Teguh
Pre-emptive treatment with hyperbaric oxygen following radiation therapy for head and neck cancer may prevent the onset of late radiation tissue injury.
| Clinical bottom line: 1. Early treatment with HBOT was associated with improved Quality of Life scores for a wide range of symptoms including dry mouth, swallowing difficulties and pain in the mouth 2. Early treatment with prophylactic HBOT may prevent the onset of late radiation tissue injury. |
Citation: Teguh DN, Levendag PC, Noever I, Voet P, van der Est H, van Rooij P, Dumans AG, de Boer MF, van der Huls MPC, Sterk W, Schmitz PIM. Early hyperbaric oxygen therapy for reducing radiotherapy side effects: early results of a randomised trial in oropharyngeal and nasopharyngeal cancer. Int J Radiation Oncology 2009; 75(3):711-716.
Corresponding author name and Email: Peter Levendag, [p.levendag@erasmusmc.n
Three-part Clinical Question:For patients who have received radiation therapy for head and neck malignancies, does early treatment with hyperbaric oxygen, versus no specific treatment, prevent late radiation tissue injury?
Search Terms:radiation tissue injury, xerostomia, head and neck cancer
The Study: Non-blinded randomised controlled trial with intention-to-treat.Adult patients diagnosed with oropharyngeal or nasopharyngeal cancer and treated with radiotherapy (46 to 70 Gy).
Control group: (N = 11; analysed): No specific therapy
Experimental group: (N = 8; analysed): 100% oxygen breathing at 2.5 ATA for 90 minutes daily, five days per week to a total of 30 treatments starting within two days after completion of radiotherapy.
The Evidence:
'Outcome at'12 months |
Control group Mean |
HBOT group Mean |
P-value |
|
EORTC H&N35: Dry mouth Swallowing |
90 40 |
30 7 |
0.009 0.011 |
|
VAS (Pain in mouth) |
7 |
1 |
<0.001 |
EORTC H&N35: European Organization for Research and Treatment of Cancer Quality of Life assessment tool for head and neck symptoms, 0 = no symptoms, 100 = worst imaginable symptoms;
VAS: Visual Analogue Scale, 0 = no pain, 10 = maximal pain imaginable.
Comments:
1. Very small trial because of slow recruitment rate, and may be subject to considerable bias. In particular, the baseline QoL scores were generally worse in the control group and may explain the observed differences at follow-up.
2. All outcomes are self-reported in non-blinded subjects.
3. No sham treatment given. Authors state that this is unlike to bias the late treatment effect seen at 18 month follow up.
4. All results given in graphic format and p-values are from regression analysis based on maximum likelihood estimation using Stata 9 software.
Appraised by: Danielle Wood and Michael Bennett, Prince of Wales Hospital; Saturday, 16 April 2016
Email: danspace@gmail.com
