Differences between... Neuroblastoma and neuroblastoma radiology

Reading File

The question asks about differences between neuroblastoma (the tumor itself) and neuroblastoma radiology (its imaging features). This is a broad + visual question, so I'll run parallel searches across the library.

Finding Sources

Good. I have the key sources. Now let me read the most comprehensive ones in parallel - the radiology textbook, Robbins pathology, and Campbell Walsh Urology sections.

Reading File

I have rich content from the radiology and pathology textbooks. Now let me read more of the radiology section (imaging features) and the full Robbins clinical features section.

Reading File

I now have excellent content from both the radiology textbook (Grainger & Allison) and Robbins pathology. Let me now fetch the key images and also read a bit more of the radiology imaging section (CT, MRI, MIBG).

Reading File

Loading Image

I now have comprehensive content from both the pathology and radiology textbooks. Let me also get the imaging figures from the radiology textbook.

Reading File

I now have all the information needed for a comprehensive answer. Let me also fetch the radiology imaging figure (ultrasound) and the CT figure.

Loading Image

I now have all I need to write a thorough, well-organized answer. Here is the full comparison:

Neuroblastoma: Pathology vs. Radiology

The question asks about the differences between neuroblastoma as a disease entity (its biology, pathology, clinical behavior) and neuroblastoma radiology (its imaging findings across modalities). These are two different lenses on the same tumor - here is a structured breakdown of each.

1. THE DISEASE - What is Neuroblastoma?

Origin & Epidemiology

Neuroblastoma (NB) arises from primordial neural crest cells that populate the sympathetic ganglia and adrenal medulla. It is the second most common solid malignancy of childhood (after brain tumors), accounting for 7-10% of all pediatric neoplasms and up to 50% of malignancies in infancy. Median age at diagnosis is 2 years; 90% occur before age 5. Most cases are sporadic; 1-2% are familial (autosomal dominant, linked to germline ALK mutations).

Robbins & Kumar Basic Pathology, p. 144

Primary Sites

Site	Frequency
Adrenal glands	40-48%
Extra-adrenal retroperitoneum / paravertebral	25%
Posterior mediastinum	15-16%
Neck	3%
Pelvis	3%

About 50% of patients have metastatic disease at diagnosis, most commonly to skeleton, bone marrow, lymph nodes, and liver.

Grainger & Allison's Diagnostic Radiology, p. 1961

Gross & Histological Pathology

Neuroblastoma vs Ganglioneuroma histology

(A) Neuroblastoma: small primitive cells with dark nuclei, scant cytoplasm, Homer-Wright pseudorosettes (tumor cells concentrically arranged around neuropil). (B) Ganglioneuroma (mature end of spectrum): large ganglion cells with vesicular nuclei and abundant cytoplasm (arrow), with Schwann cells in the background stroma.

Robbins & Kumar Basic Pathology, Fig. 4.38

Key histological features:

Sheets of small round blue cells, dark nuclei, scant cytoplasm
Homer-Wright pseudorosettes - tumor cells arranged around central neuropil (no true lumen)
Faint eosinophilic neuropil in background
Immunohistochemistry positive for neuron-specific enolase (NSE)
EM shows membrane-bound catecholamine-containing secretory granules
Maturation spectrum: Neuroblastoma → Ganglioneuroblastoma → Ganglioneuroma (Schwannian stroma + gangliocytic differentiation = better prognosis)

Molecular Biology & Prognostic Factors

Factor	Detail
MYCN amplification	Present in 25-30%; most important genetic abnormality; automatically renders tumor "high risk"; EFS 50% vs 90% without amplification
ALK mutations	Somatic gain-of-function in 8-10% of sporadic cases; marker of adverse prognosis
DNA ploidy	Hyperdiploid (whole chromosome gains) = better prognosis than diploid
Age	< 18 months = far better outlook; most are stage 1, 2, or 4S
1p and 11q deletions	Associated with adverse outcome
17q gain	Adverse outcome

Staging - International Neuroblastoma Staging System (INSS)

Stage 1-3: Local/regional disease
Stage 4: Distant metastases
Stage 4S (Special): Localized tumor + spread limited to liver, skin, and bone marrow (without bone involvement) - excellent prognosis and often undergoes spontaneous regression in infants. The biologic basis is not fully clear.

Clinical Features

Young children: protuberant abdomen, fever, weight loss
Older children: hepatomegaly, ascites, bone pain from metastases
Horner syndrome with cervical/thoracic NB (stellate ganglion involvement)
Proptosis from orbital metastases ("raccoon eyes")
Catecholamine excess (~90% of tumors produce catecholamines): elevated urinary VMA (vanillylmandelic acid) and HVA (homovanillic acid) - key diagnostic biochemical finding; however hypertension is less frequent than in pheochromocytoma
VIP production: intractable watery diarrhea
MEI (myoclonic encephalopathy of infancy): a paraneoplastic syndrome associated with NB
Neonates: deep blue subcutaneous metastases ("blueberry muffin baby")
Dumb-bell syndrome: extradural extension via foramina (especially thoracic NB) causing spinal cord compression

2. THE RADIOLOGY - Imaging of Neuroblastoma

Radiology is a separate domain from the pathology - it defines how the tumor appears on each imaging modality and guides staging and surgical planning.

Plain Radiograph

Calcified retroperitoneal or thoracic mass
Erosion of vertebral pedicles suggests intraspinal extension
Skeletal metastases: ill-defined bone destruction, lytic/moth-eaten/permeative areas interspersed with sclerotic trabeculae
Periosteal new bone formation parallel to shaft (malignant but non-specific)
Most common skeletal sites: skull and metaphyses of long bones (humerus, femur)
Grainger & Allison's, p. 1962

Ultrasound (US)

The initial diagnostic modality for abdominal NB and the primary tool for fetal/neonatal masses.

Neonatal neuroblastoma - ultrasound and CT evolution

Spontaneous regression of neonatal adrenal NB: (A) initial cystic mass, (B) becoming partially solid at 6 weeks, (C) much smaller and hyperechoic with calcifications at 31 weeks, (D) coronal CT showing near-complete regression at 45 weeks.

Grainger & Allison's Diagnostic Radiology, Fig. 75.22

Key US features:

Feature	Description
Retroperitoneal location	Anterior displacement of aorta and IVC (confirms retroperitoneal origin, distinguishing from renal origin in Wilms)
Echogenicity	Variable; hyperechoic foci = calcifications
Vascularisation	Rich on Colour Doppler (vs. absent in adrenal haemorrhage)
Hypoechoic areas	Haemorrhage / necrosis
Neonatal	Can be cystic (55%), solid (17.5%), or heterogeneous (27.5%)

Neonatal differential on US: NB vs adrenal haemorrhage vs congenital adrenal hyperplasia vs subdiaphragmatic extralobar pulmonary sequestration. A mass increasing in size or persisting beyond 48 weeks favors NB; a mass regressing, becoming heterogeneous, and showing late calcifications favors adrenal haemorrhage.

CT Scan

The main modality for staging the primary tumor (chest, abdomen, pelvis with IV contrast).

CT Feature	Description
Appearance	Large, heterogeneous mass
Calcification	Present in up to 85% of cases (very characteristic)
Low-attenuation areas	Necrosis or haemorrhage
Enhancement	Mild, heterogeneous
Vascular encasement	NB often encases rather than displaces vessels (key difference from Wilms' tumor)
Spinal extension	Suspected but not fully evaluable - MRI required for dumb-bell tumors
Skull metastases	Spheno-orbital region; permeative bone destruction with spiculated changes, scalp soft-tissue extension

MRI

Preferred for paraspinal tumors and intraspinal extension.

MRI Feature	Description
T1 signal	Prolonged (low); high T1 = haemorrhage
T2 signal	Prolonged (high)
Gadolinium enhancement	Heterogeneous
Bone marrow involvement	Low T1, high T2 signal in marrow
Dumb-bell tumors	Best modality for intraspinal/extradural extension

Radionuclide Imaging - ¹²³I-mIBG Scintigraphy

This is the most specific nuclear medicine tool for NB.

mIBG (metaiodobenzylguanidine) is an analogue of noradrenaline, specifically taken up and stored in sympathetic-derived tumors
Used for: characterization of mass, localization (especially in MEI where primary may be occult), evaluation of metastatic disease, and follow-up
~20% of primary lesions do not take up mIBG - these cases require alternative nuclear medicine studies
⁹⁹ᵐTc-MDP bone scintigraphy is also used to evaluate skeletal metastases
Grainger & Allison's, p. 1964

3. KEY DIFFERENCES SUMMARY: Pathology vs. Radiology

Domain	Neuroblastoma (Pathology/Disease)	Neuroblastoma Radiology
Defines	What the tumor IS (biology, molecular, histology)	What the tumor LOOKS LIKE on imaging
Tools	Histology, IHC, molecular genetics, biochemistry	X-ray, US, CT, MRI, MIBG
Calcification	Not a primary pathologic feature	Present in up to 85% on CT - hallmark
Vascular relationship	Invasion/encasement histologically	Encasement of aorta/IVC/renal vessels on CT/MRI (vs Wilms' which displaces)
Maturation	Spectrum: NB → GNB → GN (Schwann cells, ganglion cells = better prognosis)	All three may look similar radiologically; cannot reliably distinguish by imaging alone
Staging basis	INSS stages 1-4, 4S based on surgical/pathological findings	Image-Defined Risk Factors (IDRFs) - SIOPEN/INRG system defines surgical risk from imaging pre-op
Regression	Spontaneous regression is a known biological behavior (especially < 18 months, 4S)	Serial US/CT shows shrinkage; key to follow neonatal masses before intervening
Diagnosis confirmation	Positive bone marrow + elevated urinary catecholamine metabolites sufficient	No single imaging feature is pathognomonic
Spinal cord risk	Extradural extension histologically	MRI mandatory for paraspinal masses to evaluate cord compression
Bone marrow involvement	Histological infiltration by neuroblasts	Abnormal T1/T2 signal on MRI; bone scan or mIBG for scintigraphic detection

Sources:

Robbins & Kumar Basic Pathology, p. 144-146
Grainger & Allison's Diagnostic Radiology, p. 1961-1965
Campbell Walsh Wein Urology, Chapter on Neuroblastoma

Differences between... Neuroblastoma and nephroblastoma radiology

Reading File

This is a broad + visual question. Running parallel searches for both tumors in the radiology textbook.

Finding Sources

Now let me find the nephroblastoma/Wilms section in the radiology textbook and read it along with the neuroblastoma radiology section (already known from last session) and Robbins pathology for both.

Finding Sources

Let me search for the Wilms/nephroblastoma radiology section directly in Grainger & Allison and also search paragraphs for Wilms imaging features.

Finding Sources

Excellent - found the Grainger & Allison Wilms Tumour section. Let me now read it in parallel with the Robbins Wilms section and the neuroblastoma radiology section from the previous session.

Reading File

Excellent content from both tumors in Grainger & Allison. Let me now read more of the Wilms imaging section (CT, MRI, nuclear medicine from Grainger) and also the Robbins morphology/clinical features continuation.

Reading File

I now have all the data needed. Let me fetch the key radiology images.

Loading Image

I now have comprehensive, well-sourced data to write the full comparison. Here is the detailed answer:

Neuroblastoma vs. Nephroblastoma (Wilms Tumor) - Radiology

These are the two most common abdominal malignancies of childhood and their differentiation on imaging is one of the classic pediatric radiology challenges. The key distinction starts with origin: neuroblastoma arises from the adrenal gland/sympathetic chain (extra-renal), while nephroblastoma arises within the kidney itself.

Background: The Two Tumors at a Glance

Feature	Neuroblastoma (NB)	Nephroblastoma / Wilms Tumor (WT)
Origin	Neural crest cells - adrenal medulla, sympathetic ganglia	Metanephric blastema - kidney
Peak age	2 years (90% < 5 yrs)	3 years (peak); 2-5 yrs
Gender	Slight male predominance	Equal (M = F)
Bilateral	Rare (< 5%)	~10% (two-thirds synchronous)
Biochemical marker	Urine catecholamines (VMA, HVA) elevated in ~90%	None specific
Associated syndromes	Familial ALK mutations	WAGR, Denys-Drash, Beckwith-Wiedemann
Metastases at diagnosis	~50%	~10% (stage IV)

Radiological Differences by Modality

1. Plain Radiograph

Feature	Neuroblastoma	Nephroblastoma
Calcification	Common - stippled, irregular, amorphous calcifications visible in ~50% on plain film	Rare - calcification unusual (<10%)
Mass location	Retroperitoneal, can be thoracic, cervical	Always intrarenal - right upper quadrant or left upper quadrant
Vertebral pedicle erosion	Can occur (dumb-bell extension into spinal canal)	Not a feature
Skeletal metastases	Common - lytic/moth-eaten areas with periosteal new bone formation in long bones and skull	Less common; when present - lytic lesions

2. Ultrasound (US) - First-Line in Both

Neuroblastoma on US:

Retroperitoneal mass, typically suprarenal or paraaortic
Variable echogenicity; hyperechoic foci = calcifications (very common)
Hypoechoic areas from haemorrhage/necrosis
Rich vascularisation on Colour Doppler
Anterior displacement of the aorta and IVC - confirms retroperitoneal origin
Does NOT arise from kidney - kidney is displaced but separate, with its own capsule visible
In neonates: can be cystic (55%), solid (17.5%), or heterogeneous (27.5%)

Nephroblastoma on US:

Large heterogeneous mass arising from the upper right kidney with inferior displacement of the lower renal pole - the key US finding that localizes the tumor to the kidney.

Grainger & Allison's Diagnostic Radiology, Fig. 72.50
Intrarenal - the mass arises FROM the kidney
Large, with mixture of solid hyperechoic and cystic areas; cystic components often predominate
Normal native renal tissue may be visible, stretched at the periphery ("claw sign")
Renal vein and IVC must be assessed for tumour thrombus (can extend to right atrium) - a critical US finding unique to Wilms
Movement of the mass separate from the liver suggests no direct invasion (important surgical planning)
Calcification is NOT a typical finding

Key US differential point: If you can identify normal kidney tissue separate from the mass - it's neuroblastoma displacing the kidney. If the mass appears to arise FROM within the kidney - it's nephroblastoma.

3. CT Scan

CT Feature	Neuroblastoma	Nephroblastoma
Origin	Extra-renal (adrenal/retroperitoneal) - kidney displaced, beak-sign may show adjacent kidney	Intrarenal - kidney origin; "claw" of residual renal parenchyma wraps around tumor
Calcification	Present in up to 85% - highly characteristic	Unusual (< 10%) - when present, may suggest prior hemorrhage
Vascular relationship	Encases vessels (aorta, IVC, celiac, mesenteric) without displacing them - "vessel encasement" sign	Displaces vessels; IVC/renal vein invaded by tumor thrombus (not encased)
Tumour thrombus	Rare	Classic feature - IVC thrombus in up to 10%; can extend to right atrium
Enhancement	Mild, heterogeneous (areas of necrosis/hemorrhage/cystic change)	Heterogeneous; enhances less than normal renal parenchyma
Crossmidline	Frequently crosses midline (especially large retroperitoneal masses)	Less commonly crosses midline
Contralateral kidney	Assess for bilateral involvement (rare)	Must assess - 10% bilateral; nephroblastomatosis (nephrogenic rests) in contralateral kidney
Lymph nodes	Paraaortic, celiac, mesenteric	Para-aortic, regional; peritoneal invasion in stage III
Skeletal metastases	Common - skull (spheno-orbital), long bone metaphyses	Stage IV (10%) - lungs first, then liver
Pulmonary metastases	Less common	Present in stage IV; chest CT value controversial vs. plain CXR
Spinal canal extension	Characteristic - dumb-bell through foramina; CT can suggest but MRI required to evaluate	Not a feature

4. MRI

MRI Feature	Neuroblastoma	Nephroblastoma
T1 signal	Prolonged (low); high T1 = hemorrhage	Hypointense on T1 (similar)
T2 signal	Prolonged (high)	Variably hyperintense on T2
Gadolinium	Heterogeneous enhancement; required for spinal canal and vessel assessment	Heterogeneous, often poor enhancement; required for contralateral kidney (nephroblastomatosis), biopsy guidance
Bone marrow	Abnormal signal (low T1, high T2) in metastatic disease	Not a primary concern
Spinal canal	MRI is mandatory for any paraspinal mass - evaluate extradural/cord extension	Not indicated for spinal canal
Gold standard for bilateral	Not applicable	MRI is gold standard for bilateral disease assessment (Fig. 72.51)

Bilateral Wilms tumor on coronal MRI - large right-sided mass and a smaller left-sided mass that could easily be overlooked. MRI is now the gold standard for bilateral Wilms assessment.

Grainger & Allison's Diagnostic Radiology, Fig. 72.51

5. Nuclear Medicine

Modality	Neuroblastoma	Nephroblastoma
¹²³I-mIBG scintigraphy	Essential - mIBG (noradrenaline analogue) is specifically taken up by sympathetic-derived tumors; used for primary characterization, metastasis mapping, and follow-up. ~20% of NB are mIBG-negative	Not used - no catecholamine uptake
⁹⁹ᵐTc-MDP bone scan	Used to detect skeletal metastases (alongside mIBG)	Less commonly used
PET/CT	Used in mIBG-negative cases	Not standard
Chest CT vs CXR	For pulmonary metastases staging	Standard staging uses CXR; chest CT controversial as small CT-only metastases have uncertain prognostic significance

mIBG scintigraphy is the single biggest nuclear medicine differentiator - it is pathognomonic for neuroblastoma (sympathetic origin) and has no role in nephroblastoma.

The Single Most Important Differentiating Imaging Sign

Criterion	Neuroblastoma	Nephroblastoma
Relationship to kidney	Displaces the kidney - kidney is separate, intact, with its own capsule	Arises from the kidney - residual renal tissue forms a "claw" or "beak" around the mass
Calcification on CT	Very common (85%), coarse/amorphous	Rare (<10%)
Vessel relationship	Encases aorta/IVC without thrombus	IVC tumor thrombus (displaces, not encases)
Crossing midline	Common	Less common
mIBG uptake	Yes (80%)	No
Catecholamines	Urine VMA/HVA elevated	Normal

Staging Systems (Radiologically Relevant)

Neuroblastoma uses the International Neuroblastoma Risk Group (INRG) system based on Image-Defined Risk Factors (IDRFs) - pre-operative imaging defines the surgical risk by assessing whether the tumor encases major vessels, invades adjacent structures, or extends into the spinal canal.

Nephroblastoma uses the NWTSG staging (North American) or SIOP (European) - both are primarily surgical/pathological staging systems, though imaging (CT/MRI) is required to identify lymphadenopathy, peritoneal deposits, liver lesions, and contralateral kidney involvement pre-operatively.

Sources:

Grainger & Allison's Diagnostic Radiology, p. 1861 (Wilms Tumour) and p. 1961-1964 (Neuroblastoma)
Robbins & Kumar Basic Pathology, p. 144-148 (Neuroblastoma & Wilms Tumor)

Make a quick reference PDF on pediatric abdominal mass differential diagnosis

Reading File

Running Command

mkdir -p /home/daytona/workspace/peds-abdominal-mass && ls /home/daytona/workspace/

Writing File

~/peds-abdominal-mass/generate_pdf.py

from reportlab.lib.pagesizes import A4
from reportlab.lib import colors
from reportlab.lib.units import cm, mm
from reportlab.platypus import (
    SimpleDocTemplate, Paragraph, Spacer, Table, TableStyle,
    HRFlowable, KeepTogether
)
from reportlab.lib.styles import getSampleStyleSheet, ParagraphStyle
from reportlab.lib.enums import TA_LEFT, TA_CENTER, TA_RIGHT
from reportlab.platypus import PageBreak

OUTPUT = "/home/daytona/workspace/peds-abdominal-mass/Pediatric_Abdominal_Mass_DDx.pdf"

# ── Colours ──────────────────────────────────────────────────────────────────
NAVY      = colors.HexColor("#1a3a5c")
TEAL      = colors.HexColor("#1e7f8e")
AMBER     = colors.HexColor("#e67e22")
RED       = colors.HexColor("#c0392b")
GREEN     = colors.HexColor("#27ae60")
PURPLE    = colors.HexColor("#8e44ad")
LTGRAY    = colors.HexColor("#f5f6fa")
MIDGRAY   = colors.HexColor("#dfe3e8")
DARKGRAY  = colors.HexColor("#4a4a5a")
WHITE     = colors.white
BLACK     = colors.black

# ── Styles ────────────────────────────────────────────────────────────────────
styles = getSampleStyleSheet()

def S(name, **kw):
    """Create a named ParagraphStyle extending 'Normal'."""
    return ParagraphStyle(name, parent=styles["Normal"], **kw)

Title      = S("MyTitle",    fontSize=22, textColor=WHITE,    fontName="Helvetica-Bold",
               alignment=TA_CENTER, leading=28, spaceAfter=4)
Subtitle   = S("MySub",      fontSize=11, textColor=WHITE,    fontName="Helvetica",
               alignment=TA_CENTER, leading=16)
SectionH   = S("SH",         fontSize=12, textColor=WHITE,    fontName="Helvetica-Bold",
               alignment=TA_LEFT,   leading=16, leftIndent=6)
Body       = S("Body",       fontSize=8.5, textColor=DARKGRAY, fontName="Helvetica",
               leading=12, spaceAfter=2)
Bold       = S("Bold",       fontSize=8.5, textColor=BLACK,    fontName="Helvetica-Bold",
               leading=12)
SmallBold  = S("SmallBold",  fontSize=7.5, textColor=DARKGRAY, fontName="Helvetica-Bold",
               leading=11)
Small      = S("Small",      fontSize=7.5, textColor=DARKGRAY, fontName="Helvetica",
               leading=11)
TblHdr     = S("TblHdr",     fontSize=8,   textColor=WHITE,    fontName="Helvetica-Bold",
               alignment=TA_CENTER, leading=11)
TblCell    = S("TblCell",    fontSize=7.5, textColor=DARKGRAY, fontName="Helvetica",
               leading=11, leftIndent=3)
TblCellB   = S("TblCellB",   fontSize=7.5, textColor=BLACK,    fontName="Helvetica-Bold",
               leading=11, leftIndent=3)
FooterS    = S("Footer",     fontSize=7,   textColor=DARKGRAY, fontName="Helvetica-Oblique",
               alignment=TA_CENTER)
RedAlert   = S("RedAlert",   fontSize=8,   textColor=RED,      fontName="Helvetica-Bold",
               leading=12)
GreenText  = S("GreenText",  fontSize=8,   textColor=GREEN,    fontName="Helvetica-Bold",
               leading=12)

# ── Page canvas (header/footer) ───────────────────────────────────────────────
PAGE_W, PAGE_H = A4
MARGIN = 1.5*cm

def on_page(canvas, doc):
    canvas.saveState()
    # top stripe
    canvas.setFillColor(NAVY)
    canvas.rect(0, PAGE_H - 1.2*cm, PAGE_W, 1.2*cm, fill=1, stroke=0)
    canvas.setFillColor(WHITE)
    canvas.setFont("Helvetica-Bold", 8)
    canvas.drawString(MARGIN, PAGE_H - 0.82*cm,
                      "PEDIATRIC ABDOMINAL MASS — QUICK REFERENCE")
    canvas.setFont("Helvetica", 8)
    canvas.drawRightString(PAGE_W - MARGIN, PAGE_H - 0.82*cm,
                           f"Page {doc.page}")
    # bottom stripe
    canvas.setFillColor(MIDGRAY)
    canvas.rect(0, 0, PAGE_W, 0.9*cm, fill=1, stroke=0)
    canvas.setFillColor(DARKGRAY)
    canvas.setFont("Helvetica-Oblique", 6.5)
    canvas.drawCentredString(PAGE_W/2, 0.32*cm,
        "For educational use only. Clinical decisions require full patient assessment.")
    canvas.restoreState()

def on_first_page(canvas, doc):
    on_page(canvas, doc)

# ── Helpers ───────────────────────────────────────────────────────────────────
def section_banner(text, color=TEAL):
    """Coloured banner that acts as a section header."""
    data = [[Paragraph(text, SectionH)]]
    t = Table(data, colWidths=[PAGE_W - 2*MARGIN])
    t.setStyle(TableStyle([
        ("BACKGROUND", (0,0), (-1,-1), color),
        ("TOPPADDING",    (0,0), (-1,-1), 5),
        ("BOTTOMPADDING", (0,0), (-1,-1), 5),
        ("LEFTPADDING",   (0,0), (-1,-1), 8),
        ("RIGHTPADDING",  (0,0), (-1,-1), 8),
        ("ROUNDEDCORNERS", [4]),
    ]))
    return t

def colored_table(headers, rows, col_widths, hdr_color=NAVY, alt_color=LTGRAY):
    """Generic coloured table."""
    hdr_row = [Paragraph(h, TblHdr) for h in headers]
    data = [hdr_row]
    for i, row in enumerate(rows):
        data.append([Paragraph(str(c), TblCell) for c in row])
    t = Table(data, colWidths=col_widths, repeatRows=1)
    style = [
        ("BACKGROUND",    (0,0),  (-1,0),  hdr_color),
        ("ROWBACKGROUNDS",(0,1),  (-1,-1), [WHITE, alt_color]),
        ("GRID",          (0,0),  (-1,-1), 0.4, MIDGRAY),
        ("TOPPADDING",    (0,0),  (-1,-1), 4),
        ("BOTTOMPADDING", (0,0),  (-1,-1), 4),
        ("LEFTPADDING",   (0,0),  (-1,-1), 4),
        ("RIGHTPADDING",  (0,0),  (-1,-1), 4),
        ("VALIGN",        (0,0),  (-1,-1), "TOP"),
    ]
    t.setStyle(TableStyle(style))
    return t

def mini_box(title, lines, bg=LTGRAY, title_color=NAVY):
    """Small info box with a title bar."""
    content = [[Paragraph(title, ParagraphStyle("BT", parent=styles["Normal"],
                fontSize=8, textColor=WHITE, fontName="Helvetica-Bold", leading=11))]]
    body_rows = [[Paragraph(l, Small)] for l in lines]
    t_title = Table(content, colWidths=[PAGE_W - 2*MARGIN])
    t_title.setStyle(TableStyle([
        ("BACKGROUND", (0,0), (-1,-1), title_color),
        ("TOPPADDING",    (0,0), (-1,-1), 4),
        ("BOTTOMPADDING", (0,0), (-1,-1), 4),
        ("LEFTPADDING",   (0,0), (-1,-1), 8),
    ]))
    t_body = Table(body_rows, colWidths=[PAGE_W - 2*MARGIN])
    t_body.setStyle(TableStyle([
        ("BACKGROUND", (0,0), (-1,-1), bg),
        ("TOPPADDING",    (0,0), (-1,-1), 3),
        ("BOTTOMPADDING", (0,0), (-1,-1), 3),
        ("LEFTPADDING",   (0,0), (-1,-1), 10),
        ("GRID",          (0,0), (-1,-1), 0.2, MIDGRAY),
    ]))
    return [t_title, t_body]

# ── Document build ────────────────────────────────────────────────────────────
doc = SimpleDocTemplate(
    OUTPUT,
    pagesize=A4,
    leftMargin=MARGIN, rightMargin=MARGIN,
    topMargin=2.0*cm, bottomMargin=1.5*cm,
    title="Pediatric Abdominal Mass - Quick Reference",
    author="Orris Medical",
    subject="Pediatric Radiology - Differential Diagnosis"
)

story = []
SP = lambda n=4: Spacer(1, n)

# ══════════════════════════════════════════════════════
#  COVER BLOCK
# ══════════════════════════════════════════════════════
cover_data = [[
    Paragraph("PEDIATRIC ABDOMINAL MASS", Title),
    Paragraph("Differential Diagnosis · Quick Reference Card", Subtitle),
    Paragraph("Radiology · Pathology · Key Distinguishing Features", Subtitle),
]]
cover = Table(cover_data, colWidths=[PAGE_W - 2*MARGIN])
cover.setStyle(TableStyle([
    ("BACKGROUND",    (0,0), (-1,-1), NAVY),
    ("TOPPADDING",    (0,0), (-1,-1), 18),
    ("BOTTOMPADDING", (0,0), (-1,-1), 18),
    ("LEFTPADDING",   (0,0), (-1,-1), 12),
    ("RIGHTPADDING",  (0,0), (-1,-1), 12),
    ("ROUNDEDCORNERS", [6]),
]))
story.append(cover)
story.append(SP(10))

# ══════════════════════════════════════════════════════
#  SECTION 1 - OVERVIEW TABLE (by age)
# ══════════════════════════════════════════════════════
story.append(section_banner("1. AGE-BASED DIFFERENTIAL OVERVIEW", NAVY))
story.append(SP(4))

age_headers = ["Age Group", "Most Likely Diagnosis", "Other Diagnoses to Consider"]
age_rows = [
    ["Neonate\n(0–1 month)",
     "Hydronephrosis (most common)\nMulticystic dysplastic kidney\nAdrenal haemorrhage",
     "Ovarian cyst (girls)\nMesenteric/enteric cyst\nNeuroblastoma (cystic)\nHepatoblastoma"],
    ["Infant\n(1–12 months)",
     "Wilms tumour (nephroblastoma)\nNeuroblastoma\nHepatoblastoma",
     "Hydronephrosis\nSplenic cyst\nLymphoma\nTeratoma (sacrococcygeal)"],
    ["Toddler\n(1–5 years)",
     "Wilms tumour (peak 3 yrs)\nNeuroblastoma (peak 2 yrs)\nHepatoblastoma",
     "Lymphoma\nRhabdomyosarcoma\nGerm cell tumour\nDuplication cyst"],
    ["School age\n(5–10 years)",
     "Wilms tumour\nLymphoma (NHL)\nHepatocellular carcinoma",
     "Rhabdomyosarcoma\nNeuroblastoma (rare)\nGastrointestinal stromal tumour\nTeratoma"],
    ["Adolescent\n(>10 years)",
     "Lymphoma (most common solid mass)\nOvarian tumour/cyst (girls)\nHepatocellular carcinoma",
     "Rhabdomyosarcoma\nDesmoplastic small round cell tumour\nGastrointestinal tumours\nRenal cell carcinoma"],
]
W = PAGE_W - 2*MARGIN
story.append(colored_table(age_headers, age_rows,
    col_widths=[3.2*cm, W*0.38, W*0.42], hdr_color=TEAL))
story.append(SP(10))

# ══════════════════════════════════════════════════════
#  SECTION 2 - MAJOR DIAGNOSES COMPARISON TABLE
# ══════════════════════════════════════════════════════
story.append(section_banner("2. MAJOR DIAGNOSES — COMPARISON TABLE", NAVY))
story.append(SP(4))

comp_headers = ["Feature", "Neuroblastoma", "Nephroblastoma\n(Wilms)", "Hepatoblastoma",
                "Lymphoma", "Hydronephrosis"]
comp_rows = [
    ["Origin",
     "Adrenal / sympathetic chain (neural crest)",
     "Kidney (metanephric blastema)",
     "Liver (hepatic progenitors)",
     "Lymph nodes / reticuloendothelial",
     "Renal collecting system"],
    ["Peak Age",
     "2 yrs\n(90% < 5 yrs)",
     "3 yrs\n(2–5 yrs)",
     "< 3 yrs\n(90%)",
     "Any; NHL peak\n5–10 yrs",
     "Any age\n(often prenatal)"],
    ["Location",
     "Retroperitoneal;\nsuprarenal;\ncan be thoracic/cervical",
     "Intrarenal;\nRUQ or LUQ",
     "RUQ; hepatic",
     "Mesenteric, retroperitoneal, splenic",
     "Flank; renal fossa"],
    ["Bilateral",
     "Rare (< 5%)",
     "10% of cases",
     "Rare",
     "Diffuse/multi-nodal",
     "10–30% depending on cause"],
    ["Calcification on CT",
     "Very common\n(up to 85%)",
     "Rare (< 10%)",
     "Present in 50%\n(dystrophic)",
     "Rare; post-treatment",
     "None"],
    ["Vessel relationship",
     "Encases aorta/IVC\n(does NOT displace)",
     "Displaces vessels;\nIVC tumour thrombus",
     "Hepatic vein/IVC\ninvasion possible",
     "Displacement;\nno encasement",
     "Hydronephrotic pelvis; ureter dilated"],
    ["Tumour thrombus",
     "Absent",
     "IVC/renal vein\n(up to 10%)",
     "Hepatic vein/IVC",
     "Absent",
     "Not applicable"],
    ["Crosses midline",
     "Common (large masses)",
     "Less common",
     "Can cross midline\nif large",
     "Often bilateral/midline\nnodal disease",
     "Does not cross"],
    ["US key finding",
     "Anterior displacement\nof aorta & IVC;\nhyperechoic foci (Ca++)",
     "Intrarenal mass;\nclaw of normal\nrenal tissue;\nIVC thrombus",
     "Hepatic mass;\nhyperechoic or mixed",
     "Hypoechoic nodal\nmasses; splenomegaly",
     "Dilated pelvicalyceal\nsystem with anechoic\npelvis"],
    ["CT hallmark",
     "Calcification 85%;\nencasement of vessels;\nextension into spinal canal\n(dumb-bell)",
     "Intrarenal; claw sign;\nlow-attenuation areas;\nheterogeneous enhancement",
     "Large liver mass;\nmay show tumour\nthrombus in hepatic veins",
     "Bulky retroperitoneal\nor mesenteric nodes;\nhomogeneous",
     "Dilated pelvis ± ureter;\nthin cortex if longstanding"],
    ["MRI best use",
     "Spinal canal invasion;\nbone marrow metastases",
     "Gold standard for\nbilateral disease;\nnephroblastomatosis",
     "Resectability;\nvascular involvement",
     "Extent of disease;\nspinal cord compression",
     "Anatomy of\nobstruction; duplex"],
    ["Nuclear medicine",
     "¹²³I-mIBG (specific);\n⁹⁹ᵐTc bone scan",
     "Not used",
     "Not standard",
     "FDG-PET/CT",
     "MAG3 / DTPA renogram\n(function + drainage)"],
    ["Biochemical marker",
     "Urine VMA/HVA ↑\n(~90% of cases)",
     "None specific",
     "AFP ↑↑ (> 90%)",
     "LDH ↑; uric acid ↑",
     "None specific"],
    ["Metastases",
     "Bone, bone marrow,\nliver, LN, skin\n(~50% at diagnosis)",
     "Lung (first), liver\n(stage IV ~10%)",
     "Lung most common;\nliver",
     "Widespread by staging;\nbone marrow in NHL",
     "Not a malignancy;\nbilateral = systemic cause"],
    ["Prognosis",
     "Stage & MYCN dependent;\noverall 5-yr ~75%;\nhigh-risk ~40%",
     "Excellent overall;\n5-yr OS 86–96% (I–III);\n70% bilateral (V)",
     "Good if resectable;\n5-yr OS ~70%",
     "Good with\nchemotherapy",
     "Excellent after\nrelief of obstruction"],
]
cw2 = [2.8*cm, (W-2.8*cm)/5, (W-2.8*cm)/5, (W-2.8*cm)/5, (W-2.8*cm)/5, (W-2.8*cm)/5]
story.append(colored_table(comp_headers, comp_rows, col_widths=cw2, hdr_color=NAVY))
story.append(SP(10))

# ══════════════════════════════════════════════════════
#  SECTION 3 - NEUROBLASTOMA vs NEPHROBLASTOMA (focused)
# ══════════════════════════════════════════════════════
story.append(section_banner("3. NEUROBLASTOMA vs. NEPHROBLASTOMA — KEY DIFFERENTIATORS", TEAL))
story.append(SP(4))

diff_headers = ["Criterion", "Neuroblastoma", "Nephroblastoma (Wilms)"]
diff_rows = [
    ["Kidney relationship", "DISPLACES kidney — kidney is separate, intact, with own capsule visible", "ARISES FROM kidney — residual tissue forms a 'claw/beak' around mass"],
    ["Calcification (CT)", "Very common (85%) — coarse, amorphous", "Rare (< 10%)"],
    ["Vessel encasement", "Encases aorta/IVC/mesenteric vessels — classic NB pattern", "Displaces vessels; IVC tumour thrombus in up to 10%"],
    ["mIBG scintigraphy", "POSITIVE in ~80% — pathognomonic", "NEGATIVE — no catecholamine uptake"],
    ["Urine catecholamines", "VMA/HVA elevated in ~90%", "Normal"],
    ["Spinal extension", "Dumb-bell through foramina — MRI mandatory", "Not a feature"],
    ["Bilateral disease", "Rare", "10% — MRI gold standard for bilateral"],
    ["Crossing midline", "Common (large retroperitoneal masses)", "Less common"],
    ["IVC thrombus", "Absent", "Present in up to 10% — assess on US and CT"],
    ["Associated syndromes", "Familial ALK mutations", "WAGR, Denys-Drash, Beckwith-Wiedemann"],
]
cw3 = [3.8*cm, (W-3.8*cm)/2, (W-3.8*cm)/2]
story.append(colored_table(diff_headers, diff_rows, col_widths=cw3, hdr_color=PURPLE))
story.append(SP(10))

# ══════════════════════════════════════════════════════
#  SECTION 4 - IMAGING APPROACH ALGORITHM
# ══════════════════════════════════════════════════════
story.append(section_banner("4. RECOMMENDED IMAGING APPROACH", NAVY))
story.append(SP(4))

algo_data = [
    [Paragraph("<b>STEP 1 — ULTRASOUND (FIRST LINE for all)</b>", Bold)],
    [Paragraph("• Determine organ of origin (renal vs. hepatic vs. retroperitoneal vs. pelvic)\n"
               "• Assess vascularity with Colour Doppler\n"
               "• Identify IVC / renal vein tumour thrombus (critical for Wilms)\n"
               "• Assess for calcification (hyperechoic foci)\n"
               "• Determine cystic vs. solid vs. mixed nature", Body)],
    [Paragraph("<b>STEP 2 — CT ABDOMEN/PELVIS with IV contrast (staging & characterisation)</b>", Bold)],
    [Paragraph("• NB: chest + abdomen + pelvis — early contrast phase for chest, portal phase for abdomen\n"
               "• Wilms: abdomen + pelvis; add chest CT if CXR equivocal for pulmonary metastases\n"
               "• Assess calcification, vessel encasement vs. displacement, lymphadenopathy\n"
               "• Evaluate contralateral kidney (bilateral Wilms / NB)", Body)],
    [Paragraph("<b>STEP 3 — MRI (specific indications)</b>", Bold)],
    [Paragraph("• Paraspinal NB: MANDATORY to evaluate spinal canal extension (dumb-bell)\n"
               "• Bilateral Wilms: GOLD STANDARD — superior to CT for bilateral disease\n"
               "• Nephroblastomatosis: gadolinium required to detect nephrogenic rests\n"
               "• Hepatic tumours: vascular anatomy for surgical planning\n"
               "• Preferred over CT in neonates/infants (no ionising radiation)", Body)],
    [Paragraph("<b>STEP 4 — NUCLEAR MEDICINE (disease-specific)</b>", Bold)],
    [Paragraph("• ¹²³I-mIBG scintigraphy: NEUROBLASTOMA only — characterisation + metastasis staging\n"
               "• ⁹⁹ᵐTc-MDP bone scan: NB skeletal metastases (+ mIBG)\n"
               "• MAG3 / DTPA renogram: HYDRONEPHROSIS — functional assessment\n"
               "• FDG-PET/CT: LYMPHOMA staging and treatment response\n"
               "• Hepatobiliary scan (HIDA): biliary atresia / choledochal cyst", Body)],
]
algo_t = Table([[row[0]] for row in algo_data], colWidths=[W])
algo_style = []
for i in range(len(algo_data)):
    if i % 2 == 0:
        algo_style.append(("BACKGROUND", (0,i), (-1,i), LTGRAY))
    else:
        algo_style.append(("BACKGROUND", (0,i), (-1,i), WHITE))
algo_style += [
    ("GRID",          (0,0), (-1,-1), 0.3, MIDGRAY),
    ("TOPPADDING",    (0,0), (-1,-1), 5),
    ("BOTTOMPADDING", (0,0), (-1,-1), 5),
    ("LEFTPADDING",   (0,0), (-1,-1), 8),
    ("RIGHTPADDING",  (0,0), (-1,-1), 8),
    ("VALIGN",        (0,0), (-1,-1), "TOP"),
]
flat_algo = [[Paragraph(row[0].text if hasattr(row[0], 'text') else "", Body)] for row in algo_data]

# Build as simple alternating table
algo_rows_built = []
for i, row in enumerate(algo_data):
    algo_rows_built.append([row[0]])

algo_t2 = Table(algo_rows_built, colWidths=[W])
algo_t2.setStyle(TableStyle([
    ("ROWBACKGROUNDS", (0,0), (-1,-1), [LTGRAY, WHITE]),
    ("GRID",           (0,0), (-1,-1), 0.3, MIDGRAY),
    ("TOPPADDING",     (0,0), (-1,-1), 6),
    ("BOTTOMPADDING",  (0,0), (-1,-1), 6),
    ("LEFTPADDING",    (0,0), (-1,-1), 10),
    ("RIGHTPADDING",   (0,0), (-1,-1), 10),
    ("VALIGN",         (0,0), (-1,-1), "TOP"),
]))
story.append(algo_t2)
story.append(SP(10))

# ══════════════════════════════════════════════════════
#  SECTION 5 - BIOCHEMICAL & CLINICAL RED FLAGS
# ══════════════════════════════════════════════════════
story.append(section_banner("5. CLINICAL & BIOCHEMICAL CLUES", TEAL))
story.append(SP(4))

clue_headers = ["Finding", "Think of...", "Notes"]
clue_rows = [
    ["Elevated urine VMA/HVA", "Neuroblastoma", "Present in ~90%; sufficient with +ve bone marrow to confirm Dx"],
    ["AFP markedly elevated", "Hepatoblastoma", "AFP normally high in neonates; use age-adjusted reference ranges"],
    ["Haematuria", "Wilms tumour", "Microscopic haematuria in 25%; macroscopic less common"],
    ["Hypertension", "Neuroblastoma, Wilms, Phaeochromocytoma, Renovascular", "Catecholamines (NB) vs. renin excess (Wilms)"],
    ["Horner syndrome", "Cervical/thoracic Neuroblastoma", "Stellate ganglion involvement"],
    ["Opsoclonus-myoclonus\n(dancing eyes/feet)", "Neuroblastoma", "Paraneoplastic — MEI; may have small/occult primary"],
    ["Watery diarrhoea\n(intractable)", "Neuroblastoma", "VIP (vasoactive intestinal peptide) production"],
    ["Proptosis /\n'raccoon eyes'", "Neuroblastoma\n(orbital metastases)", "Periorbital ecchymosis — classic NB presentation"],
    ["Aniridia", "Wilms tumour (WAGR syndrome)", "WT1 deletion on 11p13"],
    ["Hemihypertrophy /\nmacroglossia", "Wilms tumour\n(Beckwith-Wiedemann)", "Also hepatoblastoma risk; screen with USS"],
    ["Blueberry muffin skin\nnodules (neonate)", "Neuroblastoma\n(Stage 4S)", "Excellent prognosis; often undergoes spontaneous regression"],
    ["Elevated LDH + uric acid", "Lymphoma", "Tumour lysis risk; treat before biopsy"],
    ["Abdominal mass +\nback pain/leg weakness", "NB (dumb-bell tumour)", "Emergency MRI to rule out cord compression"],
    ["Painless abdominal mass,\ncrossing midline", "Neuroblastoma or\nLymphoma", "Wilms rarely crosses midline"],
    ["Palpable mass with\nrapid growth + fever", "Wilms tumour, Lymphoma", "Ruptured Wilms: peritoneal seeding → upstages to Stage III"],
]
cw5 = [3.8*cm, 3.5*cm, W - 3.8*cm - 3.5*cm]
story.append(colored_table(clue_headers, clue_rows, col_widths=cw5, hdr_color=AMBER))
story.append(SP(10))

# ══════════════════════════════════════════════════════
#  SECTION 6 - STAGING SYSTEMS
# ══════════════════════════════════════════════════════
story.append(section_banner("6. STAGING SYSTEMS", NAVY))
story.append(SP(4))

# Two-column layout: NB staging | Wilms staging
nb_stage = [
    ["<b>NEUROBLASTOMA — INSS Staging</b>"],
    ["Stage 1: Localised, complete gross excision ± microscopic residual"],
    ["Stage 2A: Localised, incomplete excision; ipsilateral LN negative"],
    ["Stage 2B: Localised ± incomplete excision; ipsilateral LN positive"],
    ["Stage 3: Unresectable unilateral tumour infiltrating across midline ± LN; or contralateral LN +ve"],
    ["Stage 4: Distant mets (bone, BM, liver, LN, other) except 4S"],
    ["Stage 4S: Localised primary + spread limited to liver, skin, BM only (no bone); age < 1 yr — excellent prognosis"],
    ["<b>Risk Groups:</b> Low / Intermediate / High (based on INSS + MYCN + ploidy + histology + ALK)"],
    ["<b>MYCN amplification</b> → auto High Risk regardless of stage"],
]

wilms_stage = [
    ["<b>NEPHROBLASTOMA — NWTSG Staging (post-surgical)</b>"],
    ["Stage I (43%): Confined to kidney; complete excision"],
    ["Stage II (23%): Beyond renal capsule; complete excision; vessel infiltration or biopsy pre-resection"],
    ["Stage III (23%): +ve abdominopelvic LN; peritoneal invasion; residual tumour or unresectable"],
    ["Stage IV (10%): Haematogenous spread (lung, liver, bone, brain) or outside abdomen/pelvis"],
    ["Stage V (5%): Bilateral tumours at diagnosis — nephron-sparing surgery goal"],
    ["<b>SIOP (European):</b> Biopsy + neoadjuvant chemo first; then surgical staging"],
    ["<b>Anaplasia:</b> Focal (5%) — better outcome; Diffuse — very poor (4-yr OS: III=45%, IV=7%)"],
    ["<b>4-yr OS:</b> Stage I–III: 86–96% | Stage IV: 83% | Stage V: 70%"],
]

def stage_table(rows, color):
    data = [[Paragraph(r[0], Small)] for r in rows]
    t = Table(data, colWidths=[(W/2) - 0.3*cm])
    row_styles = []
    for i in range(len(data)):
        bg = LTGRAY if i % 2 == 0 else WHITE
        row_styles.append(("BACKGROUND", (0,i), (-1,i), bg))
    t.setStyle(TableStyle([
        ("BACKGROUND",    (0,0), (-1,0), color),
        ("ROWBACKGROUNDS",(0,1), (-1,-1), [LTGRAY, WHITE]),
        ("GRID",          (0,0), (-1,-1), 0.3, MIDGRAY),
        ("TOPPADDING",    (0,0), (-1,-1), 4),
        ("BOTTOMPADDING", (0,0), (-1,-1), 4),
        ("LEFTPADDING",   (0,0), (-1,-1), 6),
        ("RIGHTPADDING",  (0,0), (-1,-1), 6),
        ("VALIGN",        (0,0), (-1,-1), "TOP"),
    ]))
    return t

# Override first row text color
nb_t = stage_table(nb_stage, TEAL)
wilms_t = stage_table(wilms_stage, PURPLE)

combo = Table([[nb_t, wilms_t]], colWidths=[W/2 - 0.1*cm, W/2 - 0.1*cm],
              hAlign="LEFT")
combo.setStyle(TableStyle([
    ("VALIGN", (0,0), (-1,-1), "TOP"),
    ("LEFTPADDING",  (0,0), (-1,-1), 0),
    ("RIGHTPADDING", (0,0), (-1,-1), 3),
]))
story.append(combo)
story.append(SP(10))

# ══════════════════════════════════════════════════════
#  SECTION 7 - OTHER IMPORTANT ABDOMINAL MASSES
# ══════════════════════════════════════════════════════
story.append(section_banner("7. OTHER IMPORTANT PEDIATRIC ABDOMINAL MASSES", TEAL))
story.append(SP(4))

other_headers = ["Tumour / Mass", "Key Features", "Radiology Clues", "Biochemistry"]
other_rows = [
    ["Hepatoblastoma",
     "Most common hepatic malignancy in children < 3 yrs. Associated with BWS, prematurity, FAP.",
     "US/CT: hepatic mass, often heterogeneous, calcification in 50%. MRI best for resectability & vascular involvement.",
     "AFP markedly elevated (> 1,000 ng/mL)"],
    ["Hepatic haemangioma\n(infantile)",
     "Most common benign hepatic tumour in infants. Often multiple. Most regress by age 5.",
     "US: hyperechoic nodules. Contrast CT/MRI: peripheral nodular enhancement with centripetal fill-in. No AFP rise.",
     "Normal AFP; Kasabach-Merritt if large"],
    ["Ovarian cyst / tumour",
     "Most common abdominal mass in newborn girls. Follicular cysts common. Germ cell tumours in adolescents.",
     "US: anechoic (simple cyst) or complex (teratoma — fat + calcification on CT). MRI for complex masses.",
     "AFP/βhCG elevated in germ cell tumours"],
    ["Sacrococcygeal teratoma",
     "Most common neonatal tumour. 80% benign in neonates; risk of malignancy rises with age.",
     "US/MRI: mixed cystic-solid mass at sacrococcygeal region extending into pelvis. CT: calcification.",
     "AFP elevated if malignant"],
    ["Choledochal cyst",
     "Congenital biliary dilatation. Presents with pain, jaundice, mass (Charcot's triad). Risk of cholangiocarcinoma.",
     "US: cystic mass in RUQ closely related to bile duct. MRCP for delineation of biliary anatomy.",
     "ALP/GGT elevated; bilirubin"],
    ["Mesenteric / duplication cyst",
     "Benign; may be symptomatic from torsion/bleeding. Duplication cysts share bowel wall.",
     "US: anechoic thin-walled cyst. Duplication cysts: thick muscular wall with 'gut signature' on US.",
     "Normal"],
    ["Lymphoma (NHL / Hodgkin)",
     "Most common abdominal malignancy in > 5 yrs. Burkitt NHL common in ileocaecal region.",
     "US/CT: bulky homogeneous hypoechoic retroperitoneal/mesenteric lymphadenopathy. Splenomegaly. FDG-PET/CT for staging.",
     "LDH ↑, uric acid ↑; β2-microglobulin ↑"],
    ["Rhabdomyosarcoma",
     "Most common soft tissue sarcoma in children. Abdominopelvic sites: bladder, prostate, vagina, retroperitoneum.",
     "US/CT: heterogeneous soft tissue mass. Bladder RMS: irregular filling defect. MRI for pelvic extension.",
     "Non-specific; urinalysis if bladder involved"],
    ["Multicystic dysplastic\nkidney (MCDK)",
     "Benign non-functioning kidney replaced by non-communicating cysts. Usually unilateral. Contralateral UPJO in 15%.",
     "US: multiple non-communicating cysts of varying sizes with no central large cyst. No reniform shape.",
     "MAG3 confirms non-function"],
    ["Hydronephrosis",
     "Most common cause of neonatal abdominal mass. UPJO (ureteropelvic junction obstruction) most common cause.",
     "US: dilated pelvicalyceal system. MAG3 renogram for function and drainage. MRI urography for anatomy.",
     "Creatinine if bilateral"],
]
cw7 = [3.0*cm, (W-3.0*cm)*0.34, (W-3.0*cm)*0.40, (W-3.0*cm)*0.26]
story.append(colored_table(other_headers, other_rows, col_widths=cw7, hdr_color=GREEN))
story.append(SP(10))

# ══════════════════════════════════════════════════════
#  SECTION 8 - RED FLAG BOX
# ══════════════════════════════════════════════════════
story.append(section_banner("8. RED FLAGS — URGENT ASSESSMENT REQUIRED", RED))
story.append(SP(4))

red_lines = [
    "• Abdominal mass + lower limb weakness / bladder/bowel dysfunction → MRI SPINE STAT (NB dumb-bell / lymphoma cord compression)",
    "• Hypertensive crisis + abdominal mass → Neuroblastoma / Phaeochromocytoma (urine catecholamines urgently)",
    "• Rapidly expanding mass + fever + anaemia → Wilms tumour (avoid palpation — risk of rupture)",
    "• Bilateral renal masses → Wilms Stage V — nephron-sparing imperative; consult paediatric oncology",
    "• IVC tumour thrombus extending to right atrium (Wilms) → cardiac surgery team involvement pre-op",
    "• Opsoclonus-myoclonus (dancing eyes) in an infant → Search for occult neuroblastoma (mIBG + urine catecholamines)",
    "• Neonatal abdominal mass crossing the midline → Most likely neuroblastoma; urgent US + catecholamines",
    "• Wilms tumour rupture (peritoneal seeding) → Upstages to Stage III; bilateral coverage chemotherapy required",
]
red_data = [[Paragraph(l, Body)] for l in red_lines]
red_t = Table(red_data, colWidths=[W])
red_t.setStyle(TableStyle([
    ("ROWBACKGROUNDS", (0,0), (-1,-1), [colors.HexColor("#fff5f5"), WHITE]),
    ("GRID",           (0,0), (-1,-1), 0.3, colors.HexColor("#f5c6c6")),
    ("TOPPADDING",     (0,0), (-1,-1), 5),
    ("BOTTOMPADDING",  (0,0), (-1,-1), 5),
    ("LEFTPADDING",    (0,0), (-1,-1), 8),
    ("RIGHTPADDING",   (0,0), (-1,-1), 8),
    ("VALIGN",         (0,0), (-1,-1), "TOP"),
]))
story.append(red_t)
story.append(SP(10))

# ══════════════════════════════════════════════════════
#  SECTION 9 - MNEMONICS
# ══════════════════════════════════════════════════════
story.append(section_banner("9. MNEMONICS & MEMORY AIDS", TEAL))
story.append(SP(4))

mnem_data = [
    [Paragraph("<b>Neuroblastoma — 'NEAT'</b>", SmallBold),
     Paragraph("<b>Wilms — 'RAVEN'</b>", SmallBold),
     Paragraph("<b>Hepatoblastoma — 'HALF'</b>", SmallBold)],
    [Paragraph("<b>N</b>eural crest origin\n<b>E</b>levated VMA/HVA\n<b>A</b>drenal / paraaortic\n<b>T</b>umour calcification 85%", Small),
     Paragraph("<b>R</b>enal origin\n<b>A</b>FP normal\n<b>V</b>ein (IVC) thrombus\n<b>E</b>xcellent prognosis\n<b>N</b>o calcification usually", Small),
     Paragraph("<b>H</b>epatic location\n<b>A</b>FP very high\n<b>L</b>iver mass < 3 yrs\n<b>F</b>AP / BWS association", Small)],
]
mnem_t = Table(mnem_data, colWidths=[W/3, W/3, W/3])
mnem_t.setStyle(TableStyle([
    ("BACKGROUND",    (0,0), (-1,0), LTGRAY),
    ("ROWBACKGROUNDS",(0,1), (-1,-1), [WHITE]),
    ("GRID",          (0,0), (-1,-1), 0.4, MIDGRAY),
    ("TOPPADDING",    (0,0), (-1,-1), 6),
    ("BOTTOMPADDING", (0,0), (-1,-1), 6),
    ("LEFTPADDING",   (0,0), (-1,-1), 8),
    ("VALIGN",        (0,0), (-1,-1), "TOP"),
    ("ALIGN",         (0,0), (-1,-1), "LEFT"),
]))
story.append(mnem_t)
story.append(SP(10))

# ══════════════════════════════════════════════════════
#  SECTION 10 - SOURCES
# ══════════════════════════════════════════════════════
story.append(HRFlowable(width="100%", thickness=0.5, color=MIDGRAY))
story.append(SP(4))
story.append(Paragraph(
    "<b>Sources:</b> Grainger &amp; Allison's Diagnostic Radiology (Elsevier) · "
    "Robbins &amp; Kumar Basic Pathology (Elsevier) · "
    "Campbell Walsh Wein Urology · "
    "Robbins, Cotran &amp; Kumar Pathologic Basis of Disease",
    FooterS))

# ── Build ──────────────────────────────────────────────────────────────────────
doc.build(story, onFirstPage=on_first_page, onLaterPages=on_page)
print(f"PDF created: {OUTPUT}")

Running Command

cd /home/daytona/workspace/peds-abdominal-mass && python generate_pdf.py 2>&1

Running Command

ls -lh /home/daytona/workspace/peds-abdominal-mass/Pediatric_Abdominal_Mass_DDx.pdf

Your quick reference PDF is ready:

Pediatric Abdominal Mass DDx

PDF Document · PDF

Here's what's inside the document:

Section 1 - Age-Based Differential Overview A table organizing the differential by age group: neonate, infant, toddler, school age, and adolescent - showing what to think of first at each stage.

Section 2 - Major Diagnoses Comparison Table Side-by-side comparison of Neuroblastoma, Nephroblastoma (Wilms), Hepatoblastoma, Lymphoma, and Hydronephrosis across 15 features: origin, peak age, location, calcification, vessel relationship, tumour thrombus, US/CT/MRI findings, nuclear medicine, biochemistry, metastases, and prognosis.

Section 3 - Neuroblastoma vs. Nephroblastoma (focused) The 10 most important differentiating criteria (kidney relationship, calcification, mIBG, vessel encasement, IVC thrombus, spinal extension, etc.).

Section 4 - Imaging Approach Algorithm Step-by-step: US first → CT → MRI (with specific indications for each) → Nuclear medicine (which modality for which diagnosis).

Section 5 - Clinical & Biochemical Clues 15 findings mapped to diagnoses - from elevated VMA/HVA and AFP, to opsoclonus-myoclonus, Horner syndrome, raccoon eyes, and blueberry muffin skin.

Section 6 - Staging Systems INSS (Neuroblastoma) and NWTSG/SIOP (Wilms) staging in a two-column layout with survival data.

Section 7 - Other Important Masses Hepatoblastoma, ovarian tumours, sacrococcygeal teratoma, choledochal cyst, duplication cyst, lymphoma, rhabdomyosarcoma, MCDK, and hydronephrosis.

Section 8 - Red Flags 8 urgent scenarios requiring immediate action (cord compression, IVC thrombus, hypertensive crisis, tumour rupture, etc.).

Section 9 - Mnemonics NEAT (Neuroblastoma), RAVEN (Wilms), HALF (Hepatoblastoma).

stepwise imaging approach for non infective inflammatory arthritis DNB 10 marks question answer radiology

Reading File

This is a broad + visual radiology question. Running parallel searches across the library for inflammatory arthritis imaging.

Finding Sources

Good - found the Firestein & Kelley rheumatology textbook. Let me search more specifically in the radiology textbook and this rheumatology text.

Finding Sources

Found the key Grainger & Allison rheumatoid arthritis section. Let me read it in full, plus the inflammatory arthritis imaging section in the rheumatology textbook.

Reading File

Finding Sources

Excellent. Now let me read the ankylosing spondylitis section and get more of the RA imaging section (summary box and remaining content).

Reading File

Rich content. Let me continue reading the AS section (psoriatic arthritis, reactive arthritis) and the earlier RA section for more imaging details.

Reading File

Excellent data. Now let me fetch the key images (RA erosion X-ray, AS MRI sacroiliitis, bamboo spine) and read a bit more of the psoriatic/reactive arthritis section.

Loading Image

Reading File

I have comprehensive, authoritative data from Grainger & Allison. Let me fetch the RA hand erosion X-ray and the psoriatic arthritis pencil-in-cup X-ray for visual support.

Loading Image

I now have everything needed for a comprehensive, exam-standard 10-mark answer. Here it is:

Stepwise Imaging Approach for Non-Infective Inflammatory Arthritis

DNB Radiology — 10 Marks

Introduction

Non-infective inflammatory arthritis encompasses a group of immune-mediated joint diseases including Rheumatoid Arthritis (RA), Seronegative Spondyloarthropathies (SpA) — ankylosing spondylitis (AS), psoriatic arthritis (PsA), reactive arthritis, and enteropathic arthritis — as well as crystal arthropathies (gout, CPPD) and connective tissue disease-related arthritis (SLE, JIA). With the advent of disease-modifying antirheumatic drugs (DMARDs) and biologics, early and precise imaging has become indispensable for diagnosis, disease activity assessment, and monitoring treatment response.

The imaging approach is stepwise: beginning with plain radiographs as the baseline, escalating to ultrasound and MRI for early/subtle disease, and using CT and nuclear medicine for specific indications.

STEP 1 — PLAIN RADIOGRAPHY (First Line, Always)

Indications: All patients presenting with suspected inflammatory arthritis. Provides a structural baseline and is used for serial follow-up.

Standard views:

Hands and wrists (PA, both together on one film)
Feet (DP and oblique)
Pelvis AP (sacroiliac joints)
Lateral spine (cervical, thoracic, lumbar as indicated)
Any symptomatic joint

What Radiographs Detect:

Feature	RA	AS	Psoriatic Arthritis
Soft-tissue swelling	Fusiform/spindle-shaped; periarticular	Diffuse (dactylitis)	Dactylitis ("sausage digit")
Bone density	Periarticular osteopenia	Preserved	Preserved
Joint distribution	Symmetric; MCP, PIP, wrist; DIP spared	Sacroiliac joints, spine	DIP, asymmetric; can mimic RA
Erosions	Marginal; "dot-dash" early; radial index/middle MCPs	Entheseal; "shiny corners" (Romanus lesions)	Marginal + central; "pencil-in-cup"
New bone formation	Absent	Syndesmophytes (thin, vertical); bamboo spine	Coarse, bulky syndesmophytes; periosteal reaction
Joint space	Panarticular uniform loss	Disc space preserved until late; ankylosis	Variable; loss with preserved density
Ankylosis	Wrist (intercarpal, carpometacarpal)	Sacroiliac joints + spine	Interphalangeal joints

Radiographic Images:

RA - Earliest erosion: Marginal erosion at the radial border of the middle MCP head with periarticular localised osteopenia showing a characteristic "dot-dash" pattern:

RA earliest erosion - radial metacarpal head with dot-dash osteopenia and frank erosion

Fig. 44.14 - Earliest RA erosive change along radial border of middle MCH. White arrow: "dot-dash" deossification. Black arrow: frank erosion of ulnar border.

AS - Romanus lesions and syndesmophytes:

Ankylosing spondylitis with Romanus lesions and bridging syndesmophytes in thoracic spine

Fig. 44.18 - AS: Romanus lesions (sclerotic shiny corners) and bridging syndesmophytes in thoracic spine.

Psoriatic Arthritis - Pencil-in-cup deformity:

Psoriatic arthritis pencil-in-cup erosive deformity at DIP joints of toes

Fig. 44.25 - PsA: Pencil-in-cup erosive deformity with subluxation of DIP joints and acro-osteolysis giving peg-like phalanges.

Limitations of radiographs:

Insensitive for early disease (synovitis, early erosions not visible)
Structural damage seen only after significant bone loss
Cannot image soft tissue (tendons, bursae, cartilage)

STEP 2 — ULTRASOUND (US)

Indications: Early disease where radiographs are normal; assessment of synovitis; detecting subclinical disease in additional joints; treatment monitoring; guiding aspiration/injection.

Technique: High-frequency linear probe (15-18 MHz for small joints); Power Doppler for vascularity.

What US Detects:

A. Synovitis:

Thickened, hypoechoic synovium that is non-compressible
Power Doppler signal within the synovium = active inflammation (vascularity)
US and MRI are both superior to clinical examination for detecting synovitis
Power Doppler correlates well with MRI detection of synovitis
US has the advantage of examining multiple joints in one session + dynamic imaging capability

B. Joint Effusion:

Anechoic or hypoechoic fluid within the joint capsule, compressible

C. Erosions:

Cortical defect in two perpendicular planes
More sensitive than radiography for early erosion detection
A cortical break must be demonstrated in two planes to avoid overcalling

D. Tenosynovitis:

Fluid/synovial thickening within tendon sheaths
Common in RA (flexor/extensor tendons at wrist); power Doppler shows active inflammation

E. Enthesitis:

Thickening and hypoechogenicity at tendon/ligament insertions
Power Doppler positivity
Hallmark of SpA (enthesopathy at Achilles, plantar fascia, patella tendon insertions)

F. Bursitis and Nodules:

Rheumatoid nodules: hypoechoic subcutaneous masses along extensor surfaces
Bursal distension (olecranon, subacromial, retrocalcaneal bursae)

Advantages of US:

Real-time, dynamic assessment
No radiation, no contrast
Can examine multiple joints rapidly
Guides aspiration and injection
Detects active synovitis (power Doppler) vs. inactive/fibrotic synovium

Limitations:

Operator-dependent
Limited for deep joints (hip, sacroiliac joints, spine)
Cannot assess bone marrow or cartilage

STEP 3 — MRI (Magnetic Resonance Imaging)

Indications: Earliest structural changes; sacroiliac joint disease (AS); spinal involvement; when radiograph and US are inconclusive; monitoring response to biologics; preoperative planning.

Standard MRI protocol for joints:

T1 weighted SE (anatomical)
T2/STIR fat-suppressed (oedema/inflammation)
Post-gadolinium T1 fat-saturated (synovitis)

MRI in Specific Conditions:

A. Rheumatoid Arthritis:

Gold standard reference modality for synovitis detection
IV gadolinium contrast is required to accurately diagnose synovitis — enhancing synovium vs. joint fluid
Bone marrow oedema (low T1, high T2/STIR) adjacent to eroded joints = active inflammation
Erosions: cortical break with bright marrow signal (oedema-type) or dark (fibrous/inactive)
Detects tenosynovitis — fluid signal around tendons in tendon sheaths
MRI of the cervical spine in RA: atlantoaxial subluxation, basilar invagination (C1/C2 pannus)

B. Ankylosing Spondylitis (MRI of Sacroiliac Joints - Most Important):

AS sacroiliac joints MRI - T1 erosions and T2 fat-saturated subchondral oedema

Fig. 44.17 - AS MRI sacroiliac joints. (A) T1 weighted: erosions (arrows) and joint space loss = structural damage. (B) T2 fat-suppressed: extensive bilateral subchondral bone marrow oedema (arrows) = disease activity.

Key AS MRI features:

Sequence	Finding	Significance
T1	Erosions at SIJ; joint space loss; fatty deposition	Structural damage (chronic)
STIR/T2 fat-sat	Subchondral bone marrow oedema at SIJ (bilateral, symmetrical)	Earliest feature - active sacroiliitis; appears BEFORE radiographic changes
STIR spine	Bone oedema at vertebral corners (MRI equivalent of Romanus lesion)	Pre-radiographic AS
STIR thoracic	Oedema within costovertebral joints at multiple levels	Enthesitis
Post-Gd	Enhancement of SIJ and discovertebral junctions	Active inflammation

MRI has greater sensitivity and specificity than CR, CT, and scintigraphy for early sacroiliitis
According to ASAS criteria, a positive SIJ MRI = at least one subchondral bone marrow oedema lesion on ≥2 consecutive slices, or ≥2 lesions on a single slice

C. Psoriatic Arthritis:

DIP joint synovitis, enthesitis (distinguishes from RA)
Dactylitis: diffuse tenosynovitis + peritendinous inflammation of the entire digit
Bone oedema: periarticular and entheseal

D. Crystal Arthropathy (Gout):

DECT (dual-energy CT) is now the imaging investigation of choice for detecting uric acid deposits (tophi appear green-coded on DECT)
MRI: tophi as low-signal T2 masses; bone marrow oedema adjacent to tophi

STEP 4 — COMPUTED TOMOGRAPHY (CT)

Indications: Specific situations; NOT a primary investigation for most inflammatory arthritis.

Uses:

Indication	Detail
Sacroiliac joints (AS)	CT shows erosions, subchondral sclerosis, and ankylosis with greater anatomical detail than plain films; but cannot detect pre-radiographic oedema — MRI is preferred for early disease
Bamboo spine / complications	CT of spine for fracture in fused AS spine (transverse fractures are unstable and life-threatening)
Andersson lesion	Inflammatory pseudarthrosis at unfused segment in AS; CT + MRI differentiate from infection
Dual-Energy CT (DECT)	Gout: uric acid crystal deposits (monosodium urate) coded green; highly specific
Pre-surgical planning	Complex deformity assessment; atlanto-axial subluxation in RA
CPPD	Chondrocalcinosis in fibrocartilage (menisci, symphysis pubis, triangular fibrocartilage); better seen on CT than radiograph

Limitations: Ionising radiation; no direct synovitis/marrow oedema assessment; inferior to MRI for soft tissue and early inflammatory changes.

STEP 5 — NUCLEAR MEDICINE (Specific Indications)

A. Bone Scintigraphy (⁹⁹ᵐTc-MDP):

Whole-body scan shows pattern of active joint involvement
Increased uptake in inflamed joints (periarticular)
Historically used for early sacroiliitis detection in AS — now superseded by MRI (lower sensitivity/specificity)
Still useful for whole-body disease mapping and unsuspected joint involvement

B. PET/CT (¹⁸F-FDG):

Used when large-vessel vasculitis is suspected (e.g., giant cell arteritis, Takayasu) — FDG uptake in vessel walls
Assessment of systemic inflammation in connective tissue diseases
Not routine for standard inflammatory arthritis

C. SPECT-CT:

Combined functional and anatomical imaging
Useful for complex joints (feet, sacroiliac joints, spine) where planar scintigraphy is inadequate

IMAGING ALGORITHM SUMMARY

SUSPECTED NON-INFECTIVE INFLAMMATORY ARTHRITIS
              ↓
STEP 1: PLAIN RADIOGRAPHS
(Hands/wrists/feet PA + oblique; pelvis AP for SIJ;
lateral spine; affected large joints)
              ↓
Early/normal X-ray + clinical suspicion of active disease?
         YES ↓
STEP 2: ULTRASOUND
(Small joints: hands, feet, wrists; Power Doppler for synovitis;
tendons for tenosynovitis; entheses for SpA)
              ↓
Peripheral joint disease → US sufficient for most monitoring
Axial disease / sacroiliitis suspected?
         YES ↓
STEP 3: MRI
(SIJ: coronal STIR + T1; Spine: sagittal STIR + T1;
Peripheral joints: post-Gd T1 fat-sat for synovitis)
              ↓
Specific indications only:
STEP 4: CT → AS fracture, erosion detail, DECT for gout
STEP 5: Nuclear medicine → Bone scan (whole body mapping),
         PET-CT (large vessel vasculitis)

Imaging Features of Major Entities — Quick Summary

	Rheumatoid Arthritis	Ankylosing Spondylitis	Psoriatic Arthritis	Reactive Arthritis
Distribution	Symmetric; MCP, PIP, wrist	Axial; bilateral SIJ; spine	Asymmetric; DIP + MCP; axial	Asymmetric; lower limb; SIJ
DIP involvement	Spared	Rarely	Classic	Common
Osteopenia	Periarticular (early)	Absent	Absent	Absent
Erosions	Marginal, bare area	Entheseal (shiny corner)	Marginal + central (pencil-in-cup)	Marginal
New bone	Absent	Syndesmophytes (thin, vertical)	Bulky, coarse syndesmophytes	Periosteal; calcaneal spurs
Sacroiliitis	Absent	Bilateral, symmetric	Asymmetric or bilateral	Asymmetric
MRI key finding	Synovitis (Gd); erosions	Bilateral SIJ subchondral BME (STIR)	Dactylitis; enthesitis	SIJ BME (asymmetric)
US key finding	Synovitis; power Doppler	Enthesitis (Achilles etc.)	Dactylitis; enthesitis	Enthesitis

Key Points for DNB Exam Answer

Plain radiographs remain the first line but detect only established structural damage (erosions, joint space loss); they miss early synovitis.
US and MRI are superior to clinical examination for detecting synovitis and are more sensitive than radiographs for early erosion detection.
IV gadolinium is required for accurate MRI diagnosis of synovitis - enhancing synovium must be distinguished from joint fluid.
MRI of sacroiliac joints is the gold standard for early AS — subchondral bone marrow oedema on STIR/T2 fat-suppressed sequences is the earliest feature, appearing before any radiographic change.
MRI has greater sensitivity and specificity than CR, CT, and scintigraphy for early sacroiliitis.
CT is reserved for specific situations: fracture in AS, detailed erosion morphology, or DECT for gout.
Bone scintigraphy has largely been replaced by MRI for early sacroiliitis detection.
A cortical defect must be confirmed in two perpendicular planes on US/MRI to confidently diagnose an erosion.

Sources: Grainger & Allison's Diagnostic Radiology, Chapter 44 (Inflammatory Arthritis), p. 1117-1124; Firestein & Kelley's Textbook of Rheumatology

This is a shared conversation. Sign in to Orris to start your own chat.