Noninvasive monitoring and monitoring of maximalinspiratory and expiratory flows(MIF and MEFrespectively)by electrical impedance tomography(EIT)could allowan early detection of changes torespiration’s mechanical parametersduetochanges in conditions orresponse totreatment.We aimed to validateEIT-basedmeasuresofMIFas well asMEF against spirometrymeasurements in intubatedhypoxemic patients during controlled ventilationand spontaneous breathing.In addition, the distribution in the region ofmaximum airflows may interact withlungdisease and increasethelikelihood of additional ventilationinjury.We also wantedtoanalyze the effectsofventilation settings that affectlocalMIFandMEF.


We performed a new analysisofdataof two randomised, prospectivecrossoverstudies.The study included intubated patientsadmitted to theintensive care unit suffering fromacute hypoxemic respiratory failure(AHRF)or acute respiratory distress(ARDS)under pressure supportbreathing(PSV, n10) andVolume-controlled ventilation(VCV, n=20).We assessed MIF and MIFwith spirometry and EIT throughoutseveral combinations of settings for ventilation which were:highervs. lower supportduringPSV and greatersupport vs. lowerpositive end-expiratory pressure(PEEP)duringbothVCV and PSV.Regional airflows were assessed byEITin both non-dependent and dependentlung regionsas well.


MIF and impedanztomographie determinedthroughEIT werestrongly correlative withthose measured by spirometry duringall conditions(rangefromR2 0.629-0.776 and R2 0.606-0.772,respectively, p<0.05onall) in accordance with the clinically acceptableranges of agree.The higher PEEP levels significantly improvedhomogeneity in the regionalpatternof MIF and MEFwhen ventilated with volume control,by increasing airflows in theaffected lung regions while reducingthem in non-dependent areas.


EITallows for accurate and noninvasive monitoringforMIFas well asMEF.The current study also positsthepossibilitythat EITcan help to determinePSV and PEEPset-upsto increase the homogeneity and consistency ofthe regional airflows that are extending and deflating.


The electrical impedance imaging(EIT)isanoninvasive bedside, radiation-freedynamic lung imaging technique. EITproduces maps of the intrathoraciclung impedance variations that are compared tothe baseline(i.e.the end-expiratory lung volume fromthe previousbreath) every20 to 50 milliseconds].Changes in intrathoracic impedance as measuredthroughEIT are linearlyrelated toglobal and regional tidal volume andheld at increasing positive end-expiratorypressure (PEEP) levels [2].This means thatEITgives a noninvasive bedside continuousmeasure of regionallung volumechanges duringexhalation and inspiration.

Inspiratory and expiratory airflows correspondto thespeed at whichlung volume changeintime.In intubated patients,they areusually measured witha spirometer appliedto the ventilator circuit either beforetheendotracheal tube or withinthe ventilator.Global maximal inspiratory andexpiratoryflow(MIF and MEF, respectively)determined bystandard spirometry rely onphysical properties in the respiratory system(namely lung compliance, lung volume andairway resistance) [33.Thus, monitoringMIF andMEF canbe beneficial to help guideventilatory settings(e.g. selectingthelevel of positive pressure associatedwithbettermechanics)and/or to assesstheeffectiveness of treatments with pharmacological agents(e.g., increasedMIFand/or MEF followingbronchodilator drugs) [4].Spirometry can only provideglobal measurements of MIF andMEF, whereas heterogeneous distributionofaltered lung mechanics is acharacteristic of acute hypoxemicfailing(AHRF)as well as acute respiratory distress syndrome(ARDS) [55.The damage to the alveoli causescollapse of lung units tightlylocated between normal-, part-or over-inflated ones which can result inan imbalanceof regionalMIFandMEF values.These imbalances could increasetherisk of ventilator-induced lung injury(VILI)by a variety of mechanisms[6], while settingthat create more homogenous regional flowcould reduce the risk. Externalspirometry can leadtoaltered patterns in the respiratory system andincorrect measurements,as well[77.Thus, a noninvasivebedsidedynamic technique to measureglobal and regional MIF andMEFlevels wouldbe a significant addition tothe understanding ofAHRF and ARDSpatients’ pathophysiology as well asto aid in the development of personalized treatment.

In the current study,after preliminary results obtained in anthe animal model[8], our goal wastoverify inpatients who are intubatedAHRFandARDS patientsreceivingcontrolledbreathing andEIT-based breathing tests that are based on spontaneous breathing to measureMIF and MIF global againststandardspirometry.Furthermore, we exploredtheeffects of higher. lowerPEEP andpressure supports on is our beliefthat higherPEEPand lower pressure support mayresult in a more uniform distribution oflocalMIFas well asMEF.

Materials and methods


We performed a new analysis of data collected during two prospective randomized crossover studies: in the first (pressure support ventilation (PSV) study) [9], ten intubated patients recovering from ARDS [10], lightly sedated (RASS – 2/0), undergoing PSV and admitted to the intensive care unit (ICU) of the university-affiliated San Gerardo Hospital, Monza, Italy, were enrolled; and in the second (volume-controlled ventilation (VCV) study) [11], twenty intubated, deeply sedated and paralyzed patients with AHRF (i.e., PaO2/FiO2 <=300, PEEP >=5 cmH2O, acute onset, no cardiac failure) or ARDS admitted to the same ICU were enrolled. Theethics committee ofSan Gerardo Hospital, Monza, Italy, approved thestudieswhile informed consent was soughtfollowinglocalregulations.Additional information ontheinclusion and exclusion criteriaforbothstudies are providedinan online data supplement(Additionalfile1.).

Demographic data collection

Wecollected data on sex, ages, Simplified Acute Physiology Score IIscores, etiology, diagnosis andthe severityof ARDS, days onmechanical ventilationbefore study enrollmentforeachpatient.Mortality in the hospital was reported,too.

EIT andventilation monitoring

Ineach patient, EIT-dedicatedbelt,containing 16 equallyspaced electrodes, was positionedon the thorax inthesixth or fifthintercostalspaces and connected toan industrialEIT monitor (PulmoVista 500, Drager Medical GmbH, Lubeck, Germany).Throughout all study phases,EITinformation was generated throughapplying smallelectrical currents rotating aroundpatient’s thorax, continuously recordedat 20 Hz. The data were then savedfor offline analysis, just asexplained [12in the previous article [12, 13].In sync withEITtracer the airway pressure as well asairflows ofthe mechanical ventilator wererecorded continuously.


More information onthe two protocolsare availablein thedata supplement online(Additionaldata supplement file1).

In short, inThePSV study,participants underwentthe followingsteps of crossover that lasted for 20 minutes each:

  1. 1.Low support at clinical PEEP(PSV low)in comparison to.higher support forPEEP at the clinic(PSV high);
  2. 2.Clinical supportforlow PEEP(PSV-PEEP low)compares to.medical support for higher PEEP(PSV-PEEP high).

As part of theVCV study,instead,the following phaseswere carried outin a randomized order of crossovers,each lastingfor 20 minutes:

  1. 1.Secure VCV during low PEEP(VCV-PEEP low)in comparison to.VCV with a protective function atPEEP+ 5cmH 2O (VCV-PEEP high).

EIT anddata on ventilation

Based on offline analysis ofEITtracings obtained duringthefinal minutesduring each stage(analysis of10breaths) We determinedboth the global and regional(same-sizeregion of lung that is dependent as well as non-dependent) noninvasive airflows’ waveform,as previously described[88.In brief, instantaneous globalregionalexpiratory and inspiratoryairflowswere assessed asvariationsof the global and regionalimpedance measured every 50 ms and multiplied by thevolume/tidal impedance ratio fromthesame study period anddivided by 50ms. EIT airflow data werethen converted toL/min (Fig. 1) and the maximumMIFs from EITs that were global and regionaland MEF (MIFglob MIFglob, MIFnondepand MIFdepMEFglob with MEFnon-dep, MEFglob andMEFdep for MEFglob, MEFnon-dep and MEFdep) wereidentified , and theresults were averaged over5-10respiratorycycles.