| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, University of Ryukyus School of Medicine, 207 Uehara,
Nishihara-cho, Okinawa Prefecture 903-0215, Japan.
2 Diagnostic Radiology Division, National Cancer Center Hospital, 1-1 Tsukiji
5-chome, Chuou-ku, Tokyo 104-0045, Japan.
3 Clinical Laboratory Division, National Cancer Center Hospital, Tokyo 104-0045,
Japan.
4 Thoracic Surgery Division, National Cancer Center Hospital, Tokyo 104-0045,
Japan.
5 Department of Radiology, Kurume University School of Medicine, 67-Asahimachi,
Kurume 830-0011, Japan.
6 Pathology Division, National Cancer Center Research Institute, 1-1 Tsukiji
5-chome, Chuou-ku, Tokyo 104-0045, Japan.
Received December 20, 2002;
accepted after revision July 10, 2003.
Address correspondence to M. Kusumoto.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. The cases of all patients who underwent surgical resection for primary lung cancer in a single institution from 1993 to 2000 and who received an initial diagnosis of poorly differentiated non–small cell lung carcinoma, small cell carcinoma, carcinoid tumor, and large cell neuroendocrine carcinoma were histologically reviewed. The findings for 43 patients were histologically reclassified and confirmed as large cell neuroendocrine carcinoma. The CT scans available for 38 patients were evaluated by two observers.
RESULTS. In the 38 patients, six central tumors and 32 peripheral tumors, with diameters ranging from 12 to 92 mm (mean ± SD, 32 ± 19 mm), were identified. None of the tumors had air bronchograms or calcification in the mass or nodule. Of the 19 patients with thin-section CT scans, 14 (74%) showed the tumor–lung interface as well defined and five (26%) showed the interface to be ill defined. Lobulation was identified on 15 scans (79%) and spiculation was evident on six scans (32%). On contrast-enhanced CT scans, inhomogeneously enhanced tumors appeared to be larger (51 ± 18 mm) than homogeneously enhanced tumors (25 ± 10 mm; p < 0.001). At histopathologic examination, gross necrosis was noted in 20 of 28 patients who had undergone contrast-enhanced CT, and the cause of inhomogeneous enhancement on CT scans was determined to be intratumoral necrosis. Multiple microscopic necroses were present in all 28 patients.
CONCLUSION. Large cell neuroendocrine carcinoma usually appears as a well-defined and lobulated tumor with no air bronchograms or calcification. The inhomogeneous enhancement (caused by necrosis) seen in large cell neuroendocrine carcinomas with large diameters is not necessarily apparent in small-diameter (< 33 mm) large cell neuroendocrine carcinomas, even if the tumor contains necrosis.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In 87 of the 572 patients, the carcinomas were reclassified and confirmed as large cell neuroendocrine carcinomas. CT scans were available for review in 34 of the 38 patients who underwent surgery during the most recent 7 years of the study period (1993–1999). In addition, we identified four patients with a histologically determined diagnosis of large cell neuroendocrine carcinoma who had undergone CT in 2000. Therefore, our study population consisted of 38 patients, 36 men and two women, whose ages ranged from 45 to 82 years (mean, 66 years). Of the 38 patients, 20 had lesions classified as large cell neuroendocrine carcinomas at the initial diagnosis. The initial pathologic diagnoses of the lesions in the remaining 18 patients were as follows: eight small cell carcinomas (six intermediate cell type), seven poorly differentiated adenocarcinomas, two large cell carcinomas, and one poorly differentiated squamous cell carcinoma. Thirty-one of 38 patients who presented without pulmonary symptoms had abnormal findings on chest radiographs obtained during a routine examination. Symptoms in the remaining seven patients included low-grade fever (n = 1), chest and back pain (n = 1), hemosputum (n = 3), and cough (n = 2). Thirty-one patients had been smokers for between 12 and 200 pack-years (mean, 62 pack-years). Six patients had quit smoking 3–17 years (mean, 7 years) before diagnosis. In one patient, a smoking history could not be obtained. Surgical procedures included a partial resection of the mass in four patients, segmentectomy in one patient, lobectomy in 27 patients, and pneumonectomy with regional lymph node dissection in six patients.
CT scans were obtained with X-Vigor or TCT-900S units (Toshiba, Tokyo, Japan). The helical technique (collimation, 10 mm; pitch, 1.5) was used and covered the area from the lung apices to the diaphragm. In 28 patients, nonionic iodinated contrast material (Iopamiron [iopamidol 300 mg I/mL], Nihon Schering, Osaka, Japan) was administered IV at a rate of 1–2 mL/sec. Contrast-enhanced CT scans were obtained within 120 sec (range, 70–120 sec) after the onset of contrast material infusion. Unenhanced scans were obtained in 20 of 28 patients who underwent contrast-enhanced CT. In 19 of 38 patients, additional thin-section (collimation, 2.0 mm; pitch, 1.0) scans were obtained at the level of the lesion. All CT scans were obtained with 120 kVp and 200 mA. The scans were viewed on standard mediastinal window setting (window level, 60 H; window width, 550 H) and lung window setting (window level, –600 H; window width, 1500–2000 H). The CT scans were assessed by two radiologists who reached conclusions by consensus.
CT scans were reviewed for location, size, and internal characteristics of the tumor. The descriptions of tumor location included identification of the affected lung lobe and of the position—central versus peripheral—of the tumor in the lobe. Central tumors were defined as those that involved the carina or a main segmental bronchus. Peripheral tumors were defined as those surrounded by lung parenchyma or distal to the subsegmental bronchi. The scans were evaluated for the presence of endobronchial growth, obstructive pneumonia or atelectasis, and pleural effusion. On thin-section CT, tumor–lung interface characteristics were assessed and the presence or absence of surrounding emphysema was noted.
Enhancement of the nodules was assessed subjectively by comparing the enhancement of the chest wall muscle on the unenhanced scans with corresponding enhancement on the contrast-enhanced scans. The enhancement patterns of the nodules were noted as either homogeneous or inhomogeneous. Inhomogeneity was defined as the presence of an area of low attenuation in the highly attenuated background of the nodule. Nodules with peripheral rim enhancement were also considered to have inhomogeneous enhancement. For the 28 patients who underwent contrast-enhanced CT, we reviewed the pathologic reports to identify those patients whose reports mentioned that necrosis was present in gross specimens and then correlated the presence of necrosis with the CT findings. Statistical analyses of size and macroscopic necrosis between homogeneously and inhomogeneously enhanced tumors were performed using Student's t test and Fisher's exact test, respectively. Mediastinal or hilar lymph nodes that were larger than 1 cm in the short-axis diameter were regarded as evidence of lymphadenopathy. Tumors were staged according to the TNM classification scheme [8]. No patients had extrathoracic metastases at the time of diagnosis. Distribution of the clinical stages was as follows: stage IA (n = 20); stage IB (n = 7); stage IIA (n = 1); stage IIB (n = 3); stage IIIA (n = 5); and stage IIIB (n = 2). The CT findings regarding the tumor–lung interface and intratumoral necrosis were correlated with findings at surgery and pathologic examination.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Among the 19 patients who underwent thin-section CT, the more common tumor–lung interface characteristics were a well-defined interface (74%) and lobulation (79%) (Fig. 1). Spiculation (Figs. 2A, 2B and 3A, 3B, 3C) was observed in six patients (32%). In four patients, however, spiculated margins appeared as fibrotic strands because of paracicatricial emphysema or linear opacities made more pronounced by preexisting emphysema (Fig. 2A, 2B).
|
|
|
|
|
|
On contrast-enhanced CT, attenuation of all the tumors varied from slightly less than to more than the attentuation of the chest wall muscle. Among the 28 patients who had contrast-enhanced CT, 15 (54%) had tumors with homogeneous enhancement (Figs. 2A, 2B and 3A, 3B, 3C), and 13 tumors (46%) showed inhomogeneous enhancement. Of the inhomogeneously enhanced tumors, five had peripheral rim enhancement. As shown in Table 2, the inhomogeneously enhanced tumors were generally larger than the homogeneously enhanced tumors. At gross pathology examination of the resected tumors, necrosis was identified in seven of the 15 homogeneously enhanced carcinomas and in all of the 13 inhomogeneously enhanced carcinomas (Table 2). The largest homogeneously enhanced tumor with necrosis was 33 mm.
|
At histopathologic examination, pulmonary alveoli were filled entirely with tumor cells or had a compressive growth pattern, giving the appearance of a well-defined margin. Lobulated margins observed on CT reflected the interruption of tumor growth by anatomic structures or tumor extension due to differential tumor growth. In four patients, spiculation was attributable to paracicatricial or preexisting emphysema. In all other patients, spiculation corresponded to vascular and lymphatic invasion, a normal vascular structure, a thickened interstitium, or an irregularly protuberant tumor nest. In two patients, tumors associated with the bronchial wall were identified as endobronchial polypoid lesions that invaded the surrounding lung parenchyma. In one of the two patients, the invading lesion was accompanied by obstructive pneumonia. The cause of inhomogeneous enhancement was intratumoral necrosis.
Necrosis was noted in the gross pathology specimens of 20 of 28 patients who had contrast-enhanced CT scans. The difference between homogeneously and inhomogeneously enhanced tumors in the incidence of macroscopic necrosis was statistically significant (p < 0.003). Although the gross necrotic area was difficult to determine in the cut surface of the small tumors, multiple punctate necroses were observed microscopically in each case. In the large tumors, necrotic foci were larger and tended to be confluent. Consequently, the tumors contained extensive necroses. Inhomogeneous enhancement with peripheral rim enhancement represented a large central necrosis.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The incidence of large cell neuroendocrine carcinoma in cases of resectable primary lung cancers is 2.8–3.1% [2, 3]. The clinical features of the 38 patients with large cell neuroendocrine carcinoma in our study—the mean age, predominance of men, and strong association with smoking—were similar to those previously documented in the literature [2, 10]. However, 31 patients (82%) were asymptomatic and had large cell neuroendocrine carcinoma detected incidentally, which is a larger percentage than the percentages previously reported [6, 7]. This difference is presumably due to the number of patients with earlier stages of large cell neuroendocrine carcinoma in our study population.
Large cell neuroendocrine carcinomas have been reported as occurring in either central or peripheral locations [10]. Our experience supports a predominately peripheral location. In our series, large cell neuroendocrine carcinomas were slightly more common in the right lung than in the left and slightly more common in the upper lobe than in the middle or lower lobes. The reason for the higher incidence in these regions is uncertain; however, pulmonary carcinomas are known to occur with a relative frequency of 3:2 in the right versus left lung and in the upper versus lower lobe [12]. A well-defined and lobulated appearance without air bronchograms or calcification was the most common finding in the large cell neuroendocrine carcinomas observed in our study. These results are compatible with the recently reported descriptions of large cell neuroendocrine carcinoma [6, 7].
Jung et al. [7] reported that spiculation was present in eight (73%) of 11 cases of large cell neuroendocrine carcinoma. In our study, spiculation was found in six (32%) of 19 patients who had thin-section CT scans available. Spiculations are frequently seen in adenocarcinomas and occasionally in squamous cell carcinomas, corresponding to a diffuse desmoplastic response to tumor growth [13]. In our series, spiculated margins were caused by paracicatricial emphysema with extension into surrounding emphysematous lung or by linear opacities made more prominent by preexisting emphysema in four (67%) of the six tumors with spiculation. Because emphysema is thought to be caused by smoking, the spiculation of large cell neuroendocrine carcinomas in our study may be associated with the high incidence of smokers in our patients.
The scans of 13 of the 28 patients who had undergone contrast-enhanced CT showed inhomogeneous enhancement caused by gross intratumoral necrosis. Statistically, inhomogeneously enhanced tumors were larger than homogeneously enhanced tumors. Microscopic focal necrosis was present in each case, and the extent of necrosis varied considerably. In adenocarcinomas and squamous cell carcinomas, central necrosis is frequent and may be more extensive in larger tumors [12]. Similarly, in large-diameter large cell neuroendocrine carcinomas, the necrotic area seemed to be confluent and more extensive. Although small-diameter tumors contained macroscopic necrosis, they tended to show homogeneous enhancement. Identifying these necroses on contrast-enhanced CT was difficult because the necrotic foci were too small to be detected. Endobronchial growth has been associated with typical carcinoid tumors [12], but we observed it only infrequently in the large cell neuroendocrine carcinomas in our study. We found the incidence of postobstructive pneumonia or atelectasis on CT to be 8%, lower than the 27% (3/11 patients) reported by Jung et al. [7]. This difference in findings may be the result of infrequent central or endobronchial growth.
Before its new classification, large cell neuroendocrine carcinoma had been categorized in several ways. In reviewing previous cases seen in our institution, we noticed that some patients with large cell neuroendocrine carcinoma had formerly been diagnosed as having small cell carcinoma, adenocarcinoma, squamous cell carcinoma, or large cell carcinoma [3].
Radiographically, small cell carcinoma in most cases exhibits extensive hilar and mediastinal lymphadenopathy, and sometimes a primary pulmonary tumor is not recognized [12]. However, 5–10% of patients with small cell carcinoma have solitary pulmonary nodules [14–16]. Among eight patients in our study initially classified as having small cell carcinoma, six had been diagnosed as having an intermediate cell type of the disease. Yabuuchi et al. [16] reported that on CT, peripheral small cell carcinoma displays a well-defined, rounded, or lobulated homogeneous mass. The CT features on the scans of our patients initially diagnosed as small cell carcinoma were similar to the CT features of peripheral small cell carcinoma described earlier. Large cell neuroendocrine carcinoma and peripheral small cell carcinoma also have similar pathologic and CT features. Thus, histologic review of peripheral small cell carcinoma cases should be performed using new WHO criteria because it is possible that some of the patients with currently accepted diagnoses of peripheral small cell carcinoma should be reclassified as having large cell neuroendocrine carcinoma.
Our study has some limitations. The CT findings in our study population may not reflect the findings in all patients with large cell neuroendocrine carcinoma because our study was limited to patients whose tumors were surgically resected. This substantial selection bias might explain the presence of peripheral tumors in so many of our patients. However, a diagnosis of large cell neuroendocrine carcinoma is usually based on a histologic findings in a specimen obtained by surgical resection, as was the case in our study. More research is needed to clarify large cell neuroendocrine carcinoma findings in all patients, including patients with large cell neuroendocrine carcinoma who are not treated surgically. Other limitations of our study include a reliance on visual estimation of the CT scans with no comparison of the region of interest in the tumor with that in unenhanced scans and the variation in both the IV contrast injection rates (range, 1–2 mL/sec) and scanning times (range, 70–120 sec), each of which may have affected the homogeneity of the tumor.
In conclusion, we found that large cell neuroendocrine carcinoma usually appeared as a well-defined and lobulated tumor without air bronchograms or calcification. Infrequently, this appearance was accompanied by spiculation due to paracicatricial or preexisting emphysema. We found that the CT findings of large cell neuroendocrine carcinomas were similar to those of other expansively growing tumors, such as peripheral small cell carcinomas, poorly differentiated adenocarcinomas, and squamous cell carcinomas. Large-diameter large cell neuroendocrine carcinomas tended to show inhomogeneous enhancement because of necrosis, but this type of enhancement was not necessarily apparent in small-diameter (< 33 mm) large cell neuroendocrine carcinomas, even if the tumors were necrotic.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  Y. Yoshida, A. Kokubu, K. Suzuki, H. Kuribayashi, K. Tsuta, Y. Matsuno, M. Kusumoto, Y. Kanai, H. Asamura, S. Hirohashi, et al. Molecular Markers and Changes of Computed Tomography Appearance in Lung Adenocarcinoma with Ground-grass Opacity Jpn. J. Clin. Oncol., December 18, 2007; (2007) hym139v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Iyoda, K. Hiroshima, Y. Nakatani, and T. Fujisawa Pulmonary Large Cell Neuroendocrine Carcinoma: Its Place in the Spectrum of Pulmonary Carcinoma Ann. Thorac. Surg., August 1, 2007; 84(2): 702 - 707. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Chong, K. S. Lee, B.-T. Kim, J. Y. Choi, C. A Yi, M. J. Chung, D.-K. Oh, and J.-Y. Lee Integrated PET/CT of Pulmonary Neuroendocrine Tumors: Diagnostic and Prognostic Implications Am. J. Roentgenol., May 1, 2007; 188(5): 1223 - 1231. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Chong, K. S. Lee, M. J. Chung, J. Han, O J. Kwon, and T. S. Kim Neuroendocrine Tumors of the Lung: Clinical, Pathologic, and Imaging Findings RadioGraphics, January 1, 2006; 26(1): 41 - 57. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Joseph and C. Reinhold Statistical Inference for Continuous Variables Am. J. Roentgenol., April 1, 2005; 184(4): 1047 - 1056. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
