International Symposium on Intensive Care and Emergency Medicine (ISICEM)
The most famous ICU conference (other than SMACC) has been on this week.
Li Huey Tan, an intensivist currently working at The Alfred hospital in Melbourne was there feeding us live reports as a mini-blog to give those of us who couldn’t make it a bit of a taster…
Click “Read more” to hear about day 2.
March 20 at ISICEM
Wednesday kicked off with author of ‘Mechanical Ventilation’ Neil Macintyre presenting three topics. His discussed some key principles in “Patient-Ventilator Interactions” and “Pressure vs Volume targeting: Is one better?”. Like most experts in mechanical ventilation, it’s always important to look at the ventilator waveforms to troubleshoot dyssynchrony. Rapid flow delivery can cause premature termination of breaths in a spontaneous ventilating patient. Newer modes of ventilation such as Proportional Assist Ventilation or Neurally Adjusted Ventilation Assistance may theoretically offer a smoother patient-ventilator interaction. However, there currently lacks meaningful outcome data in favours of these new modes.
Pressure target or volume target ventilation? Despite what some neurotic intensivists may believe about various modes of ventilation, both volume targeted and pressure targeted modes are equally safe and effective. MacIntyre’s sound advice is to use the mode you understand best. From his published papers, a square pressure time wave form is most comfortable for the patient. We all know the issue with PCV is the potential to over or under deliver tidal volumes with the patient’s change in chest and lung compliance with time. The solution is to apply feedback mode on the ventilator setting to prevent this from happening. Apparently this safety feature is available in most new ventilators. Another tip he gave was in relation to pressure targeted modes. The inspiratory slope or rise time can be adjusted to troubleshoot dyssynchrony. The other method is to set a fix Ti (Inspiratory Time) if a patient is taking long inspiratory breaths.
“APRV as a lung Protective Mode” was MacIntyre’s last talk for the morning session. Suffice to say, whilst APRV has the theoretical benefit of improving VQ Mismatch by allowing spontaneous breathing, prevents overdistension, this mode of ventilation has not been showed to improve outcome. The studies which demonstrated a reduction in sedation requirements had patients in controlled mechanical ventilation in the control group. Importantly, when compared with other modes of spontaneous or assist ventilation, the sedation requirements were the same. The discussion went on to address some concerns other panel members had with the development of intrinsic PEEP as well as the rapid opening and closing of lung units when transitioning from T high and T low.
Interestingly, Laurent Brochard is an advocate of APRV but uses this mode due to the lack of inspiratory synchronization built into the function when ventilating ARDS patients. In “Is synchrony desirable during pressure controlled ventilation?”, he feels that spontaneous ventilation during the acute phase of respiratory failure is not beneficial to patient’s respiratory mechanics. This is supported by Papazian’s study of NMB in ARDS. He talks about the viability of the tidal volume decreasing with time in the setting of allowing patients to spont vent early on in their acute respiratory illness.
How to deal with ‘M’ waves on the ventilator? David Tuxen the common phenomenon of flow starvation in decelerating flows as well as “Reverse Triggering: The unrecognized dyssynchrony”. Reverse triggering is defined as ‘Patient is tracking the ventilator and the ventilator breath triggers patient inspiration’. To make the diagnosis, any adjustment in the ventilator rate in either direction, will result in a change in patient’s respiratory rate in the same direction. The treatment options include increasing sedation (but the patient is usually already obtunded), reduce sedation to allow spontaneous ventilation, or reduce tidal volume. All methods are usually marginally effective. The cause of this phenomenon is unknown but may be related to lung stretch. Common in middle aged male patients.
Rising plasma lactate disproportionate to systemic haemodynamic stability? Rising LFTs and oliguria? Start thinking the sick right heart! And don’t give more fluid! The cardiac output of the right ventricle is extremely vulnerable to altered right coronary blood flow, heart rate and increase in RV afterload (aka PA pressures). As we all know, the right side is a low pressure system, designed to generate flow rather than pressure. Even during diastole, some flow still occurs into the pulmonary artery. If this did not occur, back pressure occurs into the right ventricle. Right ventricle pressure builds up, compromises coronary perfusion, RV dilates, tricuspid valve starts regurgitating blood back into RA and you know what happens next.. Our experts discussed how the right ventricle is impaired during mechanical ventilation (Antoine Vieillard-Baron), due to myocardial infarction (Alexandre Mebazaa) and in severe pulmonary embolism (Anthony Mclean).
The next time you set the ventilator, think of the right heart. RV systolic function can be impaired especially during inspiration. High PEEP may be good for lung derecruitment but bad for the RV. Piazza has a review in CHEST about Acutely Decompensated RV which is worth a read. Mechanical ventilation induces uncoupling between the RV and pulmonary circulation. Nitric oxide is useful in lowering PAP and hence improve RV function. Due to the recent published studies, unfortunately, more units are now getting rid of nitric oxide despite an argument for using it in a select group of patients. There’s been talk of a new indwelling TOE probe which can be left for 72 hours to allow clinicians to adjust ventilator settings and observe RV changes.
Mclean states that the RA Pressure is a useful marker for RV performance. A high troponin and BNP level is a marker for RV dysfunction. The 30 days mortality in patients with RV dysfunction due to PE is 40% and can be a useful prognostication aid. However, in the intermediate risk patients whereby RV dysfunction is present and patient is normotensive, mortality is in the region of 15%! Mclean advocates to consider thrombolysing these patients given that the risk of bleeding is not different in comparison to heparin. Also refer to Piazza JAMA 2013 on management of PE.
Attending the round table discussion on “Prognostication after cardiac arrest” was useful to hear the opinions of various intensivists from mainly Europe (Niklas Nielsen, Mauro Oddo, Jerry Nolan, Raimund Helbok, Fabio Taccone). It’s also good to know the Australian practice does not defer significantly from these guys. The one difference though is the availability of biomarkers. Although biomarkers such as S100B and NSE should not be relied solely to prognosticate our patients, they seem to be more readily available in Europe. The cut of of 33 micromoles with NSE may not be entirely accurate in the setting of hypothermia. There is also a variation in NSE assays as high as 30% in different labs. S100B may be a more reliable biomarker in this settingas it has a short half life and its presence reflect ongoing neuronal damage. One of the features we should be looking out for when using EEG is the presence or absence of reactivity during EEG to noxious stimulus such as endotracheal suctioning. Lack of reactivity is an important indication for poor neurological outcome. Another point raised was the reduced drug clearance in the setting of hypothermia even with short acting drugs such as propofol and alfentanil. The conclusion: We need to wait longer – at least 5 days. We need to utilize a combination of good neurological examination, neuro physiology, biomarkers and possibly neuro imaging (to out-rule other causes). The panel agreed that generalized epileptiform activity with no reactivity to noxious stimulus is a good predictor of poor outcome.
Li Huey Tan