Indications (NG14 §1.4)

Consider SLNB for melanoma at the time of, or shortly after, wide local excision in patients with:
- pT2b–pT4 (Breslow > 1.0 mm with ulceration, or any tumour > 1.0 mm).
- pT1b–pT2a after discussion of benefits, risks and uncertainty. In AJCC 8, pT1b = Breslow < 0.8 mm WITH ulceration, OR 0.8–1.0 mm regardless of ulceration; pT2a = > 1.0–2.0 mm without ulceration. Mitotic rate and lymphovascular invasion are adverse prognostic features that inform the SLNB discussion but are no longer T-category criteria in AJCC 8.
Not offered for pT1a (< 0.8 mm, no ulceration) — yield < 5% and morbidity outweighs.
Not offered routinely for pTis (in situ) or pure desmoplastic melanoma.
Accurate nodal staging affects adjuvant treatment discussions: pembrolizumab TA837 for completely resected stage IIB/IIC; pembrolizumab TA766 or nivolumab TA684 for resected stage III; dabrafenib + trametinib TA544 for BRAF V600-mutant stage III.

Evidence base (MSLT-I and II)

MSLT-I (Morton, NEJM 2014) — SLNB-based staging in intermediate-thickness melanoma. SLNB does not improve overall survival but identifies SLN-positive patients with significantly worse melanoma-specific survival; treatment of regional recurrence later is associated with worse outcomes than upfront identification.
MSLT-II (Faries, NEJM 2017) — randomised SLN-positive patients to immediate completion lymphadenectomy vs nodal observation with ultrasound. No melanoma-specific survival difference; CLND increased lymphoedema. CLND therefore no longer routine after positive SLN — see completion lymphadenectomy.
DeCOG-SLT (Leiter, Lancet Oncol 2016) — confirmed no benefit of immediate CLND in pT1–pT3 SLN-positive melanoma.

Pre-operative lymphoscintigraphy — same-day intradermal injection of technetium-99m sulphur colloid or albumin around the primary site; sequential gamma camera images identify draining basins and the sentinel node(s).
Intra-operatively — patent blue dye injected intradermally around the primary 5–10 minutes before incision; gamma probe localises the sentinel node; combined visual blue staining and gamma activity ≥ 10% of hottest node defines the sentinel(s).
Multiple sentinels — all hot/blue nodes harvested.
Wide local excision usually performed at the same operation.
Indocyanine green / near-infrared fluorescence is an emerging alternative in some UK centres.

Step-sectioned, S100 / SOX10 / Melan-A / HMB45 immunostains.
Tumour-deposit size matters — < 0.1 mm, 0.1–1.0 mm, > 1.0 mm; the AJCC 8 classification of an SLN+ patient depends on burden, ulceration and other features.
Extranodal extension recorded.
Pathology is integrated into AJCC 8 N classification — N1a, N2a, N3a for clinically occult disease.

Routine completion lymphadenectomy no longer recommended in most cases. If the sentinel node is positive and completion lymphadenectomy is not done, NICE NG14 advises considering 2 nodal-basin ultrasound scans per year in years 1–3 alongside the stage III follow-up schedule (see completion lymphadenectomy).
Staging imaging — CT NCAP and brain MRI per the resected stage III pathway (NG14 §1.9.15).
Adjuvant systemic therapy discussion at MDT — pembrolizumab (TA766) or nivolumab (TA684); dabrafenib + trametinib (TA544) for BRAF V600-mutant stage III disease.

Wound infection 5–10%, seroma 5–15%, haematoma uncommon.
Lymphoedema — lower for SLNB alone than for CLND; cumulative risk depends on basin (groin > axilla > neck).
Blue dye anaphylaxis — rare (~1:1000); pre-procedure counselling.
False-negative rate < 5% in trained hands; recurrence in a previously negative basin warrants careful re-evaluation.