Use of the preoperative levels of CEA in patients with colorectal cancer.

BACKGROUND/AIMS: Preoperative carcinoembryonic antigen has been considered useful as a prognostic factor and recurrence indicator of colorectal neoplasms. However, its diagnostic ability related to some parameters such as resectability or tumor staging has been less studied. The aim of this study was to evaluate the use of this marker as a diagnostic test for these parameters of colorectal cancer. METHODOLOGY: In a sample of 283 patients operated on for colorectal carcinoma data were retrospectively recorded corresponding to preoperative carcinoembryonic antigen, type of surgery performed (curative vs. non-curative), tumor intramural spread, lymph node involvement, distant metastasis, TNM stage, tumoral differentiation, survival time and survival time free of disease. Bivariate analysis between carcinoembryonic antigen and the rest of the parameters was performed. Also, the sensitivity, specificity, positive predictive value, negative predictive value and diagnostic accuracy (efficiency) of the marker considered as a diagnostic test, were analyzed in relation to the tumoral resectability and tumoral spread. RESULTS: Preoperative level of serum carcinoembryonic antigen was statistically significantly associated with type of surgery performed (p < 0.001); tumoral intramural spread (p = 0.001); lymph node involvement (p < 0.001); presence of distant metastasis (p < 0.001); TNM staging (p < 0.001); overall survival period (p < 0.001) and disease-free survival time (p = 0.04). There was no relation between carcinoembryonic antigen and the degree of tumoral differentiation. The carcinoembryonic antigen's greatest sensitivity corresponded with the prediction for the type of the surgery performed and with the existence of distant metastasis, 80% and 70.9%, respectively. The negative predictive value was also high (> 90% in both cases). When considering the intramural spread, the specificity was 82% and its positive predictive value 93.1%. CONCLUSIONS: Preoperative levels of serum carcinoembryonic antigen, considered as a diagnostic test, are useful as predictors of resectability and tumor spread in colorectal carcinoma.

Factores predictivos de la recidiva de cáncer colorrectal / Predictive factors for colorectal cancer recurrence

Introducción. El seguimiento estrecho y no selectivo de todos los pacientes intervenidos de cáncer colorrectal (CCR) supone un gran consumo de recursos y tiempo, cuya necesidad se ha cuestionado. Sería ideal conocer a priori a aquellos pacientes con mayor probabilidad de recidiva que serían los que verdaderamente se van a beneficiar de una estrecha vigilancia. Objetivos. Definir una población de alto riesgo de recidiva de CCR según parámetros comúnmente empleados en la clínica y laboratorio. Pacientes y método. Estudio retrospectivo, descriptivo y longitudinal de 398 pacientes intervenidos de CCR en nuestra institución entre 1988 y 1995. Se recogieron los datos correspondientes a filiación, antecedentes personales, síntomas clínicos relevantes (masa palpable, anemia, perforación, oclusión); datos de laboratorio; marcadores tumorales, localización del tumor, tipo y carácter de la intervención, estadio TNM, grado de diferenciación, presencia de mucina, transfusiones perioperatorias, complicaciones postoperatorias, infecciones, recidiva y supervivencia libre de enfermedad. El tiempo medio de seguimiento fue de 33,6 meses. Se analizó la relación de estas variables con la supervivencia libre de recidiva mediante el test de rangos logarítmicos. Posteriormente, se introdujeron las variables significativas en un modelo de regresión de Cox para determinar los factores pronósticos independientes que se relacionaban con la recidiva. Resultados. La muestra estudiada constaba de 398 pacientes, 204 varones (51,4 por ciento) y 194 mujeres (48,7 por ciento); edad media: 65,30 años (rango: 21-91) (DE: 13,46; IC del 95 por ciento: 63,98-66,63). Recidivaron 85 pacientes (21,4 por ciento), siendo la probabilidad de estar libre de recidiva a los 5 años del 63,47 por ciento. En el análisis bivariante se relacionaron con la recidiva de forma estadísticamente significativa los siguientes parámetros: edad (p = 0,007); clínica de complicación (perforación/oclusión) (p = 0,009); valores de GGT (p = 0,02); CEA alterado (p = 0,04); localización rectal (p = 0,03); cirugía urgente (p = 0,01); penetración transmural (p = 0,001); ganglios afectados (p = 0,0001) y absceso de pared (p = 0,0005). El resto de los parámetros no se relacionó con la recidiva. En el análisis multivariante permanecieron como factores predictivos independientes los siguientes: edad (p = 0,02; OR = 0,98); valores de GGT (p = 0,001; OR = 3,15); localización en recto (p = 0,04; OR = 2,03); absceso de pared (p = 0,04; OR = 1,9), y afecció ganglionar (p = 0,0001; OR = 4,72).Conclusiones. La identificación de una población de alto riesgo de recidiva tumoral podría permitir un seguimiento postoperatorio más selectivo con vistas a disminuir recursos y costes (AU)

Anastomotic stricture with the EEA-Stapler after colorectal anastomosis.

OBJECTIVE: Analysis of colorectal anastomosis stricture incidence after anterior resection of the rectum performed with the EEA-Stapler. To find out if differences existed in stricture incidence considering factors such, as age, neoplasia, postoperative radiotherapy, tumor stage and anastomic level. DESIGN: Longitudinal descriptive study. PATIENTS AND METHOD: 67 patients who underwent rectal anterior resection using the EEA-Stapler were evaluated. Data from sex, age, indication for operation, postoperative radiotherapy, tumor staging and anastomic level were recorded and compared with presence of stricture anastomosis. Stenosis was evaluated and graded as follows: grade O, no stenosis; grade I, no symptoms, X-ray or endoscopic finding; grade II, symptoms, the patients require ballon catheter dilation; and grade III, invalidant symptoms, the patients require surgery. RESULTS: Twelve patients (20%) were recorded as grade II and 3 patients (5%) as grade III. There were no statistically significant differences between prevalence of stricture and sex, age, neoplasic or non-neoplasic conditions, previous radiotherapy, level of anastomosis, and tumor stage. CONCLUSION: Stenosis after colorectal anastomosis with stapler devices must not be considered as an uncommon complication. In 20% of patients it may be a serious state that may require repeated catheter balloon dilations or surgery. Such condition is not dependent on diverse factors studied.

[Colorectal cancer in the context of multiple primary malignant neoplasms]. / El cáncer colorrectal en el contexto de las neoplasias primarias malignas múltiples.

INTRODUCTION: The finding of more than one carcinoma throughout the life of a patient is not unusual. In fact, the occurrence of multiple primary cancers has been well documented. However, the features, implications and prognosis of colorectal cancers associated to other malignancies, have been less well documented. PATIENTS AND METHOD: We report 29 patients with multiple primary neoplasms in which at least one of them was a colorectal cancer. Number, extension and localization of colonic neoplasm, and time interval between diagnosis of both neoplasms, and mortality and survival rate were retrospectively analyzed. Follow-up ranged from 1 to 360 months. RESULTS: Twenty three patients had two cancers, and six more than three neoplasms. Colorectal cancer was primary in 8 patients, metachronous in 14 patients and synchronous in 5 patients. Breast (7 cases), prostate (5 cases), endometrium (3 patients) and stomach (3 cases), were the other malignancies most frequently detected. Overall, 13 of the 27 patients (49%) died during the follow-up period after the recognition of the colorectal cancer. Two patients were lost to follow-up. Mantel-Haenszel test did not show significative differences in survival between patients with two or more neoplasms. There were no differences in survival depending upon the timing of diagnosis of the colorectal tumor. CONCLUSIONS: These data suggest that colorectal cancer may share etiologic factors with hormone-dependent neoplasms (breast, prostate) and gastric cancer. In patients with primary multiple neoplasms, the number of neoplasms detected and the interval time between both diagnosis are not prognostic factors when one of the malignancies in colorectal cancer.