Parallel Session 8 - Towards Excellence through Evidence-based Practice II - Safety Matters

Introduction:
A variety of methods have been used to assess Nasogastric tube (NGT) placement. Only gastric aspirate pH checking  5.5 & X-ray imaging are considered as gold standard method and widely accepted as the most reliable test. The study will collect information to demonstrate the reliability of colorimeter testing/ capnography monitor. It helps to serve as an additional screening tool for identification of tube placement, when the patient fails to collect gastric aspirate for testing. 


Objectives:
To ascertain the accuracy of colorimeter/ capnography, with radiographic examination as the reference standard; as well as to assess the time lag from change of NGT to resume feeding of patient.


Methodology:
This is a prospective descriptive study. 71 samples are recruited from three Medical Extended wards by convenience sampling for colorimeter test. 12 samples were recruited from old age homes for a pilot study on capnography monitor with 15 trails of NGT insertion. The colorimeter test/ capnography monitor does not involve any invasive procedures & will not cause any short-term or long-term discomfort to the patient. Without getting gastric aspirate, by clinical judgement found no tube coiling inside the mouth, positive auscultation “Whoosh Test”, and the colorimeter test/ capnography monitor is performed. Then, comparing of result with X-ray imaging without leading to skipping procedures in usual clinical practice. (The Joanna Briggs Institute, 2016). 


Result & Outcome:
There were 26 patients in the study with a mean age of 74.07 years (SD = 16.33). Fifteen (57.7%) of them were male. The final sample consisted of 71 doubtful NGT placements and none of them was misplaced. The specificity of CO2 colorimeter was 98.6%, and the negative predictive value was 100% while the sensitivity and positive predictive value were unable to determine. The patient waiting time required for X-ray imaging verification ranged from 51 to 1095 minutes (mean = 380.75 mins., SD = 179.51). 
15 trails of NGT insertion performed for the pilot study of capnography monitor. The ETCO2 waveform were all present at nostril area before NGT insertion; none of the episodes detected ETCO2 and crisp waveforms after verify of tube placement in the stomach by the “Whoosh test”.




Conclusion:
CO2 colorimeter has high specificity and negative predictive value in verification of doubtful NGT placement. Using it to replace X-ray imaging verification may speed up resumption of NGT feeding. Meanwhile, further study to enhance the diagnostic accuracy of ETCO2 detection for inpatient in general ward is needed.

Introduction
In 2020, it was estimated that 5.2% of total Hong Kong population suffered from Diabetes Mellitus (DM). The majority of DM patients who required Community Nursing Services (CNS) for empowerment of DM self- management were frail elderly. Hypoglycaemic incidents in this patient group were at greater risk of causing severe adverse events, including functional decline, repeated admission and even death if problems unaddressed. 
Continuous glucose monitoring (CGM) system refers to a sensor continuously measuring glucose levels in the interstitial fluid for a long period of time. Patients’ glucose profile can be shown in the reader instantly. Corporate CNS teams have implemented a pilot project of using this painless smart device to enhance support for these high-risk DM elderly from June 2022 to February 2023. The aim of this study was to examine the feasibility of picking up asymptomatic hypoglycemic episodes followed by timely interventions by using the CGM system. Individualized nursing care plan would be formulated according to patients’ glucose profile for empowering their DM self-management skills. 
Objectives
1. To improve patients’ glycemic control 
2. To detect asymptomatic hypoglycemia (Hypo G) 
3. To empower patient’s DM self-management skills 
4. To prevent avoidable emergency admission due to poor DM control by early medical-nursing intervention
Methodology
A prospective study was conducted to review the variations of recruited patients’ glucose profiles and DM self- care abilities after eight weeks DM empowerment program. DM patients aged 60 or above without care by DM specialist clinic would be recruited. Patients’ self-monitoring blood glucose (SMBG) profile and the 14-item Summary of Diabetes Self-Care Activities Assessment (SDSCA) were assessed in pre-intervention phase as baseline information. Two sets of CGM were applied at the first week and the 7th week with a duration of 14 days each time. Community nurses provided personalized lifestyle modification and/or seeking consultation from specialists according to CGM results. The primary outcome was the change of patients’ glucose profile in terms of time in range (TIR: 4-10mmol/L). Secondary outcomes included (1) hypoglycaemia (hypo G) events (self-reported reading < 4 mmol/L from CGM/SMBG data); (2) time below range (TBR: hypo G occurrence consisting of 2 levels. 1. Low level: 3.9-3 mmol/L. 2. Very low level: 13.9 mmol/L). At the end of 6th week, post intervention SDSCA was conducted to compare the differences in patients’ DM self-care abilities. Pre-post intervention HbA1C comparison was conducted. Furthermore, patients’ accident & emergency attendance and unplanned admission related to DM complication would be monitored during the program. 
Result
Demographic data
Program completed at the end of February 2023, a total of 68 patients (33 females and 35 males; mean age: 77.4 years, range: 60-99 years) completed eight weeks’ DM empowerment program. Their duration of DM ranged from 1 to 37 years. Blood glucose of 44 patients (64.7%) were controlled by insulin +/- oral antidiabetic medication. 
Outcomes
The mean TIR was improved by 1.8%, from 62.3% to 64.1% (n=37) (p=0.464). 27 out of 37 TIR increased patients (73%) were improved by more than 5%, which was indicating clinically beneficial results with each 5% increase in TIR according to ADA recommendation.
Pre and post HbA1C results were available in 44 patients (64.7%). Mean of HbA1C was reduced by 0.4%, from 8.3% to 7.9% (p=0.017) after intervention. 
19 patients (27.9%) reported a total of 33 hypo G episodes detected by SMBG in pre-intervention phase. On the contrary, CGM system detected 54 patients (79.4%) experienced hypo G with total 425 episodes during the 1st application; in which 60% (n= 255) of asymptomatic hypo G occurred at midnight and during fasting. 
In evaluating the effect on time below and above target glucose range, mean TBR was reduced by 1.1%, from 6.9% to 5.8% while mean TBR very low level was reduced by 1.6%, from 2.7% to 1% (p = 0.036). The mean TAR was reduced by 0.7%, from 30.8% to 30.1% while mean TAR very high level was reduced by 2.7%, from 12.7% to 9.7% (p=0.111).
By comparing pre and post SDSCA items to assess patient’s crucial DM self-care activities, including general and special diet, exercise, glucose monitoring, foot care, medication and hypo G management, the program showed that patients reported significant improvement in all of the self-care activities (p< 0.005) except medication since high scores were observed in both phases.
35 early follow up and adhoc consultation were initiated by CN for suboptimal glycemic control. All these interventions were timely taken during CGM application. None of these patients had accident & emergency attendance or unplanned admission due to DM condition during the 8 weeks program.


Conclusion
This CQI project demonstrated effectiveness of smart devices to detect asymptomatic hypoglycemic conditions compared to the conventional practice. Furthermore, CGM result provided a precise and clear information, in which increased community nurses’ alertness for initiating consultation and conducting individualized patient care plan timely. Also, patients were markedly motivated by community nurses showing improvement in DM self-management with device assistance. Moreover, their blood glucose controls were improved as evident by narrowed down the glucose fluctuation and the post intervention HbA1c were reduced significantly.
No patient attended AED or required hospitalization for poor diabetic condition during program period. 
The overall results might not be statistically significant due to the small sample size. However, the project achieved a milestone in delivering safer patient care with smart technology for high-risk elderly in community care.

Introduction
Extracorporeal Membrane Oxygenation (ECMO) is a life-saving technology. Cannulation of major blood vessels is the first step and physicians must be competent in this procedure. Any procedural error may result in fatal complications. However, acquisition of such skill in early days was difficult due to a lack of platform for deliberate practice. 


Methods
We designed a prototype model for ECMO cannulation training in 2016. The cannulation model, made of platinum-catalyzed silicones and psyllium husks powder, allowed ultrasonography imaging and multiple puncture with large-bore cannulas. The model was later modified to allow conduction of different simulation scenarios, including bedside percutaneous cannulation and extracorporeal cardiopulmonary resuscitation (ECPR). In COVID-19 pandemic, infection control restrictions had prohibited face-to-face ECMO simulation training. We piloted the use of telesimulation to link instructors and simulators to off-site learners in an overseas ECMO training. 


Results
Our cannulation model is the first cannulation model in the world which can be operated as a partial task trainer for percutaneous ECMO cannulation or incorporated into a scenario-based simulation training environment. The model had been used in physician training courses in Hong Kong, Qatar, Japan, Singapore and USA. In a pilot run involving 78 Hong Kong clinicians, mean product evaluation rate was 4.53 out of 5 (a higher rate implied higher realism). Mean confidence score of participants before and after cannulation workshops were 4.3 and 7.1 out of 10 respectively (p

Purpose:
For a patient with coronavirus disease 2019 (COVID-19) with chest drain inserted for pneumothorax, air leakage from the chest drain system, which is not a closed system, can aerosolize when bubbling through the underwater seal and cause viral spreading. There are limited reports suggesting means to prevent viral spread with modification of the chest drain systems for patients with COVID-19. We would like to test the effectiveness of the modifications of the chest drain system to prevent COVID-19 spreading via aerosol in COVID-19 patients with chest drain system in our ICU.


Methodology & Sample collection:
We suggest three modifications to the chest drain system: use of 0.05% Aqueous-based chlorhexidine gluconate solution in water seal chamber, water column in the suction control chamber and use of bacterial viral filter at the suction port. The modified chest drain system was connected to a patient with COVID-19 who required mechanical ventilation. Several swabs from the modified chest drain system were taken and test for polymerase chain reaction (PCR) for COVID-19.


Findings:
Only sample from the chest drain tubing directly connected to patient side was tested positive with COVID-19. Samples from water seal chamber, chest drain tubing to connect for suction and area beyond the bacterial viral filter are all tested negative with COVID-19.


Limitations:
Continuous suction was not applied to the modified chest drain system. The effect of the modifications in the prevention of spread of COVID-19 if continuous suction is applied to the chest drain system needs further study.


Practical implications:
The modifications of chest drain system are easy to apply and the equipment are readily available in ICU. The modification could help the prevention of spread of COVID-19 in patients with pneumothorax using the chest drain system.