How LDRT Compares to Other Osteoarthritis Treatments
Low-dose radiation therapy (LDRT) is not a first-line treatment for osteoarthritis. The 2026 ARS Appropriate Use Criteria is explicit: LDRT is considered after 3–6 months of prior treatment failure. Response to multimodality lifestyle management in newly diagnosed OA is under 50%. Roughly 25% of patients requiring pharmacologic management will not respond or lose responsiveness over time. LDRT response rates are 60–90% per the ARS AUC and DEGRO guideline.
Evidence and Guideline Summary
Rows ordered by guideline support strength, strongest first. Three columns: Treatment | Prospective Randomized Trials | Guidelines Summary.
Topical NSAIDs: 39 RCTs / 10,631 patients. Strongly recommended as first-line. NICE: "Offer." AAOS: Strong. ACR and OARSI: strongly recommended. First-line pharmacologic agent across all major bodies.
Oral NSAIDs: 192 RCTs / 102,829 patients. Strongly recommended with comorbidity cautions. NICE: "Consider" (second line). AAOS: Strong. ACR and OARSI: strongly recommended. Not recommended with cardiovascular, gastrointestinal, or frailty comorbidity (OARSI).
IA Corticosteroid: 27 RCTs vs control. Broadly recommended for short-term relief. ACR: strongly recommended (knee and hip). OARSI: conditional for knee; not recommended for hip. AAOS: Moderate — short-term relief. NICE: "Consider."
LDRT: 12+ studies / ~1,750 patients; 4 published RCTs (2 positive: Fazilat-Panah 2025, Korean trial 2025 conference abstract; 2 negative: Mahler 2018, Minten 2018). 2026 ARS AUC (multispecialty: radiation oncologists, rheumatologists, orthopedic surgeons, patient advocate; formal consensus) concluded 60–90% treatment response rate. DEGRO national guideline recommends treatment: 0.5–1.0 Gy/fraction, 3.0–6.0 Gy total. 2026 international consensus statement standardizes target volumes. Not yet reviewed by ACR, OARSI, AAOS, or NICE.
Knee Replacement: 1 RCT vs non-surgical care (Skou et al., NEJM 2015, n=100). NICE NG226: advise considering referral when OA substantially impacts quality of life and non-surgical management has been ineffective. AAOS maintains clinical practice guidelines for surgical management of knee OA.
Hyaluronic Acid: 169 RCTs / 21,163 patients. Generally recommended against. NICE: "Do not offer." AAOS: advises against routine use. ACR: conditional against (knee, first CMC); strongly against (hip). OARSI: conditionally supports for knee — outlier among major bodies.
PRP: 40 RCTs / 3,035 patients. ACR: strongly recommended against. OARSI: strongly not recommended. AAOS: Limited — "may reduce pain and improve function" (exception). NICE: special arrangements only.
Stem Cell / MSC: 28 RCTs; ~467 patients in placebo-controlled subset. ACR: strongly recommended against. OARSI: strongly not recommended. Not addressed by AAOS or NICE.
Note on the LDRT Row
The LDRT randomized evidence divides by dose protocol. Trials using ≤0.5 Gy/fraction are positive or show benefit: LoRD-KNeA (70% vs 42% sham, p = 0.014), Fazilat-Panah 2025 (all outcomes p < 0.01), Makarova et al. (0.45 Gy/fraction; disability 9.5% vs 17.8%; arthroplasty 67% risk reduction; same background therapy both arms), and ArthroRad 2024 (~60% in both dose arms — dose comparison, no sham; LoRD-KNeA confirms ultra-low doses are active, so both ArthroRad arms were active). Trials using the older 1 Gy/fraction protocol (Mahler 2018, Minten 2018) were null. ARS AUC: 60–90% overall response; identifies 1 Gy/fraction as a methodological limitation. No major guideline body (ACR, OARSI, AAOS, NICE) has yet issued a recommendation.
Three Things This Table Shows
1. Guideline strength tracks the size and consistency of the prospective randomized evidence base. The treatments with the strongest endorsements also have the largest bases of randomized controlled trials.
2. A large evidence base is not the same as a favorable one. Hyaluronic acid has 169 randomized trials — the largest injectable base here — and most bodies recommend against it precisely because that evidence is conclusive and unfavorable.
3. Total knee replacement rests on one randomized trial against non-surgical care. Its evidence base is enormous within surgery (implants, approaches, rehabilitation) but the prospective randomized comparative effectiveness data is limited.
Medically reviewed by:Robert Warren Floyd, M.D., Ph.D. — Resident Physician, Radiation Oncology, The University of Texas MD Anderson Cancer Center. Last reviewed: August 2, 2026. Independent personal project — not affiliated with or endorsed by MD Anderson Cancer Center. Funding & independence disclosure.