Accuracy of whole-body MRI in the assessment of splenic involvement in lymphoma

Abstract

Background

Accurate evaluation of the spleen is an important component of staging lymphoma, because this may have prognostic and therapeutic implications.

Purpose

To determine the diagnostic value of whole-body magnetic resonance imaging (MRI), including diffusion-weighted imaging (whole-body MRI-DWI) in the detection of splenic involvement in lymphoma.

Material and Methods

This IRB approved, prospective multicenter study included a total of 107 patients with newly diagnosed, histologically proven lymphoma who underwent 1.5 T whole-body MRI-DWI and FDG-PET/CT. Whole-body MRI-DWI and FDG-PET/CT were independently evaluated by a radiologist and a nuclear medicine physician, in a blinded manner. Splenic involvement at MRI was defined as splenic index > 725 cm³ or discrete nodules. At FDG-PET/CT splenic involvement was defined as splenic uptake greater than liver uptake or hypodense nodules at contrast-enhanced CT. FDG-PET/CT augmented with follow-up imaging after treatment was used as reference standard.

Results

Splenic involvement was detected with FDG-PET/CT in 21 patients, all demonstrating response to treatment. The sensitivity, specificity, positive predictive value, and negative predictive value of whole-body MRI-DWI for the detection of splenic involvement were 85.7 %, 96.5 %, 85.7%, and 96.5%, respectively. Three out of six discrepancies were related to suboptimal criterion of splenic size used with whole-body MRI-DWI versus the size-independent FDG uptake.

Conclusion

Whole-body MRI-DWI is reasonably accurate in the detection of splenic lymphomatous involvement.

Keywords

Splenic involvement lymphoma magnetic resonance imaging (MRI)whole-body MRI diffusion-weighted imaging FDG PET/CT

Introduction

Accurate evaluation of splenic involvement is an important component of lymphoma staging, because this may have prognostic and therapeutic implications (1,2). At staging, the spleen is involved in 30–40% of patients with Hodgkin's lymphoma (2 –4) and in 10–40% of patients with non-Hodgkin's lymphoma (NHL) (5,6). Splenic involvement is considered nodal in Hodgkin's lymphoma and extranodal in NHL. The spleen is the most common single site of infradiaphragmatic disease in Hodgkin's lymphoma and occurs in about 10% of patients diagnosed with thoracic Hodgkin's lymphoma (7). Four patterns of involvement at imaging have been described: splenomegaly; diffuse infiltration with very small nodules (<5 mm); multiple small to large nodules (up to 10 cm); or a large solitary mass (8). Since splenectomy is no longer performed as part of staging, accurate non-invasive diagnosis is required. Previously, staging was predominantly performed with computed tomography (CT) with a reported wide range of diagnostic performances for splenic disease with sensitivity and specificity ranging from 33–94% and 0–100%, respectively (9). More recently, 18F-fluoro-2-deoxy-D-glucose positron emission tomography (FDG-PET)/computed tomography (CT) has been recommended for staging of FDG-avid lymphomas (10,11). In FDG-avid lymphomas, splenic uptake greater than hepatic uptake is considered as a relatively reliable indicator of lymphomatous involvement of the spleen (12,13). An important disadvantage of FDG-PET/CT, however, is the exposure of patients to ionizing radiation, which is particularly relevant in children and adolescents. Therefore, there is a growing interest to use whole-body magnetic resonance imaging (MRI) as a radiation-free alternative to FDG-PET/CT for staging lymphoma where accurate assessment of splenic involvement is an essential component. Previous studies have shown an overall moderate to good agreement between whole-body MRI and FDG-PET/CT for staging lymphoma (14 –17). However, to the best of our knowledge, there have not been any large studies that have investigated the value of whole-body MRI for the detection of splenic involvement, as a result of which its value in this setting is still unclear. The aim of this prospective study was to compare the diagnostic value of whole-body MRI including diffusion-weighted imaging (DWI) with FDG-PET/CT in the detection of splenic involvement in newly diagnosed lymphoma.

Material and Methods

Patients

Patients were selected from a larger cohort of 189 patients who had been included in a prospective multicenter study that investigated the diagnostic value of whole-body MRI in staging lymphoma compared with CT or FDG-PET/CT. Previously reported findings of this prospective study only dealt with the comparison of whole-body MRI with contrast-enhanced CT (18) or only included a limited number of patients with splenic involvement (19). Only patients with complete whole-body MRI-DWI and FDG-PET/CT studies were eligible for inclusion in the current analysis. Six hospitals participated in the enrolment of patients for this analysis; University Medical Centre Utrecht, Academic Medical Centre Amsterdam, Meander Medical Centre Amersfoort, Vall d'Hebron in Barcelona, Instituto Giannina Gaslini in Genoa, and KK Women's and Children's Hospital in Singapore. The local institutional review boards of all participating centers approved this prospective study. All study participants and/or parents or guardians, depending on the age of the participant, gave their written informed consent. Inclusion criteria were age 6 years and older with histologically proven lymphoma and the possibility to perform whole-body MRI-DWI within 15 days of FDG-PET/CT. Exclusion criteria were general contraindications for MRI (including implanted pacemaker and claustrophobia), previous malignancy, patients who were pregnant and patients in whom therapy had already started.

Whole-body MRI protocol

Non-contrast-enhanced whole-body MRI including DWI of head and neck, chest, abdomen, and pelvis was performed with a 1.5-T system (Achieva or Ingenia, Philips Medical Systems, Best, The Netherlands; Avanto, Siemens Medical Solutions, Erlangen, Germany; or Signa, GE Healthcare, Fairfield, CN, USA). For the T1-weighted (T1W) and T2-weighted short inversion time inversion-recovery (T2W-STIR) the body coil was used for signal reception in the majority of patients. In only the most recent included patients (n = 17), the surface coil was available. However, for acquisition of the DWI sequence, the surface coil was used for signal perception in all patients. Detailed scan parameters are listed in Table 1.

Table 1.

Scan parameters for whole-body MRI-DWI at 1.5T for Philips*, Siemens^†, and GE^‡.

Parameter	T1W	T2W-STIR	DWI-SFS	DWI-STIR
Pulse sequence	Single-shot turbo spin-echo		Single-shot spin-echo echo-planar imaging
Plane	Coronal	Coronal	Axial	Axial
Repetition time (ms)	537*^† or 400^‡	2444*^† or 6320^‡	6962^† or 7500^‡	8612*
Echo time (ms)	18	64	78
Inversion time (ms)	–	165	–	180
Slice thickness/gap (mm)	6/1		4/0
Slices per station (n)	30*^† or 26^‡		60
Cranio-caudal coverage per station (mm)	265		240
Field of view (mm²)	530 × 265	530 × 265	450 × 360
Acquisition matrix	208 × 287*^† or 512 × 352^‡	336 × 120*^† or 320 × 224^‡	128 × 81
B-values (s/mm²)	–		0 and 1000
Signals averaged (n)	1	2	3*^† or 10^‡
Respiratory motion compensation techniques for chest and abdomen	Breath holding	Respiratory triggering	Image acquisition under free breathing
Total effective scan time	5 min 48 s*^†, 5 min 12s^‡	5 min 13 s*^†, 6 min 15s^‡	14 min 4 s^†, 32 min 48s^‡	16 min 26 s*

In two patients b-values of 0, 100, and 800 s/mm² were used.

SFS, spectral fat saturation; STIR, short inversion time inversion recovery.

FDG-PET/CT protocol

All patients diagnosed with lymphoma underwent a diagnostic 18F-FDG PET/CT as a standard of care. FDG-PET was performed with five different PET systems (Biograph 16 PET-CT or Biograph 40 Truepoint PET-CT, Siemens Medical Solution, Erlangen, Germany; Gemini TOF PET-CT or Allegro, Philips Medical Systems, Best, The Netherlands). Patients fasted for at least 6 h before receiving FDG intravenously. A dose of 2–3.7 MBq/kg body-weight was used. Before injection of FDG, blood glucose levels were checked to exclude hyperglycemia (>11 mmol/L; i.e. >198 mg/dL). Approximately 60 min after FDG injection, PET emission scanning was performed covering the skull base to mid-femur level with the patient in supine position. Prior to PET acquisition, unenhanced CT images were obtained for attenuation correction. Finally, the majority of patients (n = 85) underwent contrast-enhanced CT.

MRI interpretation

Using a Picture Archiving and Communications System (PACS, Sectra Medical Systems, Linköping, Sweden) a radiologist (RJN with 17 years of MRI experience) independently evaluated the whole-body MRI-DWI using a standardized form. The reader was aware that patients were diagnosed with lymphoma but he was unaware of other imaging findings or clinical information. The whole-body MRI-DWI was reviewed in a locked sequence: the conventional sequences alone (T1W and T2W-STIR) and conventional sequences combined with DWI. Splenic size was assessed and/or focal involvement was recorded. The spleen was considered enlarged if the splenic index exceeded >725 cm³ (splenic index = length × height × thickness at the hilum) as described by others (4,9). For patients <11 years a splenic index >500 + (20 × age [in years]) cm³ was used as proposed by The Paediatric Oncology Group (20). Size of the largest nodule measured at T2W-STIR images and signal characteristics of the focal involvement at conventional and DWI sequences were recorded. The findings at whole-body MRI-DWI were considered positive for involvement if the spleen was either enlarged or contained focal lesion(s). Lesions with signal characteristics similar to cerebral spinal fluid were classified as benign lesions. ADC measurements were not performed as the majority of nodules were less than or equal to 10 mm which approaches to the spatial resolution of DWI (3.5 × 3.5 × 4 mm).

FDG-PET/CT image interpretation

The unenhanced, low-dose FDG-PET/CT images were transferred to DICOM software Osirix (available free online at http://www.osirix-viewer.com) version 3 and evaluated by a nuclear medicine physician (BdK with 9 years of clinical experience). Focal or diffuse splenic uptake, greater than hepatic uptake was considered positive for splenic involvement (12,13). FDG avidity per patient was recorded. Non-FDG-avid lymphoma was defined as clinically proved lymphoma without any site with increased FDG uptake (21). Using a Picture Archiving and Communications System (PACS, Sectra Medical Systems) another radiologist (FB with 25 years of clinical experience in CT) independently evaluated these datasets using a standardized form. Both splenic nodules and enlargement were recorded. Splenic nodule was defined as a hypodense area compared to the surrounding splenic parenchyma. Findings at FDG-PET and CT were combined to determine splenic disease involvement. Splenic involvement was defined as focal or diffuse splenic uptake greater than liver uptake at FDG-PET or hypodense nodules <1 cm at contrast-enhanced CT (nodular size around or below the spatial resolution of FDG-PET). In case of non-FDG avidity (n = 3), morphological features at contrast-enhanced CT (splenic enlargement > 725 cm³ and/or splenic nodules) were used to determine involvement.

Reference standard

An independent, unblinded expert panel assessed the discrepancies between FDG-PET/CT and whole-body MRI-DWI. This panel consisted of two observers who were not involved with the previous reading sessions (a nuclear medicine physician [MGGH with 12 years of clinical experience with FDG-PET] and a radiologist [AL with 9 years of clinical experience in MRI]). The panel had access to all available concurrent and follow-up investigations, results from bone marrow biopsy and clinical follow-up information at least 6 months after start of initial treatment. Discrepancies between whole-body MRI-DWI and FDG-PET/CT were classified as reader error or intrinsic error. Reader error was defined as being caused by either failure of detection of an abnormality or by an incorrect interpretation of an abnormal finding. Intrinsic error was defined as an error caused by limitations of the imaging acquisition or technique. In all patients follow-up information was reviewed to derive a reference standard. If the initial FDG-PET/CT demonstrated splenic involvement that showed visual decrease of FDG uptake after treatment, these were considered true positive for splenic involvement (n = 21). In case no response was seen, these were considered negative for splenic disease at initial staging (n = 0).

Statistical analysis

Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) along with 95% confidence intervals (CI) of splenic disease involvement for whole-body MRI-DWI were calculated against the reference standard using Statistical Package for the Social Sciences (SPSS Inc., version 20.0, Chicago, IL, USA).

Results

Patients

A total of 189 patients were eligible for inclusion. However, 78 patients were excluded from analysis because of the following reasons: no PET available (n = 44); logistic circumstances or no informed consent (n = 27); disease other than lymphoma (n = 4); could not tolerate MRI (n = 3). Thus, 111 patients were enrolled between June 2006 and June 2013. Four patients were excluded from final analysis due to incomplete whole-body MRI-DWI studies. Thus, a total of 107 consecutive patients (69 boys/men, 38 girls/women; mean age, 42.7 years; age range 6–82 years; Table 2) were finally included of whom 87 were included in previous analyses (18,19). In these studies either the performance of whole-body MRI compared to contrast-enhanced CT (18) was assessed or the number of patients with splenic involvement was relatively small (19). The flow diagram according to standards for reporting of diagnostic accuracy studies (STARD) is illustrated in Fig. 1.

Table 2.

Characteristics of included patients (n = 107).

Characteristics	HL	NHL
Patients (n)	33	74
Sex (n)
Male	14	55
Female	19	19
Age (years)
Mean ± SD	20.4 ± 10.4	52.7 ± 20.6
Range	6–45	8–82
Pathological subtype (n)
Nodular sclerosing HL	27
Lymphocyte-rich HL	3
Classic HL, NOS	1
Nodular paragranuloma	2
Diffuse large B-cell lymphoma		31
Follicular lymphoma		13
Mantle cell lymphoma		6
Extranodal marginal zone lymphoma		3
Nodal marginal zone lymphoma		1
Anaplastic large cell lymphoma		3
T-cell lymphoblastic lymphoma		5
Angioimmunoblastic T-cell lymphoma		2
Small lymphocytic lymphoma		3
B-cell NHL NOS		1
T-cell NHL NOS		1
Burkitt lymphoma		1
T-cell anaplastic lymphoma		1
T-cell rich B-cell lymphoma		1
Lymphoplasmacytic lymphoma		1
Unspecified lymphoproliferative disorder		1

HL, Hodgkin's lymphoma; NHL, non-Hodgkin lymphoma; NOS, not otherwise specified; SD, standard deviation.

Fig. 1.

Flow diagram according to standard of reporting of diagnostic accuracy studies (STARD).

Derivation of reference standard

In all patients follow-up imaging with FDG-PET/CT (either at early response assessment or after completion of therapy) was available for review, except in three patients. In one patient follow-up imaging was performed with contrast-enhanced CT. Because of initial presentation with only bone marrow biopsy proven disease, one patient was followed with bone marrow biopsies. Another patient who was suspected for primary marginal zone lymphoma of the spleen underwent splenectomy after initial staging. None of these three patients were excluded from the final analysis. Comparison of whole-body MRI-DWI and FDG-PET/CT enhanced with follow-up imaging resulted in seven discrepant cases. At review by our unblinded expert panel, one FDG-PET/CT reader error (interpretation error of focal involvement) and no FDG-PET/CT intrinsic error were identified. Our reference standard for splenic involvement was derived by the original FDG-PET/CT reading enhanced with follow-up information and corrected for FDG-PET/CT reader error. Twenty-one patients exhibited splenic disease involvement, all demonstrating response to treatment.

Diagnostic accuracy of whole-body MRI

Compared to the reference standard, 18 out of 21 patients with splenic involvement were diagnosed at whole-body MRI-DWI (Fig. 2). Whole-body MRI-DWI correctly excluded splenic involvement in 83 out of 86 patients. The sensitivity, specificity, PPV, and NPV for splenic disease with whole-body MRI-DWI was therefore 85.7% (CI, 62.6–96.2%), 96.5% (CI, 89.4–99.1%), 85.7%, and 96.5%, respectively. There was no additional value of DWI in any case during the initial reading. The expert panel assessed the three false positive and three false negative MRI findings to ascribe reader and intrinsic errors (Table 3). In two false negative whole-body MRI-DWI cases, the focal splenic involvement was visible on the original axial b = 1000 s/mm² diffusion-weighted images, thus labeled as reader errors (Fig. 3). Three errors (one false negative and two false positive cases) were ascribed to intrinsic failure of whole-body MRI-DWI where splenic size criterion is suboptimal compared to the robust functional information obtained with FDG-PET (Fig. 4). One splenic lesion with a size above the detection limit of PET did not show increased FDG uptake and therefore was considered benign according to our predefined criteria of involvement (Fig. 5). Unfortunately, this patient was followed up with only bone marrow biopsies.

Fig. 2.

Images of a 14-year-old boy with stage IV anaplastic large cell lymphoma illustrate diffuse splenic disease depicted at both FDG-PET and whole-body MRI-DWI. Coronal T2W-STIR (a), coronal gray-scale inverted maximum intensity projection DWI (b), axial DWI (d; b = 1000s/mm²) and axial ADC map (e) show an enlarged spleen (arrows; splenic index of 763 cm³). The impeded diffusion of the spleen is considered a normal finding. Coronal maximum intensity projection FDG-PET (c) illustrates splenic uptake greater than liver uptake. Arrowheads indicate nodal involvement at both sides of the diaphragm.

Table 3.

Whole-body MRI-DWI failures.

	PT ID	Pathologic subtype	Splenic index (cm³)	MRI T2-STIR	MRI-DWI	PET/CT	Type of error
False negative	37	Follicular lymphoma	178	–	Two nodules with restricted diffusion	Two corresponding focal lesions with increased uptake that resolved after treatment	MRI reader error
	75	Mantle cell lymphoma	409	–	–	Diffuse uptake > liver, that normalized after treatment	MRI intrinsic error
	77	Follicular lymphoma grade III	315	Hypo-intense nodule	Low signal b1000 and low signal ADC	Focal uptake that decreased after treatment	MRI reader error
False positive	12	DLBCL	1452	Enlarged	Enlarged	Uptake = liver	MRI intrinsic error
	27	Small lymphocytic lymphoma	771	Hyper-intense nodule	No restricted diffusion	No increased or decreased uptake	MRI reader error
	47	DLBCL	1436	Enlarged	Enlarged	Uptake < liver	MRI intrinsic error

DLBCL, diffusion large B-cell lymphoma; FU, follow-up; NA, not available; PT ID, patient identification.

Fig. 3.

Images of a 28-year-old man (no. 77) with stage III follicular lymphoma grade III illustrate MRI reader error. Coronal T2W-STIR (a) shows a 16-mm large hypointense splenic nodule (arrow). This nodule is hypodense at contrast-enhanced CT (b) with corresponding increased uptake at FDG-PET (e). Both b = 1000 s/mm² (c) and the ADC map (d) demonstrate low signal within this nodule (arrow).

Fig. 4.

Images of a 76-year-old man (no. 75) with stage III mantle cell lymphoma illustrate the limitations of splenic size measurements at MRI (MRI intrinsic error). Coronal T2W-STIR (a) shows a normal-sized spleen with diffuse increased uptake of FDG (b).

Fig. 5.

Images of an 81-year-old man (no. 27) with stage I small lymphocytic lymphoma illustrate a benign splenic lesion. Coronal T2-STIR (a) shows a 34 mm well-defined hyperintense nodule (arrow) with no impeded diffusion (b = 0 s/mm² (b); b = 1000 s/mm² (c); ADC map (d)). The axial FDG-PET (e) demonstrates homogenous normal FDG uptake.

Focal splenic lesions

There were 10 cases of focal splenic involvement identified by the reference standard (Table 4). On the T2W-STIR images nine out of 10 lesions were visible. One lesion was depicted on the DWI sequence, but not perceivable on the corresponding T2W-STIR images and remained undiagnosed during the original reading. In seven out of 10 cases, the splenic lesions demonstrated low T2W-STIR signal intensity relative to the surrounding splenic parenchyma (Fig. 3). In the majority of cases (6/10), the splenic lesions showed low signal intensity at DWI (b1000 s/mm²) relative to the surrounding splenic parenchyma. Two lesions (71 and 60 mm, respectively) demonstrated heterogeneous signal characteristics at all three sequences due to either central necrosis or central air related to direct extension of disease from the adjacent splenic flexure of the colon. Most focal lesions were not visible on the T1W images, except for the two previously described large lesions with marked heterogeneous appearances.

Table 4.

Focal splenic involvement.

PT ID	Age (years)	Pathologic subtype	MRI index (cm³)	MRI T2-STIR*	MRI-DWI b = 1000s/mm²*
33	45	Nodular sclerosing HL	1891	Low	Low
36	67	Follicular lymphoma grade III	1129	Low	Low
37	71	Follicular lymphoma	178	Iso	High
55	70	DLBCL	958	Heterogeneous	Heterogeneous
65	13	T-cell rich B NHL	1140	Low	Low
66	14	Large cell anaplastic lymphoma	763	Low	Low
70	28	Lymphocyte-rich HL	1044	Low	Iso
77	28	Follicular lymphoma grade III	315	Low	Low
87	39	DLBCL	541	Heterogeneous	Heterogeneous
107	11	Nodular sclerosing HL	257	Low	Low

Signal intensity relative to the surrounding splenic parenchyma.

DLBCL, diffuse large B-cell lymphoma.

Discussion

This study shows that whole-body MRI, including T1W, T2W-STIR, and DWI sequences, is reasonably accurate in the detection of splenic involvement in newly diagnosed lymphoma, although in three out of 21 patients splenic involvement was missed.

In addition, our results showed that most focal splenic lymphomatous lesions exhibit a low T2 signal and low diffusivity compared to the surrounding, normal-appearing spleen. Although low T2 signal intensity in focal splenic involvement has been reported in the literature (22,23), most commonly lymphomatous nodules tend to be isointense to splenic parenchyma on conventional MRI sequences (22,24,25). The different signal intensities detected in lymphomatous nodules probably reflects the varying content of fibrosis, edema and necrosis. The major advantage of detecting low intensity nodules at T2W images is that this may help to distinguish lymphomatous nodules from most benign focal lesions (e.g. hemangioma, hamartoma) (25). Punwani et al. reported an increase in detection rate with the addition of dynamic contrast-enhanced MRI compared to T2W images for focal involvement in pediatric Hodgkin's lymphoma (23). The increasing conspicuity at post- gadolinium images is probably related to the hypovascular nature of the lymphomatous nodules compared to the spleen (22,24,25).

Our measures of diagnostic performance apply for a whole-body MRI-DWI protocol that cannot contain similar dedicated, time-consuming sequences as used in an upper abdomen MRI scan protocol. Our whole-body MRI-DWI protocol will take at least 45 min and additional sequences for the upper abdomen to the current protocol are therefore practically impossible. However, several technical improvements may increase the diagnostic yield of whole-body MRI in the detection of lymphomatous splenic involvement. First, the consistent use of (whole-body) surface coil for signal reception – whether or not combined with acquisition at higher main field strength (3 T) – will increase the received signal to noise ratio and/or spatial resolution. Second, newly developed respiratory motion compensation techniques may decrease motion artifacts during free breathing without increasing acquisition time (26). However, three out of six discrepancies between whole-body MRI-DWI and our reference standard were due to the suboptimal criterion of splenic size used with MRI versus the size-independent functional information provided by FDG-PET. The problem of using size for splenic assessment has previously been reported (8,27); mildly-moderately enlarged spleens can be without lymphomatous infiltration and normal-sized spleens can have diffuse infiltration. Rini et al. (27) compared FDG-PET with contrast-enhanced CT for evaluation of the spleen in staging of lymphoma in seven patients and used histopathology as reference standard. The reported diagnostic accuracies of FDG-PET and CT were 100% and 57%, respectively. In this study, two patients with enlarged spleens (both > 725 cm³) did not contain lymphomatous infiltration. New functional MRI tools, e.g. by using ultrasmall superparamagnetic iron oxide (USPIO) agents, may further increase the diagnostic yield of whole-body MRI in this setting (28). Unfortunately, clinical implementation of USPIO agents is impeded due to lack of their availability.

Several study limitations need to be addressed. First, there was a lack of histopathological reference standard in all except one patient. Although the diagnosis of lymphoma was histologically confirmed in each patient, histological investigation of the spleen is no longer part of standard care. Second, there was heterogeneity of our image quality related to our multicentre design of our study. On the other hand, this reflects daily clinical practice and increases the generalizability of our results. Third, we assessed the splenic involvement for all types of lymphoma although splenic involvement is especially relevant in Hodgkin's lymphoma. Fourth, although we included over 100 patients, we acknowledge that the group with splenic involvement is relatively limited.

In conclusion, our results indicate that whole-body MRI-DWI is reasonable accurate in detecting splenic involvement in lymphoma. Performing whole-body MRI at higher field strength, applying more sophisticated respiratory motion compensation and using new functional MRI techniques may further improve the focal splenic detection, although the diffuse lymphomatous infiltration in normal sized spleens will remain a diagnostic challenge.

Footnotes

Acknowledgements

We would like to thank Willem PThM Mali, MD, PhD and Hugo J Adams, MD for their contribution to this study.

Conflict of interest

The authors declare that there is no conflict of interest.

Funding

This work was supported by the Dutch Organisation of Health Research and Development (ZonMw) Program for Health Care Efficiency Research (grant no. 80-82310-98-08012) and Kinderen Kankervrij (Children cancer free; project number 87). Data collection, data analysis and interpretation of data, writing of the paper and the decision to submit were left to theauthors' discretion and were not influenced by ZonMw orKIKA.

References

Rathore

Kadin

. Hodgkin's lymphoma therapy: past, present and future. Expert Opin Pharmacother 2010; 11: 2891–2906.

Rueffer

Sieber

Stemberg

. German Hodgkin' Lymphoma Study Group (GHSG). Spleen involvement in Hodgkin's lymphoma: assessment and risk profile. Ann Hematol 2003; 82: 390–396.

Munker

Stengel

Stäbler

. Diagnostic accuracy of ultrasound and computed tomography in the staging of Hodgkin's disease: verification by laparotomy in 100 cases. Cancer 1995; 76: 1460–1466.

Strijk

Wagener

Bogman

. The spleen in Hodgkin disease: diagnostic value of CT. Radiology 1985; 154: 753–757.

Strijk

Boetes

Bogman

. The spleen in non-Hodgkin lymphoma: diagnostic value of computed tomography. Acta Radiol 1987; 28: 139–144.

Ansell

Armitage

. Non-Hodgkin lymphoma: diagnosis and treatment. Mayo Clin Proc 2005; 80: 1087–1097.

Sandrasegaran

Robinson

Selby

. Staging of lymphoma in adults. Clin Radiol 1994; 49: 149–161.

Saboo

Krajewski

O'Regan

. Spleen in haematological malignancies: spectrum of imaging findings. Br J Radiol 2012; 85: 81–92.

de Jong

van Ufford

Baarslag

. CT and 18F-FDG PET for noninvasive detection of splenic involvement in patients with malignant lymphoma. Am J Roentgenol 2009; 192: 745–753.

10.

Juweid

. FDG-PET/CT in lymphoma. Methods Mol Biol 2011; 727: 1–19.

11.

la Fougère

Hundt

Bröckel

. Value of PET/CT versus PET and CT performed as separate investigations in patients with Hodgkin's disease and non-Hodgkin's lymphoma. Eur J Nucl Med Mol Imaging 2006; 33: 1417–1425.

12.

Salaun

Gastinne

Bodet-Milin

. Analysis of 18F-FDG PET diffuse bone marrow uptake and splenic uptake in staging of Hodgkin's lymphoma: a reflection of disease infiltration or just inflammation? Eur J Nucl Med Mol Imaging 2009; 36: 1813–1821.

13.

Liu

. Clinical significance of diffusely increased splenic uptake on FDG-PET. Nucl Med Commun 2009; 30: 763–769.

14.

Punwani

Taylor

Bainbridge

. Pediatric and adolescent lymphoma: comparison of whole-body STIR half-Fourier RARE MR imaging with an enhanced PET/CT reference for initial staging. Radiology 2010; 255: 182–190.

15.

van Ufford

Kwee

Beek

. Newly diagnosed lymphoma: initial results with whole-body T1-weighted, STIR, and diffusion-weighted MRI compared with 18F-FDG PET/CT. Am J Roentgenol 2011; 196: 662–669.

16.

Lin

Luciani

Itti

. Whole-body diffusion-weighted magnetic resonance imaging with apparent diffusion coefficient mapping for staging patients with diffuse large B-cell lymphoma. Eur Radiol 2010; 20: 2027–2038.

17.

Chan

Zhang

. Whole-body diffusion-weighted imaging: the added value to whole-body MRI at initial diagnosis of lymphoma. Am J Roentgenol 2011; 197: W384–W391.

18.

Kwee

Vermoolen

Akkerman

. Whole-body MRI, including diffusion-weighted imaging, for staging lymphoma: comparison with CT in a prospective multicenter study. J Magn Reson Imaging 2014; 40: 26–36.

19.

Littooij

Kwee

Barber

. Whole-body MRI for initial staging of paediatric lymphoma: prospective comparison to a FDG-PET/CT-based reference standard. Eur Radiol 2014; 24: 1153–1165.

20.

Mendenhall

Cantor

Williams

. With modern imaging techniques, is staging laparotomy necessary in pediatric Hodgkin's disease? A Pediatric Oncology Group study. J Clin Oncol 1993; 11: 2218–2225.

21.

Weiler-Sagie

Bushelev

Epelbaum

. (18)F-FDG avidity in lymphoma readdressed: a study of 766 patients. J Nucl Med 2010; 51: 25–30.

22.

Luna

Ribes

Caro

. MRI of focal splenic lesions without and with dynamic gadolinium enhancement. Am J Roentgenol 2006; 186: 1533–1547.

23.

Punwani

Cheung

Skipper

. Dynamic contrast-enhanced MRI improves accuracy for detecting focal splenic involvement in children and adolescents with Hodgkin disease. Pediatr Radiol 2013; 43: 941–949.

24.

Ito

Mitchell

Honjo

. MR imaging of acquired abnormalities of the spleen. Am J Roentgenol 1997; 168: 697–702.

25.

Palas

Matos

Ramalho

. The Spleen Revisited: An Overview on Magnetic Resonance Imaging. Radiol Res Pract 2013; 2013: 219297.

26.

Takahara

Kwee

van Leeuwen

. Diffusion-weighted magnetic resonance imaging of the liver using tracking only navigator echo: feasibility study. Invest Radiol 2010; 45: 57–63.

27.

Rini

Leonidas

Tomas

. 18F-FDG PET versus CT for evaluating the spleen during initial staging of lymphoma. J Nucl Med 2003; 44: 1072–1074.

28.

Harisinghani

Saini

Weissleder

. Splenic imaging with ultrasmall superparamagnetic iron oxide ferumoxtran-10 (AMI-7227): preliminary observations. J Comput Assist Tomogr 2001; 25: 770–776.