Study design, institutional review, ethical approvals and participants’ eligibility criteria
The study acronym: the EAGLE study, was shortened from “Efficacy of Air-GLovE in difficult venous access”. The study was carried out in two parts (Fig. 4). The first, an observational, proof-of-concept study for Airglove™ (EAGLE-1), recruited a total of 80 informed and consented participants from patients (inclusion-exclusion criteria; Table 4) attending the chemotherapy unit at Maidstone Hospital, Kent. Approvals for the EAGLE-1 study were granted by the local research ethics committee and the R&D Department at MTW Trust, Kent, England, UK.
The second part, EAGLE-2, was a prospective, cross-sectional, comparative study on 34 appropriately consented participants, where the degree of venodilation of the Airglove™ vs. WWI was determined by ultrasound in three veins of the forearm (Fig. 5). EAGLE-2 was approved by the institutional ethics committee (reference number: HLS/PSWAHP/18/168) and run at the clinical simulation unit of Glasgow Caledonian University, Scotland, UK. All participants to the study (EAGLE-1 and EAGLE-2) met the inclusion-exclusion criteria (Table 4) and written informed consent was obtained in accordance with the requirements of current good clinical practice (GCP) and the ethical standards of the responsible committee on human experimentation (institutional), and with the Helsinki Declaration of 1975, as revised in 2000 and 2008.
EAGLE-1 study
Evaluation of the Airglove™ on participants undergoing chemotherapy
The Airglove™ was evaluated on a total of 80 participants (demographics described in Table 1) undergoing chemotherapy at the Chemotherapy unit, Maidstone Hospital (MTW Trust), Kent, with DVA (inclusion-exclusion criteria in Table 4). The endpoint was set as the overall successful insertion of a venous-cannula (maximum two attempts) following the use of Airglove™. Total number of participants were recruited to the study over a period of one month and cannulation was carried out by a single qualified nurse to avoid operator bias. EAGLE-1 study questionnaires were completed by the operator of the Airglove™ and a total of 25 clinical staff in the chemotherapy unit normally performing venous-cannulation. Aspects and features frequently contributing to cannulation success or failure from the questionnaires are tabulated in Table 2.
EAGLE-2 study
Pre-assessment of volunteers
Blood-pressure (from both arms), age, height, weight, BMI, percentage of body-fat (using bioimpedance-analysis (BIA) measured with a Tanita Body composition analyser (TBF-300M) across the lower limbs) and hydration status (assessed clinically by capillary refill, dryness of mucus membranes, jugular-venous-pressure) were recorded prior to venodilation and ultrasound measurements.
Interventions
Warmed-water Immersion (WWI) method of venodilation
The upper limbs (to the level of mid-humerus) of participants were immersed in a container filled with a mix of cold-and warm tap-water to 38.5oC for 3 minutes in the WWI method. Water temperature was verified using a Tanma 72-2060 digital thermometer. Arms were towel-dried, and the degree of venodilation was assessed by ultrasound.
Airglove™ method of venodilation
A single-use, double-walled, polyurethane-balloon (Green Cross Medico, Scotland) was inflated around the forearm using the Airglove™ to 38.5oC at setting number-3 (Fig. 1c) for 3 minutes and then removed. The degree of venodilation was assessed by ultrasound.
Ultrasound assessment of venous dilation by Airglove™ vs. WWI
No invasive procedure, such as cannulation or venepuncture, was carried out on participants in EAGLE-2. Ultrasound assessment of veins was performed prior to and following the Airglove™ and WWI warming methods. Volunteers were placed in a semi-recumbent position while ultrasound measurements were taken from each arm, devoid of external circumferential compression-pressures such as tight clothing or watches.
Three markings, aligning with common IV cannulation sites, were made on the upper limbs: i) the prominent veins proximal to the base of the middle-digit on the dorsum of the hand (dorsal-metacarpal), ii) lateral side of arm, 5 cm proximal to the skin crease of cubital fossa, to mark cephalic vein in arm (CV at cubital fossa), and iii) medial side of arm at the skin crease of cubital fossa to mark the median-cubital vein (MCV near cubital fossa).
Ultrasound examination of the veins in the forearm was performed by a single sonographer, using a GE Logic S8, (software version R2, revision 1.1, GE Healthcare, USA) with multi-frequency linear array transducer (L6-15MHz). Measurements of each vein were recorded to 0.01 cm accuracy, using two-dimensional B-mode scanning in a transverse plane. On frozen annotated images, antero-posterior (AP) measurements were recorded at the skin-marked vessel sites, from inner anterior wall to inner posterior wall (or intima to intima) to facilitate repeatability (Fig. 2).
Baseline ultrasound measurements were recorded at 23°C ambient room temperature. The WWI method was deployed as described above and ultrasound measurements conducted on all three marked sites. Participants were rested for 30 mins before testing with Airglove™ and ultrasound measurements were recorded on the same three marked sites. This process was repeated on the opposite limb. WWI and Airglove™ venodilation (temperature setting #3, Fig. 1c) were set at 38.5oC for uniformity, enabling comparison. The first warming method used first on each participant was randomised and carried out by one of our team members.
Blinding in EAGLE-2
Blinding was achieved by physical concealment of onscreen diameter measures at time of scan, also ensuring the sonographer was unaware of the venodilation warming method used for participants. Vein diameter measurements were retrospectively recorded for statistical analysis.
Statistical analysis and sample size calculations
Personal communication with colleagues from chemotherapy units in the UK indicated a 65% cannulation-success rate for patients with DVA using the WWI method (LGD personal communication). Hence, the alternative hypothesis was formed that cannulation-success rate with Airglove™ in EAGLE-1 may approximate or exceed the success rate set for the WWI method, and the null-hypothesis was set as cannulation-success would be successful only half of the time (50%). Employing the pwr.p.test module (in R ver.3.60)34, a minimum sample-size of 77 participants was needed for a cannulation success rate of 65%, powered at 85% to correctly accept the alternative hypothesis at the p = 0.05 significance level.
The EAGLE-2 study which enrolled 34 participants, exceeded the minimum sample size of 24 required for 90% power, with α = 0.05 and the ideal anticipated Cohen’s “d” effect size of 0.833, calculated using the “pwr.T.test” module (in R ver.3.60, within RStudio ver.1.2.1335).1011 Mean venodilation by WWI and Airglove™ for input into the “pwr.T.test” script was determined from previous pilot studies carried out at the Sir Geraint Evans Wales Heart Research Institute, Cardiff. Normality of data was determined using the Shapiro-Wilks test in SPSS (ver.25) prior to employing the independent means T-test. The paired samples T-test was used to compare the degree of venodilation between Airglove™ vs WWI.
All statistical analysis was carried out with SPSS (ver.25; IBM Corp), SciPy module (ver.1.3) for Python (ver.3.7.2) and R (ver.3.60), with P < 0.05 considered as statistically significant.
Patient and public involvement in this study
The study protocol and questionnaire forms were circulated to members of the Kent Lung Awareness Charity (UK Charity no. 1166445; http://kentlungawaress.weebly.com ), a charity group consisting of families, friends and supporters of those who had succumbed to various lung diseases including primary and secondary lung cancers. Opinion was also gathered from various lay members of the EAGLE steering committee formed initially at Cardiff University.