Inhaled Oxygen and Cluster Headache Sufferers in the United States: Use, Efficacy and Economics

Results From the United States Cluster Headache Survey

Todd D. Rozen, MD, FAAN; Royce S. Fishman, BA


Headache. 2011;51(2):191-200. 

In This Article


A total of 1134 individuals completed the survey. There were 816 male responders and 318 female responders. Among them 868 patients had episodic CH while 266 had chronic CH. Ages of survey responders were as follows: less than 20 years (7 responders), 21–30 years (137 responders), 31–40 years (311 responders), 41–50 years (384 responders), 51–60 years (238 responders), and 61 years plus (57 responders). Every state was represented in the survey.

A. Oxygen Usage

Ninety-three percent of the surveyed patients were aware of oxygen being an acute therapy for CH; however, 34% had never tried oxygen. The younger the age of the CH patient the less likely they were to have tried oxygen. Seventy-one percent of those surveyed who were under the age of 20 years had never tried oxygen therapy vs only 24% of those who are currently between 51 and 60 years of age. Only 50% of the surveyed population had tried oxygen alone to abort a CH. In addition, only 25% were currently using oxygen as a sole abortive greater than 80% of the time. Of those who had utilized oxygen 81% stated oxygen was started as acute treatment for their CH months to years after the initial diagnosis of CH was made. At the time of the survey only 25% of CH patients were using oxygen to treat their CH.

B. Who Prescribed Oxygen and Reasons Why Oxygen was Never Prescribed

There was an equal distribution (28% each) of physician type (general practitioner, general neurologist, headache specialist) who initially prescribed oxygen. Forty-four percent of the surveyed patients had to first suggest oxygen therapy to their physician to get it prescribed. About 12% of physicians refused to prescribe oxygen for their CH patients. Reasons cited included: did not believe it would work (44%), did not know that oxygen was used to treat CH (32%), and stated that the medical literature was not convincing enough to prescribe oxygen (16%).

C. Prescribing Patterns for Oxygen

Fifty percent of the survey responders stated that when oxygen was prescribed to them for the first time they received no information on how to conduct or perform oxygen therapy. If they did get some training only 15% received it from their treating physician while 44% received instruction from a home care/oxygen delivery service. On the prescription itself only 45% of prescribers specified a specific flow rate for oxygen. Only 50% of the prescriptions noted CH as the diagnosis and only 28% specified a specific delivery system/mask type. Once prescribed only 55% of patients were told by their physicians where to go to get their oxygen. The remainder used the phone book, internet or found out how to obtain oxygen from peers who had CH. More than 50% of survey responders stated that it was difficult to actually find a source for their oxygen.

D. Oxygen Delivery Systems Prescribed

About 11% of patients were using a nasal cannula to deliver oxygen while 29% were using a face mask without a non-rebreather bag system. Only 53% were actually using the preferred non-rebreather face mask system. Seven percent were not using a mask or cannula although what was being used was not specified.

E. Oxygen Flow Rates Utilized

The oxygen flow rates that were initially prescribed to the patients by physicians were as follows: 7 L/minute 23%, 8 to 12 L/minute 51%, 13 to 15 L/minute 18%, and 16 L/minute and above 8%. The surveyed patients did not always adhere to the initial prescribed oxygen flow rates as 17% used 7 L/minute, 25% used 8 to 12 L/minute, 25% used 13 to 15 L/minute and 13% used 16 L/minute or greater. In regard to current oxygen flow rates being used by survey responders during an individual CH attack: 41% start and 34% end using 7–10 L/minute, 17% start and 14% end using 11–12 L/minute while 28% start and 52% end using 13 L/minute or higher flow rates (Table 1). Most CH patients will raise oxygen flow rates during CH treatment to try to achieve complete relief from pain.

F. Efficacy of Inhaled Oxygen (Table 2)

Seventy percent of survey responders stated that inhaled oxygen was an effective treatment for their CH. Oxygen effectiveness did not vary by age class: ages 21–30 years (70% stated effective), ages 31–40 years (73% stated effective), ages 41–50 years (70% stated effective), ages 51–60 years (69% stated effective), and ages 61 plus years (67% stated effective). Under age 20 years oxygen was 100% effective but only 2 individuals answered this survey question so not a significant result. The efficacy of oxygen appeared to be fairly consistent regardless of the number of CH attacks per day: 69% of patients with 1–2 attacks per day stated oxygen was effective vs 72% of those with 7–8 attacks per day. The highest efficacy was in those with 5 to 6 attacks per day (75% stated efficacious) but there was no statistical significance difference in the number of attacks per day and response to oxygen. In regard to the time on inhaled oxygen (whatever flow rate that patient would individually utilize) to have complete relief of CH pain: 17% stated between one and 10 minutes, 34% between 11 and 20 minutes, 22% between 21 and 40 minutes, and 27% over 40 minutes. Thus, in 49% of patients oxygen took 21 minutes or longer to provide complete head pain relief. The CH patients at the extremes of age seemed to have a faster response to inhaled oxygen as the largest percentage of patients aged 21–30 years and 61 plus years could fully abort a CH within 6–10 minutes after oxygen initiation compared to those aged 31 to 60 years who needed on average 11 to 15 minutes to abort a CH. Only one patient under age 20 responded to this survey question so the number is too small to interpret. The majority of surveyed CH patients (75%) do not use oxygen alone to abort a CH attack. Only 25% use oxygen as a sole abortive greater than 80% of the time. If oxygen is used with another abortive medication (triptans or dihydroergotamine [DHE] or other) over half (55%) will use the additional abortive when they realize that oxygen will not take away that particular headache, while 38% will take the other abortive before using oxygen. When using oxygen plus another abortive 52% of survey responders will have complete headache relief within 20 minutes of taking the combined abortives while only 13% need more than 45 minutes to have relief compared with 27% when using oxygen alone. Interestingly, 37% of the surveyed patients had complete relief of their CH within 15 minutes of using combined treatment vs only 28% when a non-oxygen abortive was used alone. The abortives that were used in combination with inhaled oxygen included: injectable sumatriptan 36%, triptan nasal spray (sumatriptan or zolmitriptan) 12%, oral triptan (sumatriptan, zolmitriptan, rizatriptan) 18%, DHE 2% and other non-specified abortive 32% Table 2.

G. Cluster Headache Subtype and Oxygen

There were some differences in oxygen response between episodic and chronic CH patients. Seventy-three percent of episodic CH patients stated that inhaled oxygen was effective vs 62% of chronic CH patients and this was a statistically significant difference (P < .006). Thirty-six percent of episodic CH patients had never tried oxygen vs 28% of chronic CH patients. Chronic CH patients were more likely to be currently using oxygen to treat CH (34%) than episodic CH patients who were in an active cycle (23%). Patients with episodic CH seem to get complete relief of CH faster with inhaled oxygen than chronic CH patients as 38% of episodic CH patients had complete pain relief within 15 minutes of oxygen treatment vs 28% of chronic CH patients which is a statistically significant difference (P < .02).

H. Gender and Oxygen

Almost the same percentage of male and female CH patients stated that oxygen was an effective therapy (70% vs 72%). Male CH patients appear to respond faster to inhaled oxygen than female CH patients. Nineteen percent of male CH patients had complete relief of a CH on oxygen after 10 minutes or less of treatment vs 11% of the female CH patients surveyed and this was a statistically significant difference (P < .05). Fifty-one percent of female CH patients start with oxygen flow rates of 7 to 10 L/minute while 45% of men start at 13 L/minute or higher, and thus this may have led to faster response times in men.

I. Smoking History and Oxygen

Smoking history did not appear to alter the effectiveness of inhaled oxygen. Seventy-eight percent of surveyed patients who had no personal smoking history and no secondhand smoke exposure from their parents stated that oxygen was effective compared to 70% of the entire population surveyed as a whole of which 73% had a personal smoking history. Of current oxygen users 96% of patients with a smoking history stated oxygen was an effective abortive vs 97% of non-smoking patients.

J. Economics of Oxygen Usage

In regard to the costs of oxygen therapy for CH, 65% of surveyed patients stated that their costs were under $1000 per year, while 31% had costs between $1000 and $6000 per year and 2.5% of the surveyed population spent between $8000 and $12,000 per year on oxygen. These costs were before reimbursement from insurance. Complete out of pocket expense after insurance reimbursement showed the majority (87%) spent under $1000 per year, while 13% spent between $1000 and $12,000 per year.

At present 64% of survey responders stated that their medical insurance covers oxygen therapy for CH. Seventy-six percent of the same surveyed group stated that injectable sumatriptan is covered by their insurance. Sixty-one percent noted that it was not difficult to get reimbursement from their insurance carrier for oxygen, while 7% stated it was very difficult. In order to get reimbursement 44% stated that the patient or their physician had to submit medical literature to the insurance carrier. Fifty-one percent of the CH patients in the United States stated they can easily afford their oxygen while 16% stated it is unaffordable. Twelve percent stated they have purchased non-medical grade oxygen (welders or research grade) and the main reasons for doing this were: that the non-medical grade oxygen was a lower cost than commercial grade oxygen (66%), cannot afford to go to the doctor to get a prescription for oxygen (22%), and my physician will not prescribe oxygen (18%).


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.