Solitary Pulmonary Nodules: Dynamic Enhanced Multi-Row Detector CT Evaluation And Clinical Application

BackgroundSometimes,information on the morphologic features and attenuation values of a solitary pulmonary nodule on unenhanced computed tomographic(CT) images may be insufficient to allow the differentiation of benign and malignant nodules.Up to 52%of surgery performed for indeterminate lung nodules results in resection of a benign nodule. The total inpatient and outpatient cost to remove a lung nodule by means of thoracotomy on thoracoscopy ranges from $18,000 to $30,000.This is an expensive and invasive way to obtain a benign diagnosis.Accurate prediction of a benign lesion that need not be resected could lead to considerable savings in the management of indeterminate lung nodules.Identification of an operable malignant lesion could result in prompt resection.Over the past decade,there has been considerable research interest in the enhancement of indeterminate lung nodules with various imaging methods.The hypothesis in most studies was that malignant lung nodules enhance substantially more than benign nodules.This hypothesis has been corroborated with varying degrees of statistical significance by using six imaging techniques:conventional tomography,computed tomography(CT),magnetic resonance(MR) imaging,fluorine-18 fluonodeoxyglucose positron emission tomography, and angiography.These scientific efforts have all shown similar results,perhaps because these imaging methods enable measurement of lung nodule characteristics that are interrelated.The nodule characteristics may include the type and amount of vasculature and metabolic activity.The evaluation of tumor vascularity with contrast material-enhanced CT has proved to be useful in the differentiation of malignant and benign nodules.In previous dynamic CT studies in which a conventional or single-detector row helical CT machine was used,however,investigators acquired a single scan or a limited number of scans through the whole nodule at specific times for the duration of the dynamic study.In a previous study,we discussed the differential enhancement of pulmonary nodules after the intravenous administration of iodinated contrast material.We have continued the evaluation with spinal CT.Our hypotheses are that the degree of lung nodule enhancement is directly related to the likelihood of malignancy and to nodule vascularity and improve accuracy of diagnosis.With the advent of multi-row detector CT(MDCT),we have the advantage of shorter acquisition times,greater coverage,and superior image resolution along the z-axis.Image clusters obtained at a given time throughout a nodule can be acquired sequentially by using a helical technique at short time intervals after the intravenous injection of contrast medium,thus allowing the same or very similar scans to be obtained through the nodule at various times to compare the extent of enhancement.A growing malignant nodule needs its own blood supply from adjacent tissues,which is essentially required for tumor growth and metastatic spread.This process may be caused by the increased extent of microvessel density leads to increased perfusion and permeability of the capillaries and is associated frequently with strong enhancement of a malignant nodule at CT.Therefore,the extent of enhancement can be interpreted as reflecting tumor vascularity,which may aid in the differentiation of malignancy and benignancy.The purpose of this study was to evaluate dynamic enhancement of solitary pulmonary nodules by using multi-row detector CT and to correlate the results with pathologic specimens.Objective1.To correlate incremental dynamic computed tomographic(CT) and pathologic findings in malignant SPNs.Dynamic attenuation was compared with pathologic type of lung cancer and with distribution of fibers in the pathologic specimen:2.To determine if the degree of enhancement of pulmonary nodules is directly related to the likelihood of malignancy and optimize protocol of CT scan.3.To explore the possibility of kinetic CT parameters to predict the prognosis of lung cancer in vivo by dynamic contrast enhanced CT imaging.4.To introduces the definition and expression of the computed tomography dose index(CTDI),Dose Length Product(DLP),the application in the clinical medicine and medicinal equipments testing.MethodsLung lesions smaller than 3 cm in diameter were evaluated in 63 patients.All examinations were performed with the same scanner(Toshiba Aquilion 16 medical systems,Tokyo,Japan).Scan protocol:Before dynamic CT was performed,we obtained thin-section helical CT scans(2.5mm collimation,0.8 second gantry rotation time,120kVp,200mA) through the nodule.Before the intravenous injection of contrast medium,a series of 4-10 images were obtained throughout the nodule for 30 or 40 mm(cover the whole nodule) along the z-axis,with 2.5 mm collimation,120kVp,200mA,and 0.4-second gantry rotation time. Thereafter,an additional 13 series of images were obtained at 20,30,45,60,75,90,120 seconds and 3,5,9,12,15 and 20 minutes after the injection of(Iomeprol 370;Bracco, Milan,Italy),which was administered at a rate of 4mL/sec for a total of 420 mg I/kg body weight by using a power injector(MCT Plus;Medrad,Pittsburgh,Pa).Image data of nodules were reconstructed by standard algorithm as follows:as for nodules between 1 to 2 cm in longest diameter,image data were reconstructed with a thickness of 2 mm;as for nodules >2 cm in longest diameter,image data were reconstructed at a thickness of 5 mm. The dynamic imaging at 90 seconds helical was obtained from the lung apex to the level of the middle pole of both kidneys for tumor staging.All thin-section,dynamic and staging CT data were interfaced directly to our picture archiving and communication system (TOSHIBA vitra 2).On the monitors,both mediastinal window(width,300 HU;level 40 HU) and lung window(width,1500 HU;level -500 HU) were viewed.CT values of the inner area of the nodule at plain and dynamic enhanced CT were measured on 3～4 Sequential axial CT sections(2.0-mm-thick,2.0-mm interval or 5-mm-thick,5.0-mm interval) through the SPNs were obtained.Serial scans were obtained during breath holding(middle expiration) to reduce motion artifact.Standard(soft-tissue) reconstruction algorithm(window width,300HU;window level,40HU) was used for dynamic scanning. The bone algorithm(window width,1,500 HU;window level,-500 HU) was also used for evaluation of the morphologic features of the lesions. Fourteen serial thin-section CT scans were obtained.The CT attenuation(in Hounsfield units) of SPNs were calculated for the three～four sections in which the maximum diameter of the lung lesion was greater than 70%of that in the central section to prevent partial volume averaging effect of the surrounding lung and peripheral enhancement of the nodule from influencing analysis of CT attenuations.A circular region of interest for the calculation of CT attenuation was established independently in each section,with the diameter as large as possible(more than 60%of the minimum diameter of the lung lesion in each section).1.Calculations were as follows:A parameters of early enhancement such as Peak height(PH),net enhancement,and slope-in of enhancement;B parameters of late enhancement and washout of adenocarcinomas such as 20 minutes' washout,20minutes'%washout,and 20 minutes' slope-out of enhancement;C threshold values were retrospectively used to define four different types of time-intensity curves,and calculate numbers of each histologic typing.Adenocarcinoma specimens were stained with hematoxylin-eosin stain and collagen and elastica double stain.To analyze correlation among PH,net enhancement,slope-in,slope-out and grade of fibers.Degeneration of tumor cells were also compared with the above parameters.2.Access SPNs using DCE-CT:compare the differences of enhancement,PH,time to peak(T_peak) of each time point,20 minutes' washout,20minutes'%washout,and 20 minutes' slope-out of enhancement between benign and malignant nodules.To analyze morphologic features and synthetically evaluate the clinical application of dynamic enhanced CT on differential diagnosis of SPNs.3.Four types of time-density curve(T-DC) were compared.To calculate the following DEC-CT parameters:(1) peak enhancement ratio(2) time to peak(T_peak)(3) the initial slope(Slope_i)(4) the washout ratio(W_peak-1200).Correlation between the above parameters and detected biomarkers(Ki-67,c-erbB2,CEA,and p53),were statistically analyzed with the spearman rank correlation.4.Radiation doses were measured by Toshiba Aquilion 16 CT scanner which was tested by thermoluminescent dosimetry.DPL and extent of DCE-CT scan were recorded and effective dose on nodule and lungs were calculated,respectively.The measured organ dose was compared with that of standard thoracic helical CT.Results1.Spearman test was made to analyze correlation between dynamic parameters and fiber in interstitium in tumors.There are significant difference between 20 minutes' washout(χ²=7.1637,P=0.0278<0.05),maximum relative washout ratio(χ²=6.505, P=0.039<0.05),slope of washout ratio(χ²=7.855,P=0.020<0.05) and fiber in tumor interstitium.Net enhancement(χ²=7.054,P=0.029),maximum relative enhancement ratio (χ²=6.640,P=0.036),20 minutes' wahout(χ²=6.765,P=0.034),20 minutes washout ratio (χ²=6.765,P=0.034).2.There are significant differences between 120s CT attenuation,20 minutes' washout,20 minutes' slope ratio and character of the nodule.There are significant differences with the above parameters andâ…¡,â…¢type curve.The results of receiver operator characteristic(ROC) analysis for differentiation of benign and malignant SPN showed that 120s CT attenuation,20 minutes' washout,20 minutes' slope ratio had 64.3% sensitivity and 84.2%specificity for identification of malignant tumors. Morphologic features such as round-like,triangle-like,multi-angle,spiculation,light lobulation,the degree of edge(sharp,clear,blur),Vessel convergence sign,vacuole sign, airing of bronchi,cut-off of the bronchi,depression of pleura,and so on.The results of ROC analysis for differentiation of benign and malignant SPN showed that the above morphologic features had 92.9%sensitivity and 100%specificity for identification of malignant tumors.The results of ROC analysis for differentiation of benign and malignant SPN showed that combination of morphologic features and DCE parameters had 95.2% sensitivity and 100%specificity for identification of malignant tumors.3.The type of T-DC curve correlated positively with the expression of p53(r=0.419, P=0.047).There is no significant correlation with CEA,Ki67,c-erbB2 and T-DC curve.4.The measured total organ effective dose at thinsection,dynamic,and staging CT ranged from 14.2996 mSv or 16.065 mSv at the nodule sites and ranged from 19.8388 mSv～22.7358 mSv in the lungs.The organ dose administered for a standard enhanced thoracic helical CT examination at our institution ranges from 4 to 25.5 mSv.Conclusions1.Enhancement characteristics of lung cancer reflect the distribution of fibers in the tumoral intersfitium and the degeneration of adenocarcinomas.Some dynamic parameters of DCE CT are useful in evaluating the prognosis and drawing up the treatment plan of the adenocarcinoma in vivo.2.CT analyses of solitary pulmonary nodules have been performed with morphologic criteria.Combination morphologic feature with DCE-CT characterizations improve accuracy of diagnosis.3.Some dynamic parameters of DCE CT are useful in evaluating the prognosis and drawing up the treatment plan of the lung adenocarcinoma in vivo.4.DCE-CT scan range were diminished,however,further reduction of the dose to that used at chest radiography was associated with a significant decrease in dynamic contrast CT of SPN.