Objectives:
Primary objective To determine if a protocolized rotational use of interfaces during NPPV
compared to standard care is effective in reducing the development of new pressure sores
in patients with hypercapnic ARF treated continuously i.e., for more than 24 hours, with
NPPV (to avoid intubation, as alternative to invasive ventilation, and during early
extubation and weaning).
Measurements of effectiveness include:
The primary effectiveness outcome will be the proportion of patients who develop new
pressure sores at 36 hours from randomization. Secondary outcomes will be the onset of
pressure sores measured at different time points (12, 24, 36, 48, 60, 72, 84, 96 hours),
number and stage of pressure sores and comfort measured at the same time points of the
primary outcome. The location of pressure sores per patient will be also collected. The
study will also monitor the number of patients that will interrupt NPPV treatment for
discomfort, the presence of eyes irritation, the adherence to the protocol (duration of
NPPV treatment/day, interface change, hygiene protocol application) and the presence of
vasoactive medications. Variables like nutritional status at randomization, Braden Scale
and Frailty Index at hospital entrance, total duration of NPPV from randomization,
duration of NPPV per day, hospital length of stay and in-hospital mortality will be
recorded. The economic impact of the protocolized rotational use of interfaces will be
assessed through a cost-effectiveness analysis evaluating the additional costs needed to
avoid a pressure sore due to the planned intervention as well as the costs needed to
actively treat skin ulcers due to failure of preventing measures in both groups of
patients.
Randomization:
A 1:1 stratified randomization will be performed to assigned patients to the protocolized
rotational use of interfaces (experimental treatment) or standard of care (comparison
group). Subjects will be stratified according to center and ward (intensive care unit,
intermediate respiratory care unit, respiratory medicine service or internal medicine
service according to hospital organization). Within each stratum, subjects will be
randomly assigned to one of the treatments using a predefined randomization list. To
ensure masking, randomization schemes will be input directly into the Research Electronic
Data Capture (REDCap) system. In the treatment group, the first interface will be
randomly assigned. Randomization scheme will be input directly into REDCap system.
Study treatments:
In both arms, the treatment according to the assigned protocol will be continued until
patients require NPPV.
Skin underneath the mask and occipital region, will be inspected every 12 hours to check
for any possible damage by assessors blind to the study protocol. Assessors (physicians
or nurses) will be instructed to recognize pressure sores and their stage. Before
starting the study recruitment, the assessors will take a test aiming to correctly stage
10 pressure sores. Only assessors who will recognize all of them, participate to the
study. Test will be taken trough google forms and results will be registered for the
study. To minimize detection bias, assessors will not participate to the care of the
patient and will be blind on patient group of assignment. If a pressure sore will be
detected, physician in charge of the patient will be promptly informed and a note will be
written on the nursing record.
Pressure sores will be stage as follows:
stage I: A reddened, painful area on the skin that does not turn white when pressed. This
is a sign that a pressure ulcer may be forming. The skin may be warm or cool, firm or
soft.
stage II: The skin blisters or forms an open sore. The area around the sore may be red
and irritated.
stage III: The skin now develops an open, sunken hole called a crater. The tissue below
the skin is damaged. You may be able to see body fat in the crater.
stage IV: The pressure ulcer has become so deep that there is damage to the muscle and
bone, and sometimes to tendons and joints.
In addition to the 4 main stages for pressure sores, there are 2 others: 1) "Unstageable"
is when you can't see the bottom of the sore, so you don't know how deep it is; 2)
"Suspected Deep Tissue Injury" when the surface of the skin looks like a Stage 1 or 2
sore, but underneath the surface it's a Stage III or IV.
Discomfort due to the mask will be evaluated every 12 hours using a 5-level Likert-like
scale to answer the questions "How is your level of discomfort related to the mask?" (in
Italian). The extremes of the scale are "1 - no discomfort" and "5 - unbearable
discomfort," with three intermediate levels ("light discomfort," "moderate discomfort,"
and "intense discomfort").
NPPV standard management:
In both arms, standard care (diagnostic tests, administration of systemic therapy
including antibiotics, steroids and inhaled bronchodilator therapy and hemodynamic
management) will be applied according to the clinical practice of each institution.
The right size of the mask will be selected according to the indications provided by the
manufacturer. The mask straps will be adjusted according to the " two-finger rule " not
to tighten to hard the headgear.
Following international guidelines, patients who benefit from NPPV during the first four
hours of treatment will receive NPPV for as long as possible during the first 24 hours.
Target saturation range will be 88-92% either spontaneously breathing or when receiving
NPPV. NPPV treatment will last until the acute cause has resolved, commonly about two to
three days.
NPPV will be started as suggested by the British Thoracic Society/Intensive Care Society
Guideline for the ventilatory management of AHRF in adults. NIV will be started when pH
<7.35 and PaCO2 >6.5 kilopascal (kPa) persist or develop despite optimal medical therapy,
targeting an oxygen saturation of 88-92%. Blood gas exchange will be performed before
starting NPPV treatment and after 2 hours from its starting and thereafter, when
clinically indicated. NPPV will be initially set in a pressure support mode with a
ventilator designed specifically to deliver NPPV or with an ICU ventilator provided with
noninvasive ventilation mode. Expiratory positive airway pressure (EPAP) will be set at
3-5 Centimeters of Water (cmH2O) while inspiratory positive airway pressure (IPAP) will
be set to achieve adequate augmentation of chest/abdomen movement with an expiratory
tidal volume of 6-8ml/kg and a reduction of respiratory rate (< 30 breath/min) together
with a good patient ventilator interaction (i.e., pressure-flow curve monitoring). The
presence of active humidification will be left to the choice of the doctor in charge for
that patient, although heated humidification should be considered if the patient
complains of mucosal dryness or if respiratory secretions are thick and tenacious.
The weaning strategy will be documented in the medical and nursing records. The following
might be use as general indication: continue NPPV for 16 hours on day 2; 12 hours on day
3 including 6-8 hours overnight use; NPPV may be discontinued on day 4 unless
continuation is clinically indicated.
The suggested monitoring during NPPV will be continuous peripheral oxygen saturation,
intermittent measurement of PaCO2 and pH, ECG monitoring in the patient with a pulse rate
>120 bpm or if there is dysrhythmia or possible cardiomyopathy.
As guidelines recommend keeping the skin dry and clean under medical equipment, a
protocol for skin will be adopted by all centers included in the study.
If a patient needs prolonged continuous NPPV, brakes will be programmed and annotated in
the case report form, in case of instability, interface will be removed for about 10
minutes only to ensure oxygenation.
When signs of skin trauma become apparent, a barrier dressing and a strategy of regular
breaks and alternating between two interface types will be used. In case of mask-related
rash even in the absence of allergy, topical steroids may be indicated and/or antibiotics
if the wound becomes infected. In case NPPV causes severe gastric distension, a
nasogastric tube will be inserted.
Early extubation and weaning with NPPV:
According to the guidelines, NPPV is recommended to aid weaning from invasive mechanical
ventilation in patients with AHRF, secondary to COPD and to other causes of AHRF, when
local expertise in its use exists. Patients during invasive mechanical ventilation will
be treated according to the clinical guidelines. After setting maximum medical treatment,
once with ventilation in a controlled mode pH and PaCO2 will be normalized in presence of
neurological, hemodynamic and respiratory stability, patients will be extubated and put
in NPPV. The ventilator will be set with the same positive end expiratory pressure (PEEP)
and inspiratory pressure support (PS) level applied during invasive mechanical
ventilation, setting the fastest pressure rise time at least during the first 24 hours
after extubation. Then, NPPV will be gradually withdrawn if patients tolerated
spontaneous breathing until they could permanently sustain spontaneous breathing.
Data collection:
A RedCap database will be set up to collect the information needed for the study. The
database will be made of 10 sheets: the first one collecting the baseline demographic and
clinical characteristics of the patients and the others will be made to collect the
outcomes' information at each time point and a discharge sheet including information on
the overall treatment.
The randomization procedure will be implemented within Redcap.
Sample size:
Given a type I error of 0.05, a power of 0.8, an expected proportion of patients who
develop pressure sores in the control group of 23% and an expected dropout rate of 5%,
239 subjects per group are estimated to identify a 10% absolute difference in the
proportion of patients developing pressure sores between intervention and control groups
measured at 36 h from randomization.
Statistical analysis:
The analyses will be performed according to the intention to treat principle. Descriptive
statistics will be calculated to summarized patient characteristics collected at study
entry according to randomization group. Categorical variables will be reported as
absolute frequencies and percentages and continuous variables as mean and standard
deviation or median first and third quartiles if not normally distributed according to
the Shapiro-Wilks test.
The raw risk of pressure sore will be calculated ad each time point as the ratio between
the number of subjects who developed a pressure sore at the time point divided by the
number of subjects undergoing NPPV at the time point. Patients having a planned break at
the time-point considered, but not been weaned off the NPPV treatment yet, will be
considered undergoing NPPV.
Mixed effect logistic regression models will be used to estimate the risk of pressure
sores and the difference between interfaces including a random intercept to account for
clustering of the subjects within center and repeated measures within patients. The
response variable will be the presence/absence of pressure sores and the independent
variables the treatment, time (12, 24, 36, 48, 60, 72, 84, 96) and their interaction. The
marginal and center specific predicted probabilities estimated by the model will provide
the incidence of pressure sores at different times and overall. The inclusion of time as
covariate will allow to test the presence of a time trend in incidence of pressure sores
and to investigate the variation of time trends according to treatment group.
Regarding the number of pressure sores, different mixed effect Poisson regression models
will be applied: one for each time point to estimate the average number of pressure sores
at each time point and a model in which only the highest number of pressure sores will be
considered for each patient. Time will be included as offset to evaluate the average
number of pressure sores per day.
The grade and location of pressure sores will be summarize using descriptive statistics
at each time point.
Regarding the economic evaluation. the incremental cost effectiveness ratio (ICER) will
be calculated considering two outcomes: i) the proportion of pressure sores and ii) the
comfort. The ICER as the ratio between the difference each outcome in the two groups
(difference in the proportion of pressure sores or comfort score) and the difference in
the average cost for each treatment allowing to evaluate the incremental costs needed to
reduce of 1 point the outcomes thanks to the rotational use of interfaces. The confidence
intervals of the ICER will be calculated using the non-parametric bootstrap method,
specifically the 2.5th and 9.75th percentiles of the bootstrap distribution of ICER will
be considered as the lower and higher limits of the confidence interval. The
cost-effectiveness acceptability curve will be also drawn to evaluate the probability of
the intervention to be cost-effective considering different thresholds.
Safety monitoring:
A single interim analysis will be performed when a number of patients equal to two-thirds
of the predefined sample are randomized and complete the study. Early stopping criteria
will be a predetermined 2-sided p value < 0.01 for the rejection of the null hypothesis
that the strategies are equivalent in terms of rate of pressure sores.
