Quality of Domiciliary Oxygen Therapy in Adults with Chronic Respiratory Diseases—Results of a Multicenter Cross-Sectional Study in Germany

180 000 patients in Germany are living with long term oxygen therapy (LTOT). Studies of the on re-evaluation of domiciliary oxygen therapy in Germany are lacking.

The present cross-sectional study aims to investigate the quality of domiciliary oxygen therapy. The primary endpoint was adherence of required oxygen therapy according to guideline criteria.

Methods

A cross-sectional study was performed in three hospitals with pulmonary departments from June 2019 to April 2020 (MHH ethics committee 8469_BO_K_2019). The study included patients in whom a re-evaluation was undertaken at least four weeks after starting domiciliary oxygen therapy (Clinical Trials Nr. NCT03937622). Oxygen titration was done according to established criteria and followed up by means of capillary blood gas analyses (1). Outpatient study participants with stable illness were recruited in special ambulatory clinics. Inpatients were recruited in a setting of elective hospital stays—for example, to control out-of-hospital mechanical ventilation or for inpatient pulmonary rehabilitation.

We collected data on underlying illness, sex and year of birth, date of oxygen prescription, oxygen consumption at rest and during exercise, duration of use, flow rate at rest, oxygen source, and use of a demand valve.

Oxygen therapy was categorized as oxygen administration at rest and exercise, and oxygen provision exclusively during exercise without consumption at rest.

Patients were investigated for the presence of exercise hypoxemia if exercise studies were undertaken in the context of clinical routine (for example, 6-minute walk tests). Without administration of oxygen during exercise, severe exercise hypoxemia was defined in a setting of minimal pulse oximetry oxygen saturation (SpO₂)<80%, with oxygen administration in SpO₂<85%. If no exercise examination was available or if SpO₂<88% during oxygen administration, patients were assigned to the category “unknown exercise hypoxemia.”

A multivariate analysis was performed as binary logistical regression with the endpoint “adequate oxygen therapy.” Variables with a p value of <0.10 were included in the analysis.

Results

A total of 362 patients (48% female), with a median age of 56 years (25–75% quartile: 50–61 years). The most common underlying disorder was chronic obstructive pulmonary disease (COPD), at 48%. Oxygen was prescribed median 25 (10–52) months before inclusion into the study. The daily duration of oxygen administration as reported by patients was 18 hours (25–75% quartile: 9–24 hours). Twenty-seven percent of patients used a demand valve, 90% used liquid oxygen, and 24% additionally used non-invasive ventilation.

Two hundred eighty-two of 362 patients (78%) reported having oxygen therapy at rest. 63 study participants used oxygen therapy only during exercise; of these, the data of 20 patients were not evaluable. Nine patients reported using oxygen only at nighttime (not evaluable). Eight patients reported not using the oxygen at all. It was therefore possible to evaluate data for 333 ((282+8+43) patients.

In 103 of the 282 patients who reported oxygen administration at rest (37%) the indication for LTOT was confirmed according to the defined criteria of this study (table). None of the 8 study participants not using oxygen had a titrated oxygen flow ≠ 0 L/min. Monitoring oxygen therapy by titration took a median of 16 minutes.

Table. Monitoring of oxygen therapy in the entire study population (N = 362).

Guideline criteria (4)	Study criteria	Patients included in the study*1
Long term oxygen therapy (LTOT)		Reported O2 therapy at rest: 282/362 (78 %)
– In patients with pulmonary hypertension or hyperglobulia: paO2 < 60 mmhg	– In all patients: O2 flow ≠ 0 L/min and paO2  ≥ 60 mmHg	180/282 (64 %)
– In other patients: paO2 < 55 mmhg
– Duration of use at least 15 hours everyday	– Duration of use at least 15 hours everyday (if titrated O2 flow ≠ 0 L/min)	159/180*2 (88 %)
– Flow rate at rest paO2 ≥ 60 mmHg or rise by ≥ 10 mmHg	– Deviation/difference of maximally 1 L/min to titrated flow rate at rest (if titrated O2 flow ≠ 0 L/min)	118/180*2(66 %)
	– All three criteria fullfilled	103/282 (37 %)
Ambulatory oxygen therapy (AOT)		Reported O2 therapy only during exercise 43/362 (12 %)
– SpO2 drop by ≥ 2 % to ≤ 90 % during exercise	– Severe exercise hypoxemia ie SpO2 drop < 80 % during exercise	15/43 (35 %) *1 Not evaluable 29/362 (8 %), Patients not using O 2 therapy 8/362 (2 %) *2 Only patients with a titrated O 2 flow ≠ 0 L/min
– Oxygen saturation of at least 90 % during exercise
– Improved performance ability (eg, increase in walking distance by ≥ 10 %)
– Alleviation of dyspnea, measured using the Borg Scale (1–10, respectively) by at least 1 point

O₂, oxygen; paO₂, partial oxygen pressure; SpO₂, pulse oximetry oxygen saturation

The median patient-reported oxygen flow rate at rest was 2L/min (1–2.5). Among the 180 patients with a flow rate >0L/min the patient-reported flow rate was more than 1 L/min higher in 46 (26%) and in 16 (9%), more than 1 L/min lower than the titrated rate.

In the multivariate analysis, prescription by pulmonologists (OR: 1.67; 95% confidence interval: [1.05; 2.82]) and underlying lung fibrosis (OR: 1.67 [1; 2.77]) were associated with a continued obviously required oxygen therapy.

Discussion

Follow-up re-evaluation of oxygen therapy, mostly in patients with COPD, were internationally classed as unsatisfactory. In a Swedish analysis of 8379 COPD patients, the rate of controls of oxygen therapy within three weeks after discharge from hospital was only 60% (2).

In about 15000 COPD patients having LTOT between 2001 and 2010, the therapy was stopped within 6 months in 25.5% (3). In our study, no obvious need for oxygen at rest existed in 36% at follow-up/control, although the patients did use it.

Guidelines (4) recommend follow-up controls for LTOT in patients with COPD within several weeks. Our study found no difference in the proportion of adequate O₂ therapy between COPD and other diagnoses.

A Canadian study compared prescribed duration of use and flow rate with actually registered administration rates in 115 patients with COPD who were having LTOT and found that only 60% of patients adhered to the treatment (5). By comparison, 88% of our patients who used oxygen ≥=15 hours followed our therapeutic recommendation. Adherence to oxygen therapy is under the joint influence of patients and medical staff and can be improved through training and follow-up.

A limitation of our study is the fact that data on the duration of use are based exclusively on patients’ own reports. The population under study was not representative for the total number of patients in primary care. We therefore were not able to determine whether the participants had ever been followed up for their oxygen therapy before being included in our study.

The monitoring of oxygen therapy in the setting of our study was done in accordance with valid guidelines, surrogate parameters for the indication for LTOT and ambulatory oxygen therapy were used. The rate of patients with a titrated oxygen flow of 0 L/min in order to achieve a partial oxygen pressure (paO₂) of 60 mm Hg can be assumed to be even higher, if arterial probes are used. Capillary specimens may lead to underestimation of pO₂ by 4–8 mmHg in comparsison to arterial probes. In sum, our study shows that domiciliary oxygen treatment has to be followed up on a regular basis and that oxygen provision in the domestic setting is probably excessive. On the basis of these data it makes sense to reimburse the additional expense incurred by monitoring oxygen therapy.