Lymphoedema was found in about half of patients after head and neck radiotherapy

Design

Systematic review and meta-analysis of prevalence (not prospectively registered)

Population

26 studies, 3,133 adults with mucosal head and neck cancer treated with radiotherapy

Primary outcome

Pooled prevalence of lymphoedema

Effect

49% (95% CI 36% to 62%); grade 2 or worse 32% (14% to 59%); I-squared 94%

This systematic review pooled 26 studies with 3,133 adults who had radiotherapy for mucosal head and neck cancer. The pooled prevalence of lymphoedema was 49% (95% CI 36% to 62%) with considerable heterogeneity (I-squared 94%). External lymphoedema was estimated at 37% (12% to 71%) and internal at 45% (31% to 60%). Where severity was graded, clinically significant grade 2 or worse lymphoedema was 32% (14% to 59%).

Definitions and assessment methods varied, most commonly clinical grading, and the evidence was mostly retrospective and single-arm. Overall certainty was very low and small-study effects were suggested (Egger p = 0.06). The review was not prospectively registered.

The estimates are imprecise, but even the lower end of the interval is substantial. The authors suggest routine surveillance, early identification and timely referral to lymphoedema services within survivorship pathways, which may be reasonable to consider locally while standardised measures are developed.

• Pooled lymphoedema prevalence after head and neck radiotherapy was 49% (95% CI 36% to 62%).
• About a third (32%) had grade 2 or worse, but with a wide interval.
• Both external and internal swelling count; examine the neck, face and the upper aerodigestive tract.
• Ask about swelling, tightness and swallowing difficulty at follow-up visits.
• Consider early referral to a lymphoedema therapist where one is available.

Why it matters

It points to a common late effect that is easily missed unless someone looks for it.

Don't overread it

Very low certainty evidence from mostly retrospective studies with varying definitions cannot give a reliable rate or show that screening improves outcomes.

The statistics, in plain English

A prevalence of 49% with an interval of 36% to 62% and an I-squared of 94% means studies disagreed widely, probably because they used different definitions and assessment methods, so the true figure in any clinic could be quite different.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.