ReAcT CO2: A Study to Assess TcCO2 to Guide Acute NIV

Patients with breathing difficulty need arterial blood sampling to check carbon dioxide levels. This is difficult, painful, associated with complications and often met with delays. The investigators have shown in a previous observational study, that using a skin probe to monitor carbon dioxide levels in patients with acute breathing difficulty was far more comfortable, significantly less painful and importantly closely reflected arterial carbon dioxide levels. Decisions regarding breathing support with mask (noninvasive ventilation or NIV) were however based on arterial blood samples collected simultaneously.

In this feasibility study over 72 hours, patients will either have arterial blood sampling or skin probe to monitor carbon dioxide levels.

The aim of the study is to determine whether continuous carbon dioxide monitoring with a skin probe reduces the need for arterial blood sampling by at least 30%. Investigators will also study the safety and effectiveness of skin probe monitoring to manage NIV.

Towards this the investigators will study patients admitted with sudden worsening of breathing that has led to a build up of carbon dioxide. Patients will be recruited within 8 hours, although preferably within 4 hours of starting NIV, preferably in the Emergency Department(ED). Their carbon dioxide levels will then be monitored by skin probe (TcCO2 group) or arterial blood gas (ABG) sampling.

Clinicians can perform ABGs in TcCO2 group at their discretion. Our main end point is the total number of ABGs performed in each group. Clinicians will complete a questionnaire for unscheduled ABGs in TcCO2 group.

The investigators will also measure any side effects, pain experienced by patients in each group and ease of using each technique as well as explore barriers to using transcutaneous monitoring and identify any drawbacks of this method.

• Trancutaneous CO2 monitoring (using TCM5 device) (Device): Patients in the Transcutaneous arm will undergo Trancutaneous CO2 monitoring (using TCM5 device)
• Arterial Blood Gas (ABG) sampling (Other): Arterial Blood Gas (ABG) sampling to monitor CO2 - as per standard practice.

Primary outcomes

• Average number of arterial blood gases per participant over the study period [over the study period of 72 hours per participant]: the total (Scheduled and unscheduled) arterial blood samples over 72 hours per participant in each group .
• number of unscheduled arterial blood sample in each group [over the study period of 72 hours per participant]: arterial blood sampling needed for clinical deterioration or due to unreliability of transcutaneous CO2 monitoring for each patient
• number of instances where transcutaneous monitoring was unreliable and arterial blood gas sampling was needed. [over the total study period of 72 hours per patient]: Clinicians will complete questionnaires regarding the need for unscheduled arterial blood samples and changes in management due to these incidences of arterial blood sampling in the transcutaneous monitoring group.

Secondary outcomes

• Pain and any discomfort from arterial blood sampling and transcutaneous monitoring using a visual analogue scale from 0 to 10 with 0 being no pain and 10 being the worst pain they have ever experienced [over the study period of 72 hours per participant]: Pain and any discomfort (such as skin burns, blisters, erythema, bruising or restriction of physical activity) from arterial blood sampling and transcutaneous monitoring
• Proportion of patients admitted with acute hypercapnic respiratory failure that are suitable and agree to participate in the study over one year (to help guide future studies). [Over the total study period of 1 year.]: Suitability and feasibility
• Proportion of patients who achieve 50% drop in carbon dioxide in 12 hours and those who achieve carbon dioxide level of 7 kPa or less in 72 hours in each group. [over the study period of 72 hours per participant]: Proportion of patients who achieve 50% drop in carbon dioxide in 12 hours and those who achieve carbon dioxide level of 7 kPa or less in 72 hours in each group.
• Proportion of patients who need further changes to ventilator settings beyond 4 hours and 24 hours [over the study period of 72 hours per participant]: How quickly can we optimise NIV settings
• Proportion of patients with deterioration in CO2 at 4 and 24 hours after NIV initiation [over the study period of 72 hours per participant]: Can TcCO2 identify deteriorating patients quicker
• Proportion of patients who get better and are discharged and proportion that fail to respond and either need mechanical ventilation / progress to palliative care. [over the study period of 72 hours per participant]: Proportion of patients who get better and are discharged and proportion that fail to respond and either need mechanical ventilation / progress to palliative care.
• Transcutaneous arm only - agreement between arterial and transcutaneous readings: [over the study period of 72 hours per participant]: a) absolute CO2 values obtained within 30 minutes of each other will be tested using Bland Altman plot; b) trend between serial ABG CO2 values and transcutaneous CO2 readings when ABGs performed (scheduled and unscheduled ABG will be assessed separately).
• Discrete choice: what method would patients prefer for monitoring during similar future illness requiring NIV? [over the study period of 72 hours per participant]: Do patients prefer one method of monitoring over the other. This is reasonable to ask all those randomized to TcCO2 as they will experience both ABG and TcCO2
• Clinician Preference questionnaire: whether they prefer ABG versus transcutaneous CO2 monitoring [over the total study period of 1 year]: Taking into consideration time and difficulty, plus certainty of implementing changes to NIV settings and weaning, with respect to modality of carbon dioxide monitoring.
• Time needed to wean off NIV in each group i.e. time from documented clinician decision to wean, until time of cessation of NIV. [over the study period of 72 hours per participant]: This would help assess whether transcutaneous CO2 monitoring allows for quicker weaning from NIV.
• Basic Health economics data will be gathered to assess whether transcutaneous monitoring is a cost effective way compared to arterial blood gas sampling. [over the total study period of 1 year]: Cost effectiveness of the two methods to monitor CO2

Sex: All, ages: 18 Years to any

Inclusion Criteria:

* Patients aged18 years and over with hypercapnia, requiring acute NIV and who are admitted to the Emergency Department or Medical HDU and, who need arterial blood sampling for clinical reasons.
* Recruited ideally within 4 hours (but no later than 8 hours) of initiation of acute NIV in hospital.

Exclusion Criteria:

* Reduced level of consciousness or agitation.
* Metabolic or mixed acidosis, serum bicarbonate < 24 mmol/L.
* Haemodynamic instability or reduced skin perfusion.
* Patients requiring more than 6litres/min (ormorethan60%) inspired oxygen through ventilator to maintain SpO2 of 88-92%.
* Need for multi-organ support e.g. inotropes or vasopressors or renal replacement therapy
* Arterial pH equal or less than 7.1 or carbon dioxide levels >15kPa as they are at risk of NIV failure and likely to require mechanical ventilation.
* Participating in any other interventional study