© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Clin Transplant 2014: 28: 1402–1409 DOI: 10.1111/ctr.12471

Clinical Transplantation

Cardiophrenic lymph nodes in liver transplant candidates with hepatocellular carcinoma: imaging characteristics and posttransplant outcomes Lee C, Kim A, Santos I, Cen Y, Alexopoulos S, Navarro S, Wallman M, Dhanireddy K, Grant E. Cardiophrenic lymph nodes in liver transplant candidates with hepatocellular carcinoma: imaging characteristics and post-transplant outcomes. Abstract: Background: No guidelines exist for the management of cardiophrenic lymph nodes in patients with hepatocellular carcinoma (HCC) being evaluated for liver transplantation. Methods: One hundred and seventy-eight patients with HCC listed for liver transplant received both pre-transplant computed tomography (CT) and follow-up CT scans. Enlarged cardiophrenic lymph nodes on CT were characterized and followed on subsequent scans; lymph node outcomes were assigned to “reduced” and “not reduced” categories. Tumor and patient characteristics were also recorded. Results: Seventy-one of one hundred and seventy-eight patients (39.9%) had at least one cardiophrenic lymph node larger than 8 mm in diameter on pre-transplant CT. One hundred and sixty-six total lymph nodes were characterized. Six lymph nodes (3.6%) in two patients increased in size on follow-up imaging; all six cardiophrenic lymph nodes were presumed to represent metastases. There was a statistically significant reduction in lymph node size in patients who were transplanted vs. those who were not transplanted. Furthermore, a statistically significant association was found between increasing Model for End-Stage Liver Disease score and lymph node size reduction. There were no significant differences in posttransplant survival between patients with different lymph node outcomes. Conclusion: In the absence of metastatic disease in other sites, these lymph nodes are probably reactive; further workup is likely not necessary.

Hepatocellular carcinoma (HCC) is the sixth most common tumor and the third most common cause of cancer death worldwide (1). In 2008, more than 700 000 cases of HCC were diagnosed worldwide (1). Cirrhosis is, by far, the greatest risk factor for developing HCC. When untreated, prognosis is extremely poor (2); however, early treatment can be associated with a median survival beyond five yr (3). Among the myriad therapies available, liver transplantation is the only option that can simultaneously cure both the neoplasm and underlying cirrhosis. Additionally, transplantation possesses the added advantage of treating undetected synchronous HCC. Liver transplantation has been reported to offer a four-yr survival of 75% in

1402

Christopher Leea, Andrew Kima, Idoia Santosb, Yong Cena, Sophoclis Alexopoulosc, Shannon Navarroa, Melissa Wallmand, Kiran Dhanireddyc and Edward Granta a

Department of Radiology, Keck School of Medicine of USC, Los Angeles, CA, USA, b Department of Radiology, Parc Sanitari Sant u, Barcelona, Spain, cDepartment Joan de De of Surgery, Keck School of Medicine of USC and dDepartment of Medicine, Keck School of Medicine of USC, Los Angeles, CA, USA Key words: hepatocellular carcinoma – liver transplantation – lymph nodes – metastasis – multidetector computed tomography Corresponding author: Christopher Lee, MD, Keck Hospital of USC, 1500 San Pablo Street, 2nd Floor Imaging, Los Angeles, CA 90033, USA. Tel.: 323-442-8721; fax: 323-442-8755; e-mail: [email protected] Conflict of interest: None. Accepted for publication 22 September 2014

patients who fulfill the Milan criteria (one tumor 5 cm or smaller, or up to three tumors 8 mm in diameter, these lymph nodes were likewise included and described. Sizes and segmental locations of the tumors were also recorded from the baseline scans. All subsequent CT examinations were reviewed to evaluate for changes in the indexed cardiophrenic lymph nodes, as well as for the development of new lymph nodes, progression of hepatic tumor burden, and new sites of metastatic disease. Increase or decrease in lymph node size was noted when either long- or short-axis measurement changed by at least 20%. The mean imaging surveillance period for each patient was 44.1 months (range 8–85 months). The clinical medical record was also reviewed for each patient, with a mean follow-up of 54.5 months (range 8–121 months). The following clinical parameters were recorded: Model for EndStage Liver Disease (MELD) score, tumor clinical stage, baseline Response Evaluation Criteria in

1403

Lee et al.

Solid Tumor (RECIST) score, baseline and highest alpha-fetoprotein (AFP) levels, transplant status, post-transplant tumor status, and survival status. For statistical analysis, cardiophrenic lymph node outcomes were divided into two groups: “reduced” and “not reduced.” The “reduced” category included those lymph nodes which decreased in size on follow-up CT, initially increased in size then subsequently decreased in size, or resolved completely. The “not reduced” category included those lymph nodes which were either stable or increased in size on follow-up examinations. Classification into these two groups was necessitated by the small number of lymph nodes which increased in size. Independent t-tests, Wilcoxon rank-sum tests, and chi-square tests were used for normally distributed data, not normally distributed data, and categorical data, respectively, in testing for differences in lymph node, tumor, and patient characteristics between the two lymph node outcome categories. To assess the relationship between transplant status and lymph node outcomes, patients were first stratified into “all lymph nodes reduced,” “partial lymph nodes reduced,” and “no lymph node reduced” categories, followed by chi-square test of the three groups with respect to transplant status. Kaplan–Meier survival curves and log-rank test were used to compare post-transplant survival between patients within the above three categories for both lymph nodes identified pre-transplant and those detected post-transplant. A p-value < 0.05 was considered statistically significant. All statistical analyses were conducted with SAS software version 9.4 (SAS Institute Inc., Cary, NC, USA). Results

Seventy-one of one hundred and seventy-eight (39.9%) patients had at least one cardiophrenic lymph node larger than 8 mm in long- or shortaxis diameter on pre-transplant CT; three patients had enlarged lymph nodes detected on post-transplant CT only. One hundred and sixty-six total cardiophrenic lymph nodes were characterized from these 74 patients. One hundred and twenty-three enlarged lymph nodes were recorded from the patients’ baseline examinations, and 43 lymph nodes were noted on follow-up examinations, including 33 lymph nodes detected on post-transplant CT. Overall, 84 of 166 (50.6%) lymph nodes were stable in size over at least six months followup, 67 (40.4%) lymph nodes ultimately decreased in size, nine (5.4%) lymph nodes resolved completely on follow-up CT, and three (3.6%) lymph nodes increased in size on follow-up imaging.

1404

Seventeen of seventy-four (23.0%) patients had available CT scans prior to developing HCC. However, only one enlarged lymph node in one patient developed following the appearance of HCC, which subsequently decreased in size following liver transplant. The remaining pre-transplant lymph nodes were all present prior to the development of HCC, and they either remained stable or decreased in size on follow-up scans. As shown in Table 2, anterior and right anteromedial locations were observed most frequently, with the right anteromedial location occurring 30.7% of the time. With regard to shape, elongated and ovoid were most common, with elongated occurring 48.2% of the time. The average baseline long-axis and short-axis dimensions of documented lymph nodes were 12.3  3.6 mm and 5.9  2.3 mm for “not reduced” and 12.3  3.6 mm and 6.1  1.7 mm for “reduced,” respectively. One hundred and forty-two of 166 (85.5%) lymph nodes had no visible fatty hilum, while 24 (14.5%) lymph nodes demonstrated a fatty hilum. There were no statistically significant differences in lymph node location, shape, size, and presence of a fatty hilum between the “reduced” and “not reduced” lymph node outcome categories. Fifty-seven of 71 (80.3%) patients with enlarged cardiophrenic lymph nodes on pre-transplant CT had received a liver transplant at the time of manuscript submission, with mean post-transplant surveillance period of 59.2 months. Fourteen (19.7%) patients had not received a transplant for various reasons (e.g., still on waiting list, deceased prior to transplant, removed from list due to increased tumor burden). One hundred and sixteen cardiophrenic lymph nodes in these 57 patients were documented on pre-transplant scans. Of the 116 pre-transplant lymph nodes, 56 (48.3%) were stable after transplant, 52 (44.8%) ultimately decreased in size, and eight (6.9%) resolved completely. No documented pre-transplant lymph nodes increased in size following transplantation. Figure 1 illustrates the outcomes of enlarged cardiophrenic lymph nodes based on transplant status. Among the 57 patients with pre-transplant cardiophrenic lymph nodes who had undergone transplantation, 24 (42.1%) patients had reduction in size of all the documented cardiophrenic lymph nodes, while 11 (19.3%) had reduction in size of at least one lymph node. Only 22 (38.6%) patients did not demonstrate a reduction in size of any cardiophrenic lymph node following liver transplantation. No significant association was observed between post-transplant surveillance period and lymph node outcome (p = 0.58). Among the patients who had not undergone transplantation,

Cardiophrenic lymph nodes liver transplant Table 2. Distribution of lymph node characteristics by lymph node outcomes

Location Anterior Left Anterolateral Left Anteromedial Left Posteromedial Right Anterolateral Right Anteromedial Right Posteromedial Shape Elongated Irregular Ovoid Round Triangular Short-axis dimension Long-axis dimension Presence of a fatty hilum Yes No

All

Reduced

Not Reduced

p-Value

37 (22.3) 24 (14.5) 20 (12.0) 5 (3.0) 15 (9.0) 51 (30.7) 14 (8.4)

16 (9.6) 15 (9.0) 8 (4.8) 3 (1.8) 6 (3.6) 23 (13.9) 5 (3.0)

21 (12.7) 9 (5.4) 12 (7.2) 2 (1.2) 9 (5.4) 28 (16.9) 9 (5.4)

0.64

80 (48.2) 5 (3.0) 64 (38.6) 7 (4.2) 10 (6.0) N = 166, 6.0  2.0 N = 166, 12.3  3.6

33 (19.9) 3 (1.8) 32 (19.3) 4 (2.4) 4 (2.4) N = 76, 6.1  1.7 N = 76, 12.3  3.6

47 (28.3) 2 (1.2) 32 (19.3) 3 (1.8) 6 (3.6) N = 90, 5.9  2.3 N = 90, 12.3  3.6

0.73

24 (14.5) 142 (85.5)

8 (4.8) 68 (41.0)

16 (9.6) 74 (44.6)

Fig. 1. Distribution of lymph node reduction between patients transplanted vs. not transplanted.

12 of 14 (85.7%) had no reduction in size of any cardiophrenic lymph node, while only two (14.3%) had partial reduction in size of the lymph nodes. Rates of lymph node size reduction were significantly different between patients who were transplanted vs. those who were not transplanted (p < 0.01). Only six lymph nodes (3.6%) in two patients increased in size on follow-up imaging. One patient demonstrated two enlarged cardiophrenic lymph

0.49 0.99 0.19

nodes on his baseline CT, with both lymph nodes increasing in size on the latest study with concurrent worsening intra- and extrahepatic disease (Fig. 2). Consequently, the patient was removed from the transplant list. The second patient had no appreciable cardiophrenic lymph nodes on his pretransplant scans and underwent uneventful liver transplantation. Following transplantation, however, four enlarged right cardiophrenic lymph nodes appeared on his latest post-transplant CT, along with a malignant right pleural effusion (Fig. 3). All six lymph nodes in both patients were presumed to represent nodal metastatic disease. Aside from the patient noted above, six additional patients demonstrated post-transplant hepatic recurrence and/or metastatic disease. Eight enlarged lymph nodes in five of these patients were detected on pre-transplant CT. All eight lymph nodes were stable in size during the pre-transplant period, with five decreasing in size and three remaining stable following transplantation. Six enlarged lymph nodes in three patients were initially detected on post-transplant CT, with five decreasing in size and one remaining stable on follow-up. No lymph node characteristics were present to predict post-transplant recurrence. Analogous to the study population, the majority of pre-transplant cardiophrenic lymph nodes occurred in patients with either hepatitis C or combined hepatitis C and alcohol as the etiologies of cirrhosis (Table 1). No significant association was found between the etiology of liver disease and lymph node outcome. Similarly, the study population and patients with pre-transplant lymph nodes demonstrated similar MELD scores (Table 1).

1405

Lee et al.

Fig. 2. Pre-transplant axial computed tomography (CT) image (A) reveals an 11 9 7 mm right anteromedial cardiophrenic lymph node (arrow). Follow-up CT 26 months later (B) demonstrates interval enlargement of the cardiophrenic lymph node (arrowhead), measuring 15 9 10 mm. Axial CT image more inferiorly (C) also shows a new soft tissue nodule anterior to the liver (open arrow), compatible with metastasis. There has also been progression of the tumor burden within the right hepatic lobe (not shown). The patient was subsequently removed from the transplant list.

However, there was a statistically significant association between increasing MELD score and lymph node size reduction (p = 0.03). Lymph nodes that decreased in size were associated with higher MELD scores. Moreover, when only including patients who underwent liver transplantation, the association became even stronger (p < 0.01). No significant differences were observed in tumor size or location, tumor clinical stage, baseline RECIST score, and baseline or highest subsequent AFP level between the two lymph node outcome categories. Table 3 summarizes the associations between the various tumor and patient characteristics and lymph node outcomes. Additionally, no association was present between tumor location and lymph node location (p = 0.86). Finally, there were no significant differences in post-transplant survival between patients with different lymph node outcomes, regardless of whether the lymph nodes were detected pre-transplant (Fig. 4A) or post-transplant (Fig. 4B).

1406

Discussion

As metastatic disease is a contraindication for liver transplantation, the prospective diagnosis of extrahepatic spread is essential when evaluating patients with HCC for possible transplantation, particularly in the current organ shortage era. Indeed, post-transplant recurrence likely occurs in patients who had harbored occult metastases at the time of their liver transplant. Although HCC may metastasize to cardiophrenic lymph nodes, no guidelines exist for the management of these lymph nodes in patients with HCC being evaluated for transplantation. Therefore, we sought to estimate the prevalence of metastatic cardiophrenic lymph nodes in liver transplant candidates with HCC, evaluate the outcomes of these lymph nodes following transplantation, and assess for any correlation between lymph node outcomes and post-transplant survival. In our patient population, nearly 40% of patients had at least one enlarged cardiophrenic lymph node on pre-transplant CT; yet, only 1.1% of patients harbored presumed cardiophrenic

Cardiophrenic lymph nodes liver transplant

A

B

Fig. 3. Pre-transplant axial computed tomography (CT) image (A) demonstrates no appreciable cardiophrenic lymph nodes. Esophageal varices (open arrow) are present secondary to portal hypertension. Post-transplant axial CT image (B) reveals several new, enlarged, right-sided cardiophrenic lymph nodes (white arrows). There is also a new right pleural effusion with associated pleural masses (*), compatible with pleural carcinomatosis.

nodal metastases. We observed a statistically significant reduction in lymph node size in patients who were transplanted vs. those who were not Table 3. Summary of associations between tumor and patient characteristics and lymph node outcomes Characteristic

p-Value

Etiology of cirrhosis MELD score (all patients) MELD score (only patients transplanted) Dominant tumor size Dominant tumor location Tumor clinical stage Baseline RECIST score Baseline AFP level Highest AFP level

0.22 0.03

Cardiophrenic lymph nodes in liver transplant candidates with hepatocellular carcinoma: imaging characteristics and post-transplant outcomes.

No guidelines exist for the management of cardiophrenic lymph nodes in patients with hepatocellular carcinoma (HCC) being evaluated for liver transpla...
438KB Sizes 0 Downloads 7 Views