Diagnostic Ultrasound in Colorectal Cancer

SUMMARYBackground and purpose Colorectal cancer is a common disease in Denmark with considerable morbidity and mortality. Although survival in recent years has improved, Denmark still has the lowest 5-year survival compared to the other Nordic countries. The treatment of patients depends on local tumour stage and whether there is distant spread of the disease at the time of diagnosis. Ultrasound Diagnostics is non-unionized and less costly than MDCT, MRI and PET/CT. Although MRI in recent years has gained ground in diagnostics in Denmark, there are still patients who have contraindications to MRI or are non-compliant. Also, patient movement during the MRI sequence acquisition degrades image quality on MRI and decreases accuracy. Over the last years, the ultrasound technique has improved further by Power Doppler, intravenous contrast enhancement, and elastography. Ultrasound diagnostics has the potential to contribute to the staging of colorectal cancer. The purpose of these studies was to determine the usefulness of ultrasound diagnostics in patients with colorectal cancer.The purpose of the TRUS studies was to compare staging of rectal carcinomas using digital rectal exploration with the resulting pathological examination in relation to differentiating benign from malignant polyps and determining tumour stage and lymph node status. In this context we also performed an observer comparison using both TRUS and MRI. Consistency of tumour outgrowth of rectal cancer rated by TRUS and MRI was compared. We investigated the correlation between the incidence of distant metastases and tumour outgrowth measured by ultrasound. In the elastography study, we compared the quantitative ultrasound elastography measurements of tumour response to chemoradiation with the pathological T-stage and TRG response.The diagnostic accuracy of ultrasound for the detection of liver metastases was evaluated. Both preoperative ultrasound and intraoperative ultrasound were evaluated in a prospective, blinded setting. We verified the ability of colour Doppler, power Doppler, and contrast-enhanced power Doppler to detect signals in the liver metastases. In addition, we prospectively compared contrast-enhanced ultrasound with CT scan in the detection of liver metastases.Results By transrectal ultrasound of polyps using the new AWS technique, a sensitivity of 96% and a specificity of 88% was found for cancer, whereas digital exploration had a significantly lower sensitivity and a specificity of 62% for malignant invasion of rectal polyps. The degree of growth through the rectal wall evaluated by digital exploration was correct in 73% of the cases compared to 90% by transrectal ultrasound (p <0.005). We found no difference in T-stage classification between ultrasound and MRI. In other studies overestimation of wall outgrowth was more frequent than understaging. We found increased over-and underestimation by the less experienced observer. By comparing transrectal ultrasound and MRI we observed good agreement in the tumour outgrowth measurements from the intestinal wall, with a kappa value of 0.93 in classifying early versus advanced tumours. Six percent of patients with tumour outgrowth < 5 mm on ultrasound had distant metastases, whereas there were distant metastases in 46% of patients with more than 5 mm outgrowth (p <0.00001).It is difficult to diagnose lymph node metastases by TRUS. The first study demonstrated a sensitivity and specificity of 58% and 77%, respectively. The sensitivity was even lower in the later study.By quantitative ultrasound elastography of patients with rectal cancer after 2 weeks' neoadjuvant therapy, we detected a stiffness of 1.95 m/s of tumours limited to the rectal wall at the final pathological examination. Tumours with outgrowth after chemoradiation and subsequent surgery had a hardness of 2.47 m/s (p <0.05). Tumours with good TRG response tended to be softer than the TRG 3-4 tumours.In our study of 295 patients with primary colorectal cancer we found a sensitivity of preoperative ultrasound, surgical exploration, and intraoperative ultrasound of 70%, 84%, and 97%, respectively, based on a patient-by-patient comparison (p <0.05). Finally, in a study of 271 patients we observed an equal sensitivity of contrast-enhanced preoperative ultrasound and CT of 86%. The positive predictive value tended to be higher for contrast-enhanced ultrasound than CT. Local recurrence was found in 66% of patients who had liver metastases with power Doppler signal in central tumour vessels, whereas none of the patients with central vessels within the liver metastases had local recurrence.ConclusionTransrectal ultrasound was more accurate than digital rectal exploration, both in terms of early stage determination, intestinal wall outgrowth, and detection of lymph node metastases. Experience has impact on performance. There was no difference between TRUS and MRI in the T-stage determination. We found a good correlation between TRUS and MRI in determining the extent of tumour outgrowth. Ultrasound is not optimal for lymph node diagnosis and cannot stand alone, not even with early tumours, when local surgery/TEM should be considered. The low lymph node sensitivity could lead to relapse after surgery.TRUS can be used for T-staging of early rectal tumours and in patients with contraindications to or poor compliance with MRI scan.Quantitative elastography showed promising results in detecting early response to neoadjuvant chemoradiation of advanced rectal cancer.IOUS is a safe method with a significantly higher sensitivity in the detection of liver metastases than preoperative ultrasound and surgical palpation. Patients with liver metastases, which harboured power Doppler signal centrally, more often had advanced extrahepatic disease. CEUS of the liver had sensitivity comparable to that of contrast enhanced multi-slice CT scan. There was a tendency towards fewer false positive results using CEUS compared to CT. CEUS of the liver is a safe method to apply if CT/MRI contrast study cannot be performed because of renal impairment. More patients could avoid unnecessary liver surgery with improved liver diagnostics. It may also be possible to identify more candidates for liver surgery or other treatment methods.The high solubility of ultrasound and new technological developments give ultrasound a place in the diagnosis of rectal cancer, especially in early tumours. Screening for colorectal cancer will give rise to the detection of a number of early tumours. Contrast-enhanced liver ultrasound and intraoperative ultrasound has additional space in the detection of liver metastases from colorectal cancer.