The prevalence of iron deficiency among patients presenting with colorectal cancer

OBJECTIVES: To examine prospectively the prevalence of iron deficiency among new patients presenting with colorectal cancer and to compare transferrin saturation and serum ferritin as markers of iron deficiency in this group of patients. PATIENTS AND METHODS: Data were gathered on all patients presenting with a new diagnosis of colorectal cancer over a 12-month period. Iron status was estimated and, when possible, confirmed by measurement of serum ferritin concentration and transferrin saturation. Iron status was further examined in relation to tumour site and Dukes' stage. RESULTS: During the study 157 patients presented with a new colorectal cancer. Of these, 130 could be evaluated and 78[60%] had evidence of iron deficiency. Transferrin saturation was below the reference range in 55 patients, but serum ferritin was below in only 18 patients. Among the 49 patients with right-sided cancers, 39[80%] were iron deficient. Iron deficiency was significantly more likely in patients with right sided cancers compared with those with cancers at or distal to the splenic flexure (chi2 = 13, P < 0.001). CONCLUSION: The majority of patients with a new diagnosis of colorectal cancer are iron deficient at presentation. In such patients transferrin saturation measurement is a more sensitive marker of iron deficiency than serum ferritin. The potential role of measuring serum transferrin saturation as an adjunct to faecal occult blood screening should be explored further.     

Duration of symptoms and spread of colorectal cancer: a short history does not mean early disease

The 5-year survival rates for colorectal cancer are generally lower in the UK than other European countries. In an attempt to improve prognosis, central government has stipulated that patients with suspicious symptoms ought to be seen within 2 weeks of referral from a primary care physician. In order to evaluate whether symptom duration affects stage at presentation of colorectal cancer, a retrospective analysis of all patients presenting over a 2-year period to a large district general hospital was performed. There was no significant difference (P = 0.885) in Dukes' staging in patients with symptoms lasting less or more than 6 months. Though seeing patients with symptoms suspicious of colorectal cancer in specialist out-patient clinics within 2 weeks of presentation to the primary care physician would probably reduce the number of patients presenting as an emergency, it is unlikely to improve prognosis. Thus funds diverted towards the 2-week wait are probably best utilised for other procedures such as colonoscopy and for improving care once the diagnosis of cancer has been made. Diagnosis of colorectal cancer at an earlier stage is best achieved by screening of the population.     

How has the ‘two‐week wait’ rule affected the presentation of colorectal cancer?

OBJECTIVE: To assess the impact of the 'two-week wait' rule on the presentation of colorectal cancer. METHODS: A retrospective study of all patients referred to a fast-track clinic in a colorectal cancer centre over an 18-month period, documenting outcome, especially colorectal cancer diagnosis. Comparison was made with patients diagnosed with colorectal cancer presenting via other routes in the same time period. RESULTS: Over an 18-month period, 462 patients were seen in the fast-track clinic and 64 (13.8%) were diagnosed with colorectal cancer. A further 131 patients with colorectal cancer presented to the department in the same time period through other means; 66 via standard out-patient letters, 26 from other departments and 39 (20%) as emergency admissions. Median (range) time to first clinic was 12 (2-28) days for fast-track and 24 (1-118) days for standard referrals (P < 0.0001); median time to first treatment was a further 36 (9-134) and 36.5 (1-226) days, respectively. The fast-track cohort had more advanced staging than those referred by standard letter. There were 19 Dukes' B, 22 Dukes' C and 14 Dukes' D cancers in the fast-track group compared with 28 Dukes' B, 25 Dukes' C and 6 Dukes' D in the standard referral group. After patient interview, only 337 (73%) of 462 fast-track patients appeared to fulfil the referral criteria but of the 64 diagnosed with cancer, 59 (92%) satisfied the criteria. Of the 66 patients with cancer referred by standard letter, 61 (92%) fulfilled the criteria. CONCLUSION: Patients referred to the fast-track clinic were seen quicker than those referred by standard letter, but they tended to have more advanced disease. The fast-track referral criteria were fulfilled by most patients with cancer (whether or not they were referred to the fast track clinic), confirming their validity. After detailed interview in the clinic, a quarter of fast-track referrals were found not to satisfy referral criteria, suggesting that prioritization in primary care could be improved.     

Changing patterns of colorectal cancer in a regional teaching hospital.

A study was performed of all patients who underwent either surgery or endoscopic polypectomy for colorectal cancer in Oxford between 1979 and 1983 to ascertain whether any changes had occurred in presentation, diagnosis, treatment or outcome since a previous study from the same centre undertaken between 1966 and 1971. The average number of patients treated annually had increased from 52.8 to 103.4. The growth of the population of the health district was 6.7%. A similar proportion (25%) continued to present as emergencies. There was an improvement in the proportion of cases presenting with Dukes' A and B lesions (54% vs 41%) (P < 0.001), and a decrease in the number of patients with extranodal disease (from 30.4% to 13.3%) (P < 0.001). Only 13.9% of patients underwent colonoscopy before surgery and 14 patients were treated by endoscopic removal alone. In a health district with a slowly growing population, the number of patients presenting with colorectal cancer had increased dramatically over a decade. Colonoscopy has so far made little impact on making an earlier diagnosis. The increase in colorectal cancer has important implications for resource allocation.     

Iron-deficiency anaemia and delay in the diagnosis of colorectal cancer

Aims Iron‐deficiency anaemia (IDA) is a recognized complication of colorectal cancer (CRC) especially with right‐sided tumours, and failure to investigate the anaemia in older patients may lead to a delay in diagnosis. The aims of this study were to establish the proportion of patients with CRC shown to have an IDA for more than six months before diagnosis and to establish the proportion of patients with IDA who subsequently prove to have CRC. Methods All patients presenting with confirmed CRC in a health district (catchment population 280000) in the 4 years 1996–9 were identified from the pathology database after ethical approval. The criteria for IDA were haemoglobin (Hb) < 10.1 g/dl plus mean corpuscular volume < 78 fl and/or mean corpuscular Hb concentration < 32 g/dl. The haematology data‐base serving the same population was searched for evidence of: 1. IDA at diagnosis of CRC; 2. IDA more than 6 months and more than one year before the diagnosis of CRC; 3. The number of haematology referrals per annum in women over 55 and men over 50 years of age meeting the criteria for IDA. Results Of 440 patients with colorectal cancer, 166 (38%) had IDA at diagnosis and of the latter 54 (12%) were known to have IDA for more than six months before diagnosis and 26 (6%) had IDA more than one year before diagnosis. IDA was more common in right sided tumours (65%) than in those arising in the left side of the colon and rectum (26%). The annual incidence of IDA in the sampled population was 1366 in the stated age group. Conclusion The investigation of iron‐deficiency anaemia in older patients is important but in order to detect 26 patients with colorectal cancer a year earlier, the investigation of approximately 5000 patients would be required – a detection rate of less than 1%.     

The prognostic value of nucleolar organiser regions in colorectal cancer: a 5-year follow-up study.

Nucleolar organiser regions (AgNORs) are loops of ribosomal DNA which reflect the cellular activity or malignant potential of the cell and are identified by a specific staining technique. The purpose of this study was to assess the prognostic value of AgNORs in colorectal cancer and to compare it with other accepted prognostic methods. We studied 164 patients who were surgically staged for colorectal cancer and who had complete follow-up data available for 5 years. Using a highly specific silver staining and counting technique each patient was given an AgNOR score. There were 5 Dukes' C tumours, 108 were Dukes' B and 5 were Dukes' A. No cancer deaths occurred in patients with Dukes' A tumours. The incidence of well-differentiated, moderately-differentiated and poorly-differentiated tumours was 37.2%, 53.7% and 9.1%, respectively. Non-survivors had significantly higher AgNOR scores compared with survivors (mean value +/- SD, 14.2 +/- 0.9 vs 8.2 +/- 0.6, P less than 0.0001). In a regression analysis model AgNOR score was the most significant individual variable for predicting survival (chi 2 = 15, P less than 0.01) when compared with Dukes' classification, histological grade, tumour depth or vascular invasion.     

Acute-phase protein response, survival and tumour recurrence in patients with colorectal cancer

INTRODUCTION: An acute-phase protein response (APPR) has been associated with reduced crude survival rates and increased recurrence following apparently curative resection in patients with colorectal cancer. This study investigated the prognostic significance of a preoperative and postoperative APPR in relation to disease-specific mortality rate. METHODS: Some 202 patients with colorectal cancer were followed for at least 5 years. C-reactive protein concentration, measured before and at 3 months after operation, was used as an index of the APPR. Univariate and multivariate analyses were performed on a number of potential prognostic factors. RESULTS: Thirty-six per cent of patients had an APPR and this was associated with a higher rate of local tumour invasion, fewer curative resections and a higher carcinoembryonic antigen (CEA) concentration. There was no difference in Dukes' stage between patients with or without an APPR. The most important prognostic factor related to both disease-specific and crude survival was Duke's stage (P < 0.0001). Subgroup analysis demonstrated that APPR had prognostic significance only in patients with advanced disease (P = 0.013). An APPR was present in a minority of patients (11 per cent) after operation and was not associated with increased likelihood of tumour recurrence. CONCLUSION: The APPR is increased in more than a third of patients presenting with colorectal cancer and is associated with more frequent local tumour invasion, fewer curative resections and a higher CEA level. An APPR at 3 months after operation does not have the prognostic significance reported by earlier studies.     

Local recurrence after radical surgery for colorectal cancer

Local recurrence of colorectal cancer after curative surgery is a major clinical problem. The aim of our study was to present our experience in this field. Between January 1990 and December 2000, 572 patients underwent resection for colorectal cancer in our department; 66 of them had local recurrence within the first 2 years. Most of those patients had Dukes' stage B (n = 24) or stage C (n = 37) tumors, which were located mainly in the rectum (n = 40) and sigmoid colon (n = 18). The incidence of local recurrence was 11% and 15.9% for tumors that were Dukes' stages B and C, respectively. Thirty-five of 66 patients received palliative treatment, and 28 of them died within 9 months. The remaining 31 patients underwent radical excision of the recurrent tumor: 11 of these patients died within 2 years, and 20 were still alive after 30 months. The only hope for long-term survival for patients presenting with local recurrence from colorectal cancer after primary radical treatment is to identify local recurrence at an early stage and treat it in a radical manner.     

Colorectal cancer in South Tyneside: a personal audit.

A detailed retrospective audit over a 10-year period of 162 patients presenting with colorectal cancer to one surgeon without a specific interest in colorectal surgery was undertaken. The modes of presentation and potential sources of delay in treatment were analysed. The 30-day hospital mortality rate was 10.1% and most of these patients had incurable disease. The crude survival 1 year after the study was 34% (mean follow-up 6 years); at 3 years after study, it was the same (8 years mean follow-up). Survival data are also given in relation to the presentation, and operative and histological findings. Preoperative radiotherapy for rectal cancers is discussed and assessed. During the last 3 years, 49 further patients have presented. The operative mortality rate was 8%. It is suggested that improved surgical treatment, particularly in the acute case, has been responsible for the reduced mortality rate. Only 51% of patients underwent operation with a reasonable prospect of cure. It is suggested that no improvement in the prospects of patients with colorectal cancer will occur until pre-symptom diagnosis from screening techniques is more commonly available.     

Screening practices of patients presenting for resection of a colorectal neoplasm

Multiple studies demonstrate the efficacy of colorectal cancer (CRC) screening in patients over 50 years of age. However, there is a lack of consensus regarding which screening method to use, and compliance has been poor. The objective of this study was to identify the CRC screening practices at two institutions and determine the relationship between screening and pathologic stage for patients presenting with a colorectal neoplasm. This study, conducted at the University of Virginia (UVA) Health System and the Salem Veterans Affairs Medical Center (VAMC) between October 30, 2000, and September 1, 2004, included 198 patients > or = 50 years who presented for resection of a primary colorectal neoplasm. Pathologic stage and prior screening were identified retrospectively through chart review and patient response to an anonymous survey. Prior screening was demonstrated in 71 per cent of patients. Colonoscopy was the most commonly used modality. There was a higher percentage of CRC screening at VAMC compared with UVA (80% vs 62%, P < 0.0008). Patients at UVA were more likely screened with colonoscopy, whereas fecal occult blood testing (FOBT) was most common at VAMC (P < 0.0001). Prior CRC screening and cancer stage were inversely related. Ninety-one per cent of patients with benign polyps had been screened prior to diagnosis, compared with 72 per cent of patients with stage I and II cancer and 54 per cent of patients with stage III and IV cancer (P < 0.05). Of patients presenting for surgery, 71 per cent underwent CRC screening. Variability exists in the methods employed for CRC screening. CRC screening facilitates diagnosis at an early stage.     

Early diagnosis in colorectal cancer. Still no benefit?

Time to diagnosis from first symptoms has been assessed in 332 patients with colorectal cancer treated by the three general surgeons in Trafford Health Authority. When compared with other series, delay to diagnosis has been shortened, particular general practitioner and hospital delay. Fewer patients presented as emergencies and a greater proportion of patients had early-stage disease. There was no significant difference in delay times between Dukes' stage B and C patients but there was a significant difference in survival at 2 years between these two stages. Delay times for patients with risk factors, family histories or diverticular disease were not significantly different from times in patients without these factors.     

Reported symptoms, diagnostic delay and stage of colorectal cancer: a population-based study in Denmark

OBJECTIVE: The primary prognostic factor for colorectal cancer (CRC) is stage. Any association between symptoms, diagnostic delay and stage may have implications for the clinical course of the disease. We examined the association between symptoms and diagnostic delay and between symptoms and stage, and assessed whether the associations differed for colon cancer (CC) and rectal cancer (RC). PATIENTS AND METHODS: Population-based prospective observational study based on 733 Danish CRC patients. Diagnostic delay and patients' reported symptoms were determined through questionnaire-interviews. Dukes' stage was obtained from medical records and pathology forms. Diagnostic delay was categorized into three delay groups: < or = 60, 61-150 and > 150 days. Stage was classified into nonadvanced (Dukes' A and B) or advanced (Dukes' C and D) cancers. We calculated the frequency of the most frequently reported initial symptom or symptom complex for CC and RC patients, and evaluated the frequency of patients with different initial symptoms/symptom complexes in the three delay groups. For the most frequent initial symptoms/symptom complexes, we calculated the frequencies according to stage, and estimated the relative risk of having an advanced stage, with 95% confidence intervals. RESULTS: The most frequent initial symptoms/symptom complexes were very vague symptoms for CC and rectal bleeding for RC. For both CC and RC, rectal bleeding was significantly associated with nonadvanced stage. The relative risk of having an advanced cancer was 0.6 for monosymptomatic rectal bleeding and 0.7 for rectal bleeding combined with other symptoms. CONCLUSIONS: Initial symptoms of CC were often very vague, making it difficult to identify a precise start date. The most frequent initial symptom/symptom complex for RC - rectal bleeding - was better defined. Rectal bleeding was significantly associated with nonadvanced CC and RC and a significantly decreased relative risk of having an advanced cancer.     

Colorectal carcinoma in the young. Prognostic factors

Colorectal carcinoma is the third most frequently diagnosed malignant neoplasm. Usually patients affected by this neoplasia belong to VI decade of life. However approximately 2-8% of tumors arise in patients with age under 40 years. Aim of the study was to analyse the results of surgical treatment of colorectal cancer in patients aged under forty. From January 1987 to December 2002, 46 patients under forty years with colorectal cancer underwent surgical procedure. No perioperative mortality was registered, and complications were evidenced in nine patients (20%). Actuarial five years survival was 33%, and overall mean survival was 53 months. Univariate and multivariate analyses identified as prognostic factors the tumor grade, Dukes' stage, nodal status, and length of symptoms.     

Prognosis in Duke's B colorectal carcinoma: the Jass classification revisited.

OBJECTIVE: To assess whether Jass staging enhances prognostic prediction in Dukes' B colorectal carcinoma. DESIGN: A historical cohort observational study. SETTING: A university tertiary care centre, Switzerland. SUBJECTS: 108 consecutive patients. INTERVENTIONS: Curative resection of Dukes' B colorectal carcinoma between January 1985 and December 1988, Patients with familial adenomatous polyposis; hereditary non-polyposis colorectal cancer; Crohns' disease; ulcerative colitis and synchronous and recurrent tumours were excluded. A comparable group of 155 consecutive patients with Dukes' C carcinoma were included for reference purposes. MAIN OUTCOME MEASURES: Disease free and overall survival for Dukes' B and overall survival for Dukes' C tumours. RESULTS: Dukes' B tumours in Jass group III or with an infiltrated margin had a significantly worse disease-free survival (p = 0.001 and 0.0001, respectively) and those with infiltrated margins had a significantly worse overall survival (p = 0.002). Overall survival among those with Dukes' B Jass III and Dukes' B with infiltrated margins was no better than overall survival among all patients with Dukes' C tumours. CONCLUSION: Jass staging and the nature of the margin of invasion allow patients undergoing curative surgery for Dukes' B colorectal carcinoma to be separated into prognostic groups. A group of patients with Dukes' B tumours whose prognosis is inseparable from those with Dukes' C tumours can be identified, the nature of the margin of invasion being used to classify a larger number of patients.     

Prognostic power of Doppler perfusion index in colorectal cancer. Correlation with survival.

OBJECTIVE: The authors assessed the relative value of Dukes' staging and Doppler perfusion index (DPI) as prognostic indices of early death in patients with colorectal cancer who have had apparently curative surgery. SUMMARY BACKGROUND DATA: Dukes' classification is recognized as the gold standard against which other prognostic factors should be compared; however, its discriminatory power is limited. No established method exists for accurately identifying patients who, despite having undergone an apparently curative resection for colorectal carcinoma, are at high risk. METHODS: Eighty consecutive patients undergoing apparently curative surgery for colorectal cancer were staged using Dukes' classification. In addition, DPI (ratio of hepatic arterial to total liver blood flow) was measured before surgery by means of a duplex/color Doppler sonography. RESULTS: On the basis of 2-year follow-up data, Dukes' classification failed to define clearly those patients who died or who developed recurrent disease. In contrast, DPI clearly identified two groups of patients: 78% of patients with an abnormally elevated DPI value developed recurrent disease or died, whereas 97% of patients with a normal DPI value survived. CONCLUSION: This technique can identify patients at high risk for colorectal cancer, after having undergone apparently curative resection, who would be suitable for adjuvant therapy.     

Iron deficiency anaemia and delayed diagnosis of colorectal cancer: a retrospective cohort study

Aim The extent to which different referral pathways following a primary care diagnosis of iron deficiency anaemia (IDA) are associated with delay in diagnosis of colorectal cancer (CRC) was determined. Method Eligible patients aged 40 or more years, with IDA diagnosed in primary care, and a subsequent diagnosis of CRC, were studied retrospectively. Referral pathways were identified using the specialty of first recorded GP referral following IDA diagnosis. Differences in time to diagnosis of CRC were assessed by referral specialty. Differences in the proportion of cases referred before and after the re‐issue of the NICE urgent referral guidelines for suspected lower gastrointestinal (GI) cancer were also assessed. Results Of 628 882 eligible patients, 3.1% (n = 19 349) were diagnosed with IDA during the study period; 3.0% (n = 578) were subsequently diagnosed with CRC. Two hundred and fifty‐nine (44.8%) patients had no recorded referral or a referral unrelated to anaemia or the GI tract. Only 35% (n = 201) of patients were referred to a relevant specialty. Median time to CRC diagnosis ranged from 2.5 months (referral to a relevant surgical specialty) to 31.9 months (haematology). Time to diagnosis was longer in patients referred to a medical compared with a relevant surgical specialty (P = 0.024). There was no significant difference in time to CRC diagnosis before and after the NICE guidelines were re‐issued in 2005. Conclusion Significant differences exist between referral specialties in time to CRC diagnosis following a primary care diagnosis of IDA. Despite NICE referral recommendations, a significant proportion of patients are still not managed within recommended care pathways to CRC diagnosis.     

Preoperative anemia in colon cancer: assessment of risk factors

Anemia is common in cancer patients and is associated with reduced survival. Recent studies document that treatment of anemia with blood transfusion in cancer patients is associated with increased infection risk, tumor recurrence, and mortality. We therefore investigated the incidence of preoperative anemia in colorectal cancer and assessed risk factors for anemia. Prospective data were collected on 311 patients diagnosed with colorectal cancer over a 6-year period from 1994 through 1999. Patients were stratified by age, gender, presenting complaint, preoperative hematocrit, American Joint Committee on Cancer (AJCC) stage, and TNM classification. Discrete variables were compared using Pearson's Chi-square analysis. Continuous variables were compared using Student's t test. Differences were considered significant when P < 0.05. The mean age of the study cohort was 67 +/- 9.2 with 98 per cent of the study population being male. The mean AJCC stage was 2.2 +/- 1.2 and the mean preoperative hematocrit was 35 +/- 7.9 with an incidence of 46.1 per cent. The most common presenting complaints were hematochezia (n = 59), anemia (n = 51), heme-occult-positive stool (n = 33), bowel obstruction (n = 26), abdominal pain (n = 21), and palpable mass (n = 13). Preoperative anemia was most common in patients with right colon cancer with an incidence of 57.6 per cent followed by left colon cancer (42.2%) and rectal cancer (29.8%). Patients with right colon cancer had significantly lower preoperative hematocrits compared with left colon cancer (33 +/- 8.5 vs 36 +/- 7.4; P < 0.01) and rectal cancer (33 +/- 8.5 vs 38 +/- 6.0; P < 0.0001). Patients with right colon cancer also had significantly increased stage at presentation compared with left colon cancer (2.3 +/- 1.3 vs 2.1 +/- 1.2; P < 0.02). Age was not a significant risk factor for preoperative anemia in colorectal cancer. We conclude that there is a high incidence of anemia in patients with colon cancer. Patients with right colon cancer had significantly lower preoperative hematocrits and higher stage of cancer at diagnosis. Complete colon evaluation with colonoscopy is warranted in patients with anemia to improve earlier diagnosis of right colon cancer. A clinical trial of preoperative treatment of anemic colorectal cancer patients with recombinant human erythropoietin is warranted.     

Colorectal cancer. A blueprint for disease control through screening by primary care physicians

The Virginia Colorectal Cancer Control Project is a statewide effort to reduce morbidity and mortality from colorectal cancer by stimulating the adoption of screening and early detection practices by primary care physicians. The project emphasizes use of the three-day fecal occult blood test, digital rectal examination, and endoscopy. Recruitment strategies included personal contact, newsletters, journal articles, and screening workshops. Of the 33 318 patients screened over 26 months, positive fecal occult blood test reactions were recorded in 3.3% of asymptomatic patients and in 14.8% of symptomatic patients. Polyps were found in 149 and cancer was diagnosed in 94 patients of whom one third were asymptomatic. Eighty percent of the latter had Dukes' A and B lesions, 12% had Dukes' C lesions, and 8% had Dukes' D lesions. In contrast, only 36% of the symptomatic cancers were Dukes' A and B lesions, and 69% were Dukes' C and D lesions. These results suggest that primary care physicians can be effective in the screening and detection of precancerous polyps and early-staged colorectal cancers.     

Factors affecting survival in patients presenting with synchronous hepatic metastases from colorectal cancer: a clinical and computer analysis

A review of 90 patients presenting in Leeds over the period 1976-80 with synchronous hepatic metastases from colorectal cancer has been undertaken. The median survival for the group was 10.3 months (range 1-48 months). A multivariate analysis identified factors in the history and examination of the patient, results of biochemical and haematological investigations and characteristics of both the primary and secondary tumour, which influenced survival. Recognition of these factors allowed both a clinical scoring index and computer survival model to be constructed. When applied to the group as a whole both were reasonably accurate at predicting survival (62 per cent for the survival model and 66 per cent for the scoring index). However, the computer model correctly identified 80 per cent of the 'long-term' survivors (greater than 12 months from the time of diagnosis). It is concluded that many trials of therapy for hepatic metastases from colorectal cancer have failed to consider the natural history of the disease and the identification of prognostic indices should allow for adequate stratification within the treatment arms of any future study.