y 41.670 (41,1% T1 y 58,9% T2) pacientes en las bases de datos SEER y la Base de Datos Nacional del Cáncer, respectivamente. La sobrevida global a 5 años fue del 87,1% y el 86,2% para los pacientes con T1 frente al 82,7% y el 80,7% de los pacientes con T2 (p < 0,001) en el SEER y la Base de Datos Nacional del Cáncer, respectivamente. La sobrevida global a 10 años fue del 71,3% y el 66,3% para los pacientes con T1 frente al 62,2% y el 57,2% de los pacientes con tumores T2 (p < 0,001) en el SEER y la Base de Datos Nacional del Cáncer, respectivamente. La sobrevida específica de la enfermedad a 5 y 10 años para el cáncer colorrectal en el SEER fue del 97,0% (T1) frente al 95,2% (T2) y del 94,1% (T1) frente al 90,3% (T2), respectivamente. La grupo étnico afroamericano se asoció con una sobrevida menor (Hazard Ratio -HR 1,26 y 1,65 para la sobrevida general y sobrevida específica de la enfermedad-SEER; HR 1,20 para la sobrevida general-Base de de Datos Nacional del Cáncer).LIMITACIONES:Sesgos intrínsecos relacionados con el análisis de grandes conjuntos de datos.CONCLUSIONES:Dentro del cáncer colorrectal en estadio I, los tumores T2 han disminuido la sobrevida general y la sobrevida específica de la enfermedad, en comparación con los cánceres T1. Esta diferencia de sobrevida puede justificar la revisión del sistema de estadificación del American Joint Committee on Cancer para incluir la subclasificación del estadio Ia (T1N0M0) y el estadio Ib (T2N0M0). Consulte Video Resumen en http://links.lww.com/DCR/B659.

Post-operative Complication Rate Comparison Between Airway Surgery and Upper Airway Stimulation Using NSQIP and ADHERE.

OBJECTIVE: Postoperative complication rates were compared between obstructive sleep apnea surgery (OSAS) and hypoglossal nerve upper airway stimulation (UAS). STUDY DESIGN: Cohort. SETTING: Multi-institutional international databases. METHODS: OSAS data were collected from the NSQIP database (2014; American College of Surgeons National Surgery Quality Improvement Program). UAS data were obtained from the ADHERE registry (Adherence and Outcome of Upper Airway Stimulation for OSA International Registry; 2016-December 2019). ADHERE comorbidities and complications were categorized to match NSQIP definitions. A chi-square test was used for proportion P values. RESULTS: There were 1623 UAS procedures in ADHERE and 310 in NSQIP. The UAS group was older than the OSAS group (mean ± SD, 60 ± 11 vs 42 ± 13 years) but similarly male (75% vs 77%) and overweight (body mass index, 29 ± 4 vs 29 ± 3 kg/m2). There was a higher proportion of hypertension, diabetes, and heart disease in the UAS cohort. Palatopharyngoplasty was the most common surgical procedure (71%), followed by tonsillectomy (25%). UAS operative time was longer (132 ± 47 vs 54 ± 33 minutes). Postoperative length of stay was not normally distributed, as 71% of UAS stays were <1 day as opposed to 40% of OSA stays (P < .0001). Thirty-day return to the operating room related to the procedure was 0.1% for UAS and 4.8% for OSAS (P < .0001). Surgical site infections were 0.13% for UAS and 0.9% for OSAS (P = .046). CONCLUSION: The UAS cohort was older and more likely to have comorbid hypertension, diabetes, and heart disease. Despite baseline differences, the postoperative complication rate was lower with UAS than with OSAS.

Male sex, ostomy, infection, and intravenous fluids are associated with increased risk of postoperative ileus in elective colorectal surgery.

BACKGROUND: Postoperative ileus is a common and costly complication after elective colorectal surgery. Effects of intravenous fluid administration remain controversial, and the effect of ostomy construction has not been fully evaluated. Various restrictive intravenous fluid protocols may adversely affect renal function. We aimed to investigate the impact of intestinal reconstruction and intravenous fluid on ileus and renal function after colorectal resection under an enhanced recovery protocol. METHODS: A retrospective study of a prospectively maintained institutional database for a tertiary academic medical center following National Surgical Quality Improvement Program standards was reviewed, analyzing elective colorectal resections performed under enhanced recovery protocol from 2015 to 2018. Postoperative ileus was defined as nasogastric decompression, nil per os >3 days postoperatively, or nasogastric tube insertion. Patients with and without ileus were compared. Intravenous fluid and different anastomoses and ostomies were investigated. Acute kidney injury was a secondary outcome, due to the potential of renal damage with restriction of intravenous fluid volume during and after surgery and controversy in current literature in this matter. RESULTS: Postoperative ileus occurred in 18.5% of patients (n = 464). Male sex (odds ratio 1.97, 95% confidence interval 1.12-3.52) and postoperative infection (odds ratio 2.13, 95% confidence interval 1.03-4.35) were associated with ileus. Compared to colorectal anastomosis, ileostomy/ileorectal anastomosis had the highest risk of ileus (odds ratio 4.9, 95% confidence interval 2.33-11.3), colostomy second highest (odds ratio 3.3, 95% confidence interval 1.35-8.39), while ileocolic anastomosis did not significantly differ (odds ratio 2.06, 95% confidence interval 0.69-5.85) on multivariate analysis. Each liter of intravenous fluid within the first 72 hours significantly correlated with postoperative ileus (odds ratio 1.41, 95% confidence interval 1.27-1.59). Rates of acute kidney injury did not differ (P = .18). CONCLUSION: Each additional liter of intravenous fluid given in the first 72 hours increased the risk of postoperative ileus 1.4-fold. There is substantially higher risk of ileus with male sex, infection, ileostomy/ileorectal anastomosis, and colostomy. Judicious use of intravenous fluid, as described in our enhanced recovery protocol, is not detrimental for renal function in the setting of normal baseline.

Delirium detection by a novel bispectral electroencephalography device in general hospital.

AIM: Delirium is common and dangerous among elderly inpatients; yet, it is underdiagnosed and thus undertreated. This study aimed to test the diagnostic characteristics of a noninvasive point-of-care device with two-channel (bispectral) electroencephalography (EEG) for the screening of delirium in the hospital. METHODS: Patients admitted to the University of Iowa Hospitals and Clinics were assessed for the presence of delirium with a clinical assessment, the Confusion Assessment Method for Intensive Care Unit and Delirium Rating Scale. Subsequently, we obtained a 10-min bispectral EEG (BSEEG) recording from a hand-held electroencephalogram device during hospitalization. We performed power spectral density analysis to differentiate between those patients with and without delirium. RESULTS: Initially 45 subjects were used as a test dataset to establish a cut-off. The BSEEG index was determined to be a significant indicator of delirium, with sensitivity 80% and specificity 87.7%. An additional independent validation dataset with 24 patients confirmed the validity of the approach, with a sensitivity of 83.3% and specificity of 83.3%. CONCLUSION: In this pilot study, the BSEEG method was able to distinguish delirious patients from non-delirious patients. Our data showed the feasibility of this technology for mass screening of delirium in the hospital.

Prospective Validation of the Iowa Rectal Surgery Risk Calculator.

BACKGROUND: The Iowa Rectal Surgery Risk Calculator estimates risk for proctectomy procedures. The Iowa Calculator performed well on NSQIP 2010-2011 training and 2005-2009 validation datasets, but was not prospectively validated and did not include low anterior resections. This study sought to demonstrate validity on new independent data, to update the calculator to include low anterior resection, and to compare performance to other risk assessment tools. METHODS: Non-emergent ACS-NSQIP proctectomy and low anterior resection data from 2010 to 2015 (n = 65,683) were included. The Iowa Calculator generated risk estimates for 30-day morbidity using 2012-2015 data. An Updated Calculator used 2010-2011 training data to include low anterior resection, with validation on 2012-2015 data. NSQIP data provided NSQIP Morbidity Model predictions and a custom web-script collected ACS-NSQIP Online Surgical Risk Calculator predictions for all patients. RESULTS: Proctectomy morbidity (not including low anterior resection) decreased from 40.4% in 2010-2011 to 37.0% in 2012-2015. Low anterior resection had lower morbidity (22.4% in 2012-15). The Iowa Calculator demonstrated good discrimination and calibration using 2012-2015 data (C-statistic 0.676, deviance + 9.2%). After including low anterior resection, the Updated Iowa Calculator performed well during training (c-statistic 0.696, deviance 0%) and validation (C-statistic 0.706, deviance + 7.9%). The Updated Iowa Calculator had significantly better discrimination and calibration than morbidity predictions from the ACS Online Calculator (C-statistic 0.693, P < 0.001, deviance - 28.1%) and NSQIP General/Vascular Surgery Model (C-statistic 0.703, P < 0.05, deviance - 40.8%). CONCLUSION: When applied to new independent data, the Iowa Calculator supplies accurate risk estimates. The Updated Iowa Calculator includes low anterior resection, and both are prospectively validated. Risk estimation by the Iowa Calculators was superior to ACS-provided risk tools.

Postoperative complications and patient satisfaction: does payer status have an impact?

BACKGROUND: Patient demographics and outcomes may influence patient satisfaction. We aim to investigate the relationship between postoperative complications and survey-based satisfaction in the context of payer status. METHODS: Institutional data were used to identify major complication occurrence and linked to patient satisfaction surveys. The impact of complication occurrence on satisfaction was investigated and stratified by payer status. RESULTS: In all, 1,597 encounters were identified with an 18% major complication rate. Satisfaction scores in specific domains were significantly more likely to be above the median for patients without complications (P < .01) and for payer status Medicaid/low income (P < .05). In sensitivity analyses, we found no significant interactions among payer status, complications, and satisfaction scores. CONCLUSIONS: Significant differences exist for individual satisfaction survey domains between patients with and without major postoperative complications and by payer status. Payer status was not found to have an impact on the intersection of major complications and patient satisfaction.

Hospital Characteristics Associated with Stage II/III Rectal Cancer Guideline Concordant Care: Analysis of Surveillance, Epidemiology and End Results-Medicare Data.

BACKGROUND: Evidence suggests that high-volume facilities achieve better rectal cancer outcomes. METHODS: Logistic regression was used to evaluate association of facility type with treatment after adjusting for patient demographics, stage, and comorbidities. SEER-Medicare beneficiaries who were diagnosed with stage II/III rectal adenocarcinoma at age ≥66 years from 2005 to 2009 and had Parts A/B Medicare coverage for ≥1 year prediagnosis and postdiagnosis plus a claim for cancer-directed surgery were included. Institutions were classified according to National Cancer Institute (NCI) designation, presence of residency program, or medical school affiliation. RESULTS: Two thousand three hundred subjects (average age = 75) met the criteria. Greater proportions of those treated at NCI-designated facilities received transrectal ultrasound (TRUS) or magnetic resonance imaging (MRI)-pelvis (62.1 vs. 29.9 %), neoadjuvant chemotherapy (63.9 vs. 41.8 %), and neoadjuvant radiation (70.8 vs. 46.3 %), all p < 0.0001. On multivariate analysis, odds ratios (95 % confidence intervals) for receiving TRUS or MRI, neoadjuvant chemotherapy, or neoadjuvant radiation among beneficiaries treated at NCI-designated facilities were 3.51 (2.60-4.73), 2.32 (1.71-3.16), and 2.66 (1.93-3.67), respectively. Results by residency and medical school affiliation were similar in direction to NCI designation. CONCLUSIONS: Those treated at hospitals with an NCI designation, residency program, or medical school affiliation received more guideline-concordant care. Initiatives involving provider education and virtual tumor boards may improve care.

Association between surgeon characteristics and their preferences for guideline-concordant staging and treatment for rectal cancer.

BACKGROUND: Rectal cancer guidelines recommend transrectal ultrasound or magnetic resonance imaging for locoregional staging and neoadjuvant chemoradiation therapy (CRT) for Stage II/III disease, but studies show these are underutilized. We examined how surgeon preferences align with guidelines or vary by training. METHODS: Questionnaires on training, years of practice, and staging/treatment preferences were sent to surgeons practicing in Florida. RESULTS: Of 759 surveys distributed, 321 (42%) responded; 158 were excluded because they were trainees, not treating rectal cancer, or not board certified/eligible. Among the remaining 163, 71% were general surgeons, 18% colorectal surgeons, and 11% surgical oncologists. Colorectal surgeons and surgical oncologists were more likely than general surgeons to prefer transrectal ultrasound/magnetic resonance imaging (79% vs 50%; P < .01), and neoadjuvant CRT (71% vs 45%; P < .01). Differences remained significant after adjusting for years in practice. CONCLUSION: Increased focus on appropriate use of staging procedures and neoadjuvant CRT within general surgery training/educational programs is warranted.

Effect of BMI on outcomes in proctectomy.

BACKGROUND: The unique surgical challenges of proctectomy may be amplified in obese patients. We examined surgical outcomes of a large, diverse sample of obese patients undergoing proctectomy. OBJECTIVE: The purpose of this work was to determine whether increased BMI is associated with increased complications in proctectomy. DESIGN: This was a retrospective review. SETTINGS: The study uses the American College of Surgeons National Surgical Quality Improvement Program database (2010 and 2011). PATIENTS: Patients included were those undergoing nonemergent proctectomy, excluding rectal prolapse cases. Patients were grouped by BMI using the World Health Organization classifications of underweight (BMI <18.5); normal (18.5-24.9); overweight (25.0-29.9); and class I (30.0-34.9), class II (35.0-39.9), and class III (≥40.0) obesity. MAIN OUTCOME MEASURES: We analyzed the effect of preoperative and intraoperative factors on 30-day outcomes. Continuous variables were compared with Wilcoxon rank-sum tests and proportions with the Fisher exact or χ tests. Logistic regression controlled for the effects of multiple risk factors. RESULTS: Among 5570 patients, class I, II, and III obesity were significantly associated with higher rates of overall complications (44.0%, 50.8%, and 46.6% vs 38.1% for normal-weight patients; p < 0.05). Superficial wound infection was significantly higher in classes I, II, and III (11.6%, 17.8%, and 13.0% vs 8.0% for normal-weight patients; p < 0.05). Operative times for patients in all obesity classes were significantly longer than for normal-weight patients. On multivariate analysis, an obese BMI independently predicted complications; ORs (95% CIs) were 1.36 (1.14-1.62) for class I obesity, 1.99 (1.54-2.54) for class II, and 1.42 (1.02-1.96) for class III. LIMITATIONS: This study was a retrospective design with limited follow-up. CONCLUSIONS: Class I, II, and III obese patients were at significantly increased risk for morbidity compared with normal BMI patients. Class II obese patients had the highest rate of complications, a finding that deserves further investigation.

Development of an improved risk calculator for complications in proctectomy.

BACKGROUND: Rectal surgery is associated with high complication rates, but tools to prospectively define surgical risk are lacking. Improved preoperative risk assessment could better inform patients and refine decision making by surgeons. Our objective was to develop a validated model for proctectomy risk prediction. METHODS: We reviewed non-emergent ACS-NSQIP proctectomy data from 2005 to 2011 (n = 13,385). Logistic regression identified variables available prior to surgery showing independent association with 30-day morbidity in 2010-2011 (n = 5,570). The resulting risk model's discrimination and calibration were tested against the NSQIP-supplied morbidity model, and performance was validated against independent 2005-2009 data. RESULTS: Overall morbidity for proctectomy in 2010-2011 was 40.2%; significantly higher than the 23.0 % rate predicted by the NSQIP-provided general and vascular surgery risk model. Frequent complications included bleeding (16.3%), superficial infection (9.2%), and sepsis (7.4%). Our novel model incorporating 17 preoperative variables provided better discrimination and calibration (p < 0.05) than the NSQIP model and was validated against the 2005-2009 data. A web-based calculator makes this new model available for prospective risk assessment. CONCLUSIONS: We conclude that the NSQIP-supplied risk model underestimates proctectomy morbidity and that this new, validated risk model and risk prediction tool ( http://myweb.uiowa.edu/sksherman ) may allow clinicians to counsel patients with accurate risk estimates using data available in the preoperative setting.

Is the abdominal repair of rectal prolapse safer than perineal repair in the highest risk patients? An NSQIP analysis.

BACKGROUND: Although the perineal approach in the surgical management of rectal prolapse has a higher recurrence, it is the accepted approach for higher-risk patients because of its lower morbidity. OBJECTIVE: The aim of this study was to determine outcomes of abdominal versus perineal approaches to rectal prolapse repair. DESIGN SETTINGS: A retrospective study was performed comparing outcomes of patients undergoing different types of surgical approaches (open abdominal, laparoscopic, perineal) for rectal prolapse. PATIENTS: The American College of Surgeons National Surgical Quality Improvement Participant User Data Files (2008-2009) were queried for patients undergoing adult, elective procedures for rectal prolapse. MAIN OUTCOME MEASURES: Univariate analysis and multivariate logistic regression were performed to look at age, ASA classification, procedure type, and resultant mortality rate. RESULTS: One thousand four hundred sixty-nine patients meeting our criteria were identified. Older patients (age>80) and higher-risk patients (ASA classifications 3 and 4) were significantly associated with the selection of the perineal approach. The overall mortality rate was 0.5%. The mortality rate for all perineal procedures was 0.9% in comparison with 0.13% for all abdominal operations (p = 0.033). The mortality rate for the highest-risk groups (ASA 3 and 4) for perineal procedures was 1.3% in comparison with 0.35% in the abdominal procedure group; the relative risk for mortality was 4 times greater in the perineal procedure group than in the abdominal procedure group. LIMITATIONS: The retrospective design and standardized outcomes measured use administrative-level data and prevent the assessment of procedure-specific outcomes. CONCLUSIONS: Hospital mortality for the surgical repair of rectal prolapse is uncommon. The decision to choose the abdominal approach for the repair of rectal prolapse may not be as prohibitive as previously thought for higher-risk patients. Because of the broad range of functionality within each ASA classification, the operation offered should always be individualized, and patient selection is the most important factor.

The alpha2 and alpha3 integrins are required for morphologic differentiation of an intestinal epithelial cell line.

BACKGROUND: The molecular mechanisms controlling intestinal epithelial cell differentiation are poorly defined because of the difficulty of growing normal intestinal cells. We have taken advantage of the ability of the Caco-2 cell line to acquire a glandular phenotype in 3-dimensional (3-D) culture systems to investigate the role of alpha2 and alpha3 integrins in morphologic differentiation. METHODS: Caco-2 cells transfected with sense or antisense DNA constructs of alpha2 or alpha3 integrins were grown in 3-D Matrigel or collagen I in the presence or absence of integrin function-blocking antibodies. We used light and confocal microscopy, BrDU incorporation, TUNEL assay, a fluorometric adhesion assay, FACS analysis, and Western blot analysis to study the effect of extracellular matrix (ECM) and integrins on morphology, polarization, proliferation, apoptosis, cell adhesion, and integrin expression. RESULTS: Compared to collagen I, Caco-2 cells cultured in 3-D Matrigel display cytoskeletal and adherens junction rearrangements and decreased proliferation consistent with cellular differentiation. These changes, which are inhibited by alpha2 and alpha3 blocking monoclonal antibodies and alpha2 and alpha3 antisense DNA transfection, were associated with an increase in alpha3 integrin expression. CONCLUSIONS: We demonstrated that signaling through both constitutively expressed alpha2 integrin and Matrigel-induced alpha3 integrin expression is required to acquire a differentiated phenotype in Caco-2 cells.

Salvage radical surgery after failed local excision for early rectal cancer.

OBJECTIVES: Local recurrence after transanal excision of rectal cancer is often amenable to salvage radical proctectomy, but the long-term results remain unknown. This study was designed to determine the outcome of salvage radical surgery after failed local excision in patients with early rectal cancer. METHODS: We retrospectively reviewed the charts of 29 patients who underwent salvage radical surgery for local recurrence after a full-thickness transanal excision for Stage I rectal cancer. End points included local and distant recurrences and disease-free survival after salvage radical surgery. Comparisons between groups were performed by chi-squared test. RESULTS: Recurrence involved the rectal wall in 26 patients (90 percent) and was purely extrarectal in only 3 (10 percent). Mean time between local excision and radical operation was 26 months. The resection was considered curative in 23 patients (79 percent). The stage of the recurrent tumor was more advanced than the primary tumor in 27 patients (93 percent). At a mean follow-up of 39 (range, 2-147) months after radical surgery, 17 patients (59 percent) remained free of disease. The disease-free survival rate was 68 percent for patients with tumors with favorable histology vs. 29 percent for patients with tumors with unfavorable histology. CONCLUSION: Salvage surgery for recurrence after local excision of rectal cancers may not provide results equivalent to those of initial radical treatment. In the present study the poor results of salvage surgery emphasize the importance of appropriate selection of the initial treatment of Stage I rectal cancer.