Many of the difficulties that beset early taxonomists in the classification of NHL have been overcome with improved immunological and molecular methods of diagnosis. The Revised European–American Lymphoma (REAL) classification in 1994 depended on a triad of morphology, immunophenotype and molecular methods as well as clinical features for defining disease entities,² differentiating it from earlier morphologically based classifications. The scheme forms the backbone of the World Health Organisation (WHO) classification of tumours of haematopoietic and lymphoid tissues.³ A summary of the WHO classification (4th edition) is given in Table 70-1. The WHO classification stratifies neoplasms by lineage into clinically distinct disease entities and is a real advance in the ability to identify disease accurately and consistently. Further, it can be refined so as to improve patient management.⁴ For example, it has recently been shown that gene expression profiling in diffuse large B-cell lymphoma (DLBCL) enables recognition of discrete subsets (germinal centre B-cell type and activated B-cell type) which have independent prognostic significance, and this has been included in the 4th edition of the classification.³ Other additions include paediatric follicular lymphoma, primary DLBCL of the central nervous system (PCNSL), and two so-called ‘grey zone’ lymphomas: B-cell lymphoma with features intermediate between DLBCL and classical HL, and B-cell lymphoma with features intermediate between DLBCL and BL.

Accurate diagnosis requires adequate tissue biopsy and an experienced histopathologist. Some lesions, for example, retroperitoneal, mediastinal and mesenteric masses, may be amenable to ultrasound or CT-guided core-needle biopsy, which may safely yield adequate tissue for histological diagnosis and immunophenotyping⁶ but, as with HL, an entire lymph node is preferable for diagnosis.

Imaging Nodal Disease

At present, size is the only criterion by which lymph nodes demonstrated on CT or MRI are considered to be involved, though clustering of multiple small lymph nodes, for example within the anterior mediastinum or the mesentery, is suggestive. A maximum short-axis diameter of 10 mm is taken to be the upper limit of normal, depending upon the exact site within the neck, thorax, abdomen, or pelvis. There are exceptions: normal neck jugulodigastric node can measure up to 13 mm short-axis diameter; nodes in the gastrohepatic ligament and porta hepatis are considered abnormal if they measure more than 8 mm in diameter; retrocrural nodes greater than 6 mm are taken as enlarged;⁹ and in the pelvis the upper limit of normal is 8 mm.¹⁰ Lymph nodes at some sites, such as the splenic hilum, presacral and perirectal areas, are not usually visualised on cross-sectional imaging and, whenever demonstrated, are likely to be abnormal.

Enlarged lymph nodes in both HL and NHL are usually homogeneous, of soft-tissue density on CT. Mild or moderate uniform enhancement occurs after intravenous injection of contrast medium. Calcification is uncommon but may be seen on post-treatment images. Necrosis is rarely seen in large nodal masses in both HL (particularly nodular sclerosing HL) and aggressive NHL, more frequently after treatment. On MRI, involved lymph nodes have low-to-intermediate signal intensity on T1-weighted images, and they may have very high signal intensity on fat-suppressed T2-weighted and short tau inversion recovery (STIR) sequences. Though the signal intensity of involved nodes and the presence of necrosis do not appear to have much prognostic significance, there is some evidence that within large lymphomatous masses, heterogeneous T2 signal at magnetic resonance imaging (MRI), or heterogeneous enhancement at CT, is associated with a worse outcome.

Choice of Imaging Technique

CT has been the technique of choice for the staging and follow-up of lymphoma for some time, and enables localisation of the most appropriate lesion for consideration of percutaneous image-guided biopsy. Ultrasound has no value in whole-body staging. Ultrasonographic appearances of lymphomatous nodal disease are non-specific, though the pattern of vascular perfusion as demonstrated by power Doppler interrogation may suggest the diagnosis, lymphomatous nodes having rich central and peripheral perfusion. The main value of ultrasound in lymphoma lies in confirming the nature of a palpable mass and assessing the major viscera. The accuracy of MRI in detecting lymph node involvement is equal to that of CT (and is better in some areas such as the supraclavicular fossa and within the pelvis), but it has no particular advantage over CT in this respect and its role is essentially adjunctive, to solve problems or monitor response to treatment. Recent advances in MRI technology (high field strength magnets and parallel imaging) have enabled MRI to be used for whole-body staging: the role of whole-body diffusion-weighted imaging in staging and response assessment is a field of active research.¹¹ Major advantages in patients with HL in particular (who are often young) include the lack of ionising radiation.

Detection of disease in normal-sized nodes is not possible with cross-sectional imaging, nor is it possible to differentiate between nodal enlargement secondary to lymphoma or reactive hyperplasia. Functional imaging is often able to make this distinction. The superior diagnostic accuracy and more favourable imaging characteristics of PET with FDG and, latterly, PET/CT, has resulted in a dramatic decline in the use of gallium-67 scintigraphy, which no longer has a role as an isolated tool in the staging of lymphoma.

Numerous studies have shown that FDG-PET is at least as accurate as CT in the depiction of nodal and extranodal disease^12,13 and more sensitive than Ga-67 scintigraphy. It results in clinically significant upstaging in up to 10–20% of patients compared to CT, which may result in changes in therapy, particularly in HL.¹⁴ Most NHLs are FDG-avid, though false-negative studies can occur with low-grade lymphomas such as small lymphocytic lymphoma, cutaneous lymphoma, some peripheral T-cell lymphomas and MALT types. For these subtypes, contrast-enhanced CT remains the standard of care. The development of PET/CT with accurate co-registration means that both morphological and functional abnormalities can be assessed simultaneously with improved diagnostic performance of FDG-PET as a result.¹⁵ It is important to recognise that lymphomatous involvement of certain organs can be very difficult to recognise with FDG-PET/CT, because of physiological uptake—for example, in the stomach and central nervous system.¹⁶ Debate continues as to whether it is necessary to carry out a full diagnostic CT imaging as part of the PET/CT study and often a low-dose CT, for the purposes of attenuation correction and anatomical correlation, is sufficient.^17,18

Neck

Between 60 and 80% of patients with HL present with cervical lymphadenopathy. The spread of the disease is most frequently to contiguous nodal groups, with involvement of the internal jugular chain and spread to other deep lymphatic chains in the neck. Patients with supraclavicular or bilateral neck adenopathy are at increased risk of infradiaphragmatic disease.

Cervical adenopathy is less common in NHL, but commonly occurs in association with extranodal disease in Waldeyer’s ring. Approximately 40–60% of patients who present with head and neck involvement will have disseminated NHL. Involved nodal groups tend to be non-contiguous. Central necrosis within a lymph node is rarely seen. Imaging with contrast-enhanced CT or MRI has a useful role in evaluating the neck in patients with lymphoma, as it may identify enlarged nodes which are impalpable. It is helpful in response assessment, particularly in patients treated with radiotherapy, where post-treatment fibrosis renders clinical assessment difficult.

Thorax

Intrathoracic nodes are involved at presentation in 60–85% of patients with HL and 25–40% of patients with NHL.¹⁹ Nodes larger than 1 cm short-axis diameter are considered enlarged. Any intrathoracic group of nodes may be affected, but all the mediastinal sites other than paracardiac and posterior mediastinal nodes are more frequently involved in HL than NHL. Nearly all patients with nodular sclerosing HL have disease in the anterior mediastinum. The frequency of nodal involvement in HL is as follows:¹⁹ prevascular and paratracheal—84% (Fig. 70-1); hilar—28% (Fig. 70-1); subcarinal—22% (Fig. 70-1); others—5% (aortopulmonary, anterior diaphragmatic, internal mammary (Fig. 70-2)). In NHL involvement of the hilar and subcarinal groups is rarer, occurring in 9 and 13%, respectively, whereas superior mediastinal nodes are involved in 35%.²⁰

The great majority of cases of HL show enlargement of two or more nodal groups, whereas only one nodal group is involved in up to half of the cases of NHL. Hilar nodal enlargement is rare without associated mediastinal involvement, particularly in HL. Although paracardiac and internal mammary nodes are rarely involved at presentation in HL, they may be involved in recurrent disease, as they are not included in the classical ‘mantle’ radiation field. In HL and NHL, large anterior mediastinal masses usually represent thymic infiltration as well as a nodal mass (Fig. 70-3). A large anterior mediastinal mass in HL is recognised as an adverse prognostic feature and, as such, defines the need for more aggressive initial therapy. CT will demonstrate unsuspected mediastinal nodal enlargement despite a normal chest radiograph in 10% of patients with HL, and these patients have a poorer prognosis. CT of the chest has been shown to alter management in up to 25% of patients with HL. The therapeutic impact is less in patients with NHL, who are likely to receive chemotherapy regardless of stage. Impalpable axillary nodal enlargement is also frequently detected on CT in HL and NHL.

At presentation the retroperitoneal nodes are involved in 25–35% of patients with HL but up to 55% of patients with NHL.²¹ Mesenteric lymph nodes are involved in more than half the patients with NHL and less than 5% of patients with HL.²¹ Other sites such as the porta hepatis and splenic hilum are also less frequently involved in HL than NHL (Fig. 70-4). In HL, nodal spread is predictably from one lymph node group to another through directly connected lymphatic pathways. Nodes are frequently of normal size or only minimally enlarged. Spread from the mediastinum occurs through the lymphatic vessels to the retrocrural nodes, coeliac axis and so on. Around the coeliac axis, multiple normal-sized nodes may be seen, which can be difficult to evaluate because involved, normal-sized nodes are frequent in HL²² (Fig. 70-4). The coeliac axis, splenic hilar and porta hepatis nodes are involved in about 30% of patients and splenic hilar nodal involvement is almost always associated with diffuse splenic infiltration (Fig. 70-4). The node of the foramen of Winslow (porta caval node), lying between the portal vein and the inferior vena cava, is important, as it is often overlooked and may be the only site of disease relapse. It has a triangular shape; its normal long-axis diameter is up to 3 cm and in the anteroposterior plane is approximately 1 cm.

In NHL, nodal involvement is frequently non-contiguous and bulky and is more frequently associated with extranodal disease. Discrete mesenteric nodal enlargement or masses may be seen with or without retroperitoneal nodal enlargement. Large-volume nodal disease in both mesentery and retroperitoneum may give rise to the so-called ‘hamburger’ sign, in which a loop of bowel is compressed between two large nodal masses (Fig. 70-5). Multiple normal-sized mesenteric nodes should be regarded with suspicion for the diagnosis of lymphoma and lymphoma is a recognised cause of the ‘misty mesentery’. In NHL, regional nodal involvement is frequently seen in patients with primary extranodal lymphoma involving an abdominal viscus. Involved nodes tend to enhance uniformly and the presence of multilocular enhancement should suggest an alternative diagnosis such as tuberculosis or atypical infection.

Some authors also separate out AIDS-related lymphoma (ARL) and the post-transplant lymphoproliferative disorder (PTLD),²⁵ both of which commonly affect the lungs.

Lung involvement at presentation occurs in just under 4% of patients with NHL, but in approximately 12% of patients with HL. It is usually secondary to direct extension of nodal disease into the adjacent parenchyma, hence its paramediastinal or perihilar location. In this circumstance there is no effect on stage; the ‘E’ lesion. Patients with HL presenting with an intrapulmonary lesion in the absence of demonstrable mediastinal disease are unlikely to have lymphomatous disease of the lung unless there has been previous mediastinal or hilar irradiation, when recurrence may be confined to the lungs. Conversely, in NHL, nodal disease is absent in 50% of those patients with pulmonary or pleural involvement. As nodal disease progresses or relapses, lung involvement becomes commoner in HL and NHL, such that 30–40% of patients with HL have pulmonary involvement at some stage during the course of the disease.

The radiographic appearances are extremely variable, but the commonest pattern is of one or more discrete nodules, with or without cavitation, which tend to be less well defined than those of primary or metastatic carcinoma, which they otherwise resemble (Figs. 70-7 and 70-8).²⁶ The disease often spreads along lymphatic channels and involves lymphoid follicles around bronchovascular divisions, resulting in peribronchial nodulation spreading out from the hila, which can result in streaky shadowing visible on chest radiographs and at CT (Fig. 70-7).

Less commonly, lymphomatous cells fill the pulmonary acini, producing rounded or segmental areas of consolidation with air bronchograms (Fig. 70-8). Nodulation along the bronchial wall may enable differentiation from infective consolidation. A rare pattern of disease is widespread interstitial reticulonodular shadowing, producing a lymphangitic picture. Another rare manifestation is atelectasis, which usually results from endobronchial lymphoma rather than extrinsic compression by nodal disease.

The differential diagnosis of pulmonary involvement in lymphoma is extensive, and includes drug-induced changes, the effect of radiotherapy, and opportunistic infection during or following chemotherapy, particularly in patients with antecedent immunosuppression. Precise clinical correlation is essential in determining the most likely diagnosis.

Pleural Disease

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related posts:

Venous Access and Interventions Bone Tumours and Neuroblastoma in Children Genitourinary Tract Trauma Aortic Intervention The Neonatal and Paediatric Chest Skeletal Radiology in Children

Stay updated, free articles. Join our Telegram channel

Join