Karen E. A. Burns MD, FRCPC, MSc (Epid) Clinician Scientist, - - PowerPoint PPT Presentation

Karen E. A. Burns MD, FRCPC, MSc (Epid) Clinician Scientist, Associate Professor Unity Health Toronto (St. Michael’s Hospital) Li Ka Shing Knowledge Institute Toronto, Canada Karen.Burns@unityhealth.to

Within the last 12 months I have not had any type of financial arrangement or affiliation with commercial interests related to the content of this continuing education activity that requires disclosure.

DISCLOSURE

I have nothing to disclose

• Approximately 40% of the total time spent on invasive MV is

dedicated to weaning [1].

• Invasive MV is associated with important complications,

including VAP, sinusitis, gastrointestinal bleeding, and muscle weakness.

• Strategies to limit the duration of invasive MV and ventilator-

related complications have been identified as key research priorities in Critical Care [2].

• 1. Esteban A, Chest 1994;106:1188-93.
• 2. MacIntyre NR, Chest 2001;6 Suppl:375-95.

• Over the past two decades, investigators conducted RCTs to

evaluate aspects of weaning that reduce the time patients spend on MV including use of:

• Screening protocols
• SBT techniques
• Strategies to reduce ventilator support for patients who fail an

initial SBT

• Early RCTs of protocol-directed screening (largely led by RTs

and RNs) were mostly positive

• Ely 1996
• Kollef 1997
• Marelich 2000
• No benefit - selected populations
• Namen 2001 (neurosurgical patients)
• Randolph 2002 (pediatric patients)
• No benefit - highly structured settings
• Krishnan 2004 (Johns Hopkins)

 Blackwood et al, summarized 17 trials (n = 2,434) comparing

protocolized (largely led RTs and RNs) vs. non-protocolized weaning in a SR/MA.

 Most trials compared ‘once daily’ screening to ‘usual care’.  **Usual care required a physician order to conduct SBTs.

Blackwood B, Cochrane D, 2014; Issue 11: CD006904.

Compared to non-protocol based weaning, protocolized weaning:

 26% reduction in total duration of MV

[n = 14 trials, 95% CI (13% - 37%), p = 0.0002]

 70% reduction in weaning time

[n = 8 trials, 95% (CI 27% - 88%), p = 0.009]

 11% reduction in ICU stay

[n = 9 trials, 95% CI (3% - 19%), p = 0.01

 Only 1 trial (n=385) compared ‘twice daily screening’ to ‘usual

care’ and found a significantly shorter duration of MV and a trend toward a lower VAP in ‘twice daily screening’.

 No trial compared a strategy of ‘more frequent screening’

to ‘once daily’ screening.

Marelich, CHEST, 2000;118:459-67.

• A Cochrane review (n=9 trials) compared PS and T-piece

weaning strategies in critically ill adults.

• Nonsignificant differences between PS and T-piece weaning
• weaning success
• pneumonia
• reintubation
• In a subgroup analysis, patients were significantly more likely

to pass a PS (vs. a T-piece) SBT [RR 1.09, 95% CI 1.02 to 1.17] in 4 trials (n= 940).

Ladiera et al, Cochrane Database 2014 May 27;(5):CD006056.

 Published a SR/MA (n=12 trials, n=2161) compared PS and T-

piece weaning in critically ill adults (including tracheostomized patients).

• PS vs. T-piece weaning did not influence:
• weaning success
• reintubation
• ICU mortality
• A subgroup analysis suggested that PS weaning may be

superior to T-piece weaning with regard to weaning success

• for simple weaning patients [RR 1.44 (1.12 – 1.86)]
• but not for difficult [RR 1.45 (0.73 – 2.88)] or prolonged [RR 0.85 (0.69 –

1.05)] weaning patients.

Pelligrini et al, Respir Care. 2016;61(12):1693-1703.

• Study Design: Randomized or quasi-randomized trials
• Patients: Critically ill adults or children
• Interventions: Directly comparing 2 or more SBT techniques

Excluded trials - evaluated SBTs as part of a weaning strategy

• Primary Outcomes: SBT success, extubation success, reintubation

Burns KE, Crit Care 2017 Jun 1;21(1):127.

3,785 unique citations 3,602 citations excluded 183 potentially relevant citations 152 citations excluded 65 crossover studies 31 not randomized 27 weaning studies 13 SBT vs. no SBT 8 physiologic 8 other 31 included trials (n=3,541) *11 (T-piece vs. PS) 9 (T-piece vs. CPAP)

9 trials n=1,901 p=0.96 RR 1.00 [0.89, 1.11] I2 = 77% Post hoc 8 trials n=1,381 p=0.03 RR 1.06 [1.01,1.12] I2 = 0%

Quality assessment No of patients Effect Quality

No of tria ials [n]

Risk of bias Inconsisten cy Indirectn ess Imprecisi

• n

Pressure Support T-piece Relative (95% CI) Risk Difference

Operative trials (<24 hours): Low Pre-test Probability

2 tria ials [548] 48]

no serious risk of bias serious1 not serious serious2 173/274 (63.1%) 226/274 (82.5%) RR 0.86 (0.61 to 1.22) 115 fewer per 1000 (-322, + 181) ÅÅOO LOW

Non-operative trials: Greater than Low Pre-test Probability

7 7 tria ials [1353] 353]

no serious risk of bias not serious not serious not serious 536/680 (78.8%) 499/673 (74.1%) RR 1.07 (1.01 to 1.13) 52 more per 1000 (+7, +96) ÅÅÅÅ HIGH

p=0.3

11 trials n=1,904 p=0.007 RR 1.06 [1.02, 1.10] I2 = 0% Post hoc 10 trials n=1,384 p=0.03 RR 1.06 [1.01, 1.12] I2 = 0% Burns KE, Crit Care 2017 Jun 1;21(1):127.

• 1. PS SBTs may facilitate extubation decision-making.
• 2. Even if PS SBTs underestimate post-extubation WOB passing an

SBT may:

• offset clinician reluctance to extubate some patients
• result in more timely and successful extubation

T-piece SBTs may be appropriate in selected patients

• Especially for patients that we think have a low likelihood of

extubation success (e.g., LV dysfunction, neuromuscular weakness)

• When we wish to prioritize a low FP rate for passing an SBT (to avoid

the risks associated with extubation failure).

• However, when we use T-piece SBTs (vs. PS SBTs) in patients with a

high likelihood of extubation success they may induce a higher FN rate.

Pre-test probability

Setting ting: : 18 ICUs Spain; Inter terven entio tions ns: : 30 min PS 8 cm H2O SBT (n=557) vs

• vs. 2 hr T-piece SBT (n = 578)

SBT T succ cces ess s (PS vs. T-piece): iece): 532/575 (92.5%) vs.486/578 (84.1%); p<0.001 Extubation tubation success cess (PS vs. T-piece): iece): 473/575 (82.3%) vs. 428/578 (74.1%); p<0.001 Reintu intubat ation ion rates es: : 11.1% vs. 11.2%

Subira et al, JAMA, 2019

Bedeneau, AJRCCM, 2016; 195:772-83. Current RCTs include pts < 24 hrs and ‘pts with a high pretest probability of success’

To understand how ICU physicians discontinue MV in practice We conducted a cross-sectional survey of adult ICU physicians practicing in 6 geographic regions:

• Canada

India

• United Kingdom (UK) Europe (excluding the UK)
• Australia/New Zealand United States (USA)

In a self-administered, multinational survey, we sought to

• 1. Characterize magnitude and extent of weaning practice variation
• Identify weaning candidates
• Conduct SBTs
• Use of ventilator modes
• Use of written directives to guide aspects of care
• Use of NIV in the weaning & peri-extubation period
• Personnel involved in weaning
• 2. Assess for regional differences in weaning practices
• 3. Identify predictors of practice variation

• We used rigorous survey methodology to design, test, and

administer our survey.

• We collaborated with regional critical care societies to administer
• ur questionnaire
• Canada (CCCS, CCCTG, QICS)
• India (ISICM)
• UK (UKICS)
• Europe (ESICM)
• Aus/NZ (ANZICS)
• USA (SCCM)
• We analyzed 1,144 questionnaires (Canada, 156; India, 136; UK,

219; Europe, 260; Australia/New Zealand, 196; USA, 177).

Goal: Achieve 200 responses from each region Proportionate sampling

Across regions, most respondents screened patients once daily to identify SBT candidates (regional range, 70.0%–95.6%) Less often screened twice daily (range, 12.2%–33.1%) or more than twice daily (range, 1.6%–18.2%)

Most respondents used PS alone (range, 31.0%–71.7%) or PS with SBTs (range, 35.7%–68.1%) to wean patients. Intensivists infrequently or rarely used other modes (volume-assist control alone, SIMV, pressure-assist control, and PLVG) to wean. When used, these modes were employed in combination with SBTs.

Two modes were predominantly used to conduct SBTs:

• PS with PEEP (56.5%–72.3%)
• T-piece (without CPAP; off the ventilator; 8.9%–59.5%)

They infrequently or rarely used

• CPAP (without PS; 6.5%–18.9%)
• ATC (2.6%–21.1%)
• T-piece with CPAP = 0 on the ventilator (1.2%–11.8%)
• PS without PEEP (1.6%–7.7%)

• Use of written guidelines, protocols, or policies varied significantly across regions

with most respondents affirming having no directive

• Directives for sedation administration were the most commonly used directives

across regions (range, 37.5%–75.4%)

• Broader variation in the use of written directives to guide
• Conduct of SBTs (range, 10.4%–73.5%)
• Adjustment of ventilator support (range, 15.5%–55.6%)
• Directives to guide delirium management were infrequent (range, 21.8% -37.0%).

Across indications - NIV was most commonly used for patients with COPD Significant variation in the use of NIV for weaning and peri-extubation for patients with COPD, CPE, OSA, and in postoperative patients

• Personnel availability and work profile varied across regions
• 1. Screening to identify SBT
• Involved attending intensivists, senior trainees
• RTs in North America (87.2% and 78.5%) and RNs in the UK

(57.5%) and Australia/New Zealand (44.4%) assuming lead roles in screening

• 2. Decisions to conduct SBTs were made by intensivists and senior

trainees across regions and in collaboration with RTs in Canada (64.7%) and the USA (58.2%).

• 3. SBTs were actually conducted by different personnel
• RTs (Canada, USA)
• Attending intensivists and senior trainees (India)
• Attending intensivists, senior trainees, and nurses (UK, Europe,

and Australia/New Zealand)

• 4. Ventilator Adjustments and Extubation
• RTs in North America and
• RNs in the UK Australia/New Zealand and Europe
• 5. Across regions, attending intensivists made decisions regarding

extubation and tracheostomy

• “Having a protocol to adjust support” and “RT availability”

were determinants of conducting screening in practice

• “Region” was a significant determinant of all common

weaning practices.

Across regions, most intensivists reported

 Conducted daily screening to identify SBT candidates and

less often conducted more frequent screening.

 Used 2 techniques (PS with PEEP and T-piece – off

ventilator) to conduct SBTs

 Used PS alone and PS with SBTs to support patients during

weaning from MV

We identified significant regional variation:

• Use of written directives to guide care during weaning

(with most respondents having no directives)

• Use of NIV in the weaning and peri-extubation period
• The availability of personnel involved in weaning and their

scope of practice

• Region was a significant predictor of all common

weaning practices.

 Our findings highlight that stated weaning practices are variably

supported by current evidence and that practices have evolved where evidence is equivocal or lacking

 High quality research is needed to inform MV discontinuation

practices

Supporte ted Equivoca

• cal

Lacki king ng Daily Screening Conduct of SBTs Extubation to NIV Use of PS (generalized from RCTs failed SBT) Twice daily screening

• Weaning is a collaborative process that is heavily influenced by

contextual factors.

• In practice, evidence is applied in ICUs that differ from the

settings in which it was generated.

• Practice variation may explain, in part, why results of RCTs are

rarely replicated in practice.

• Although some practice variation may be warranted to address

patient variation, context-specific factors that produce unwarranted practice variation or negatively impact outcomes should be identified and addressed.