This is a randomized controlled trial with an allocation ratio of 1:1. Half the patients are randomized to continuous positive airway pressure (CPAP) and half to routine medical care. Included are one hundred and twenty patients aged 18-80 years plus patients with ongoing CPAP scheduled for abdominal surgery at Umeå university hospital. The primary outcome is oxygen partial pressure (PaO2) measured on postoperative day 2 compared with the day before surgery. Secondary outcomes include diffusion capacity for carbon-monoxide, vital capacity, FEV1 and carbon-dioxide partial pressure (PaCO2) on postoperative day 2 compared with the day before surgery. PaO2 and PaCO2 are recorded from blood gas measurements obtained from the radial artery. Percentage of nocturnal hypoxia defined as the percentage of oxygen saturation during 90% of the second postoperative night. Tolerance to CPAP measured by the number of hours used CPAP. Side effects related to CPAP. All patients are examined with a simplified sleep apnea examination (Noxturnal T3, Res Med) the night before surgery.

Patients in the CPAP treated group are given an auto-CPAP with a minimum pressure of 5 cm and a maximum pressure of 10 cm. They will be treated with CPAP for at least 2 hours immediately after surgery and during the first two postoperative nights. Oxygen is supplied to CPAP if oxygen saturation falls below 90%. Patients in the control group receives standard treatment and supplemental oxygen if oxygen saturation falls below 90%. In a third arm, we will include patients who already are using CPAP at night at home for previously diagnosed obstructive sleep apnea. are instructed to use CPAP for 2 hours immediately after surgery and subsequent nights during hospital stay.

Theory Research question: Does nasal CPAP abolish postoperative hypoxia and lung function decline after abdominal surgery?

Methods This is a randomized controlled trial with an allocation ratio of 1:1. Half of the patients are randomized to CPAP and half to routine medical care. Randomization is generated by a computer program and handled by a person outside the study. Randomization procedure is documented by the contact person and in the patient's study protocol.

Include are one hundred and twenty patients age 18-80 years old, scheduled for abdominal surgery at Umeå university hospital. Excluded are patients with American Society of Anesthesia (ASA) class 2-3 and patients with cognitive impairment or dementia, unable to participate. Patients with ongoing CPAP treatment are not randomized. Instead, they form a separate group with ongoing CPAP after surgery

The primary outcome is oxygen partial pressure (PaO2) measured on postoperative day 2 compared with the day before surgery.

Secondary outcomes include diffusion capacity for carbon-monoxide, vital capacity, FEV1 and carbon-dioxide partial pressure (PaCO2) on postoperative day 2 compared with the day before surgery.

PaO2 and PaCO2 are recorded from blood gas measurements obtained from the radial artery. Percentage of nocturnal hypoxia defined as the percentage of oxygen saturation during 90% of the second postoperative night. Tolerance to CPAP measured by the number of hours used CPAP. Side effects related to CPAP.

All patients are examined with a simplified sleep apnea examination (Noxturnal T3, Res Med) the night before surgery.

A need of 35 patients in each arm was estimated to detect a difference in mean (SD) in PaO2 of 1 (1.5) kilo Pascal with a significance of p<0.05 and a power of 80%. Due to dropouts, it is estimated a need to include and randomize 120 patients.

Patients in the CPAP treated group are given an auto-CPAP that can increase the pressure during respiratory arrest (AirSense 10 AutoSet, ResMed Inc) including a silicone nasal mask. It is set with a minimum pressure of 5 cm and a maximum pressure of 10 cm during instruction on the day before surgery. Pressure and mask are tested by trained personnel at the Lung section so that both pressure and mask are comfortable and can be tolerated by the patient. Immediately after surgery they will be given CPAP for at least 2 hours immediately after surgery and during the first two postoperative nights. Oxygen is supplied to CPAP if oxygen saturation falls below 90%.

Patients in the control group receive standard treatment and supplemental oxygen if oxygen saturation falls below 90%

Included in a third arm are patients who already are using CPAP at night at home for previously diagnosed obstructive sleep apnea. They are instructed to use CPAP for 2 hours immediately after surgery and subsequent nights during hospital stay.

Inclusion Criteria:

• Adults (age 18-80 years) scheduled for elective abdominal surgery
• Must be able to tolerate nasal CPAP

Exclusion Criteria:

• American Society Anesthesia (ASA) Class 3-4.
• Cognitive impairment or dementia