Advice
Clinical and technical evidence
Clinical and technical evidence
A literature search was done for this briefing in accordance with the interim process and methods statement. This briefing includes the most relevant or best available published evidence relating to the clinical effectiveness of the technology. Further information about how the evidence for this briefing was selected is available on request by contacting mibs@nice.org.uk.
Published evidence
Literature searches found 3 published studies on the Nasal Alar SpO2 sensor which are summarised in this briefing, involving 135 people in the US. These include a published accuracy and feasibility study in healthy volunteers (Morey et al. 2014), 2 published conference abstracts of an accuracy study in surgical patients (Melker et al. 2014) and a comparative study with the Nellcor forehead sensor in critical care patients (Schallom et al. 2016). Table 1 summarises the clinical evidence as well as its strengths and limitations.
Overall assessment of the evidence
The evidence for the Nasal Alar SpO2 sensor is limited in quantity and quality, and includes 1 published study and 2 conference abstracts. Most studies focussed on confirming the accuracy of the sensor using radial artery co-oximetry samples, compared with finger or forehead sensors. Other outcomes reported included the time taken to detect oxygen desaturation and the incidence of pressure ulcers. Two out of 3 studies were in clinical settings with patients, one of which studied people with low peripheral perfusion.
Because all the studies were done in the US, and some studies included healthy volunteers, this may limit their relevance to the NHS care pathway. Randomised studies comparing the Nasal Alar SpO2 sensor with other non-digit pulse oximeters done in a UK setting would be useful to confirm its equivalence for diagnostic accuracy and effects on patient outcomes.
Table 1 Summary of published evidence on the Nasal Alar SpO2 sensor
|
Study size, design and location |
Accuracy and feasibility study in 12 healthy volunteers in a non-clinical setting in the US. |
|
Intervention and comparator(s) |
Nasal Alar SpO2 sensor and conventional finger sensor. Subjects breathed hypoxic mixtures of fresh gas by a facemask to achieve SpO2 ranges of 70% to 100%. SpO2 was measured by both nasal and finger sensors and compared to the reference standard: traditional co-oximetry from radial artery samples. |
|
Key outcomes |
The Nasal Alar sensor was accurate to within ±2% for the full range of SpO2 levels when compared with the reference samples. Bias, precision and ARMS over a range of 70% to 100% were significantly better for the nasal sensor compared with the finger sensor. The mean bias for the nasal and finger sensors was 0.73% and 1.90% (p<0.001) respectively, with corresponding precision values of 1.65 and 1.83 (p=0.015) and ARMS values of 1.78% and 2.72% (p=0.047). |
|
Strengths and limitations |
Single-centre non-randomised and non-blinded study. Small sample size in healthy volunteers in non-clinical setting. Manufacturer-funded study: authors were employees or held equity shares in Xhale. |
|
Study size, design and location |
Observational study in 80 surgical patients in the US. |
|
Intervention and comparator(s) |
Nasal Alar SpO2 sensor (placed on both alae) and conventional finger sensor. Either matched pulse oximeters were used to record data from the 2 ala and finger, or an alternative stand-alone oximeter was used on 1 ala. Simultaneous finger oximetry data was collected from a multi-para meter patient monitor. |
|
Key outcomes |
Desaturations were present in 15 patients (19%): alar desaturation happened on average 9 seconds (range –5 to 33) sooner than the finger with the same oximeter (physiologic delay). Alar desaturation measured with an alternative stand-alone oximeter averaged 7 seconds slower than those from the first oximeter (device delay). The multi-parameter patient monitor introduced a further 5 seconds average delay. In all, a combination of physiologic delay and device delay results in an average of 14 seconds delay between measurement at the finger and the first measurement at the nasal ala. |
|
Strengths and limitations |
Single-centre non-randomised and non-blinded study. Peer-reviewed accepted conference abstract, but not published as a full article so details of methodology are unclear. Manufacturer-funded study: authors are employees or hold equity shares in Xhale. |
|
Study size, design and location |
Observational study in 42 critically ill patients with poor peripheral perfusion in the US. Patients were included if a peripheral signal was not able to be obtained, were on vaspressors or had a reduced temperature. All patients had arterial lines. |
|
Intervention and comparator(s) |
Nasal Alar SpO2 sensor and Nellcor forehead sensor. Arterial samples were measured by co-oximetry and values recorded from both sensors at time 0, 24, and 120 hours. Skin was assessed every 8 hours with relocation of the sensor to the opposite nasal ala or forehead side. sensor was removed when skin injury was seen. |
|
Key outcomes |
More measures were within 3% of co-oximetry values for nasal Alar (56%) compared with forehead (44%) sensors. Measurement failures were 6% for nasal ala and 21% for forehead. Mean wear time was 66.2 hours for nasal sensor and 37.4 hours for forehead sensors. 13 patients developed a pressure injury with the forehead sensor (9 at stage 1, 3 at stage 2 and 1 deep tissue injury) and 3 with the nasal sensor (2 stage 1, 1 stage 2). |
|
Strengths and limitations |
Peer-reviewed accepted conference abstract, but not published as a full article so details of methodology are unclear. Single-centre non-randomised and non-blinded study. Clinical setting in a relevant patient population. High drop-out rate: only 14 people had all 3 measurements, as 51% died before data collection completion. Prevention of sensor-related pressure injury may not be possible in all critically ill patients. |
|
Abbreviations: ARMS, accuracy root mean square error: calculated as the square root of (bias2 + precision2) with values of less than 2% to 3% considered to be acceptable. |
|
Recent and ongoing studies
-
Evaluation of the Xanas nasal pulse oximetry sensor during robot assisted laparoscopic prostatectomy. Primary comparator: ear sensors. Enrolment: 64. Completion date: 2017. Location: UK (abstract presented).