One of the most intimidating things about entering the ICU for the first time is the “life support machine” – the mechanical ventilator. Although I have posted an extensive series of tutorials on Mechanical Ventilation, covering most of the modes, oxygen therapy and applied respiratory physiology, I have attempted, in this tutorial, to distill everything to the “least you have to know” in 40 minutes. Keep in mind that modern machines look more like iPhones, and are far easier to use than the devices I grew up with that looked to me, on day 1, like something in the cartoon below.
I start with a discussion about the difference between normal breathing, CPAP and Positive Pressure Ventilation (PPV). PEEP is, effectively, CPAP during PPV. I then go on to discuss pressure limited modes of ventilation; worldwide this are the most widely used modes in ICU. I limit my discussion to Pressure Assist Control, Volume Guaranteed Pressure Control (VG_PC) and Pressure Support Ventilation (PSV). VG-PC is a popular and flexible option as an ICU’s default mode. However, as it is a pressure controlled mode, there is significant variability in tidal volume and airway pressure from minute to minute.
Several important rules are emphasized: the tidal volume should, in general be lower than 6ml/kg of ideal body weight, the plateau pressure lower than 30cmH2O and the driving pressure lower than 15cmH2O. I introduce the Spontaneous Breathing Index (SBI = RR/TV in L). The magic number is 100. We use the SBI to determine the success of weaning on PSV.
Volume Controlled Ventilation is the predominant mode use in the Operating Rooms (Theatres), and Volume Assist Control is a popular mode in North America. In ICU you must set a peak inspiratory flow and be aware that this may be insufficient during assisted breaths and lead to dys-synchrony. Volume Control is often used in ARDS to “lock in” the Tidal Volume (TV) but the operator must be aware that the TV that matters is not what is dialed up on the ventilator, but what the patient exhales.
I go on to discuss how to assess the patient on invasive mechanical ventilation, by looking at whether they are breathing spontaneously, in which case we determine whether they are suitable for a Pressure Support wean or not, or whether or not there is a problem with oxygenation (increase FiO2, PEEP, Mean Airway Pressure and seriously consider Prone Positioning) or Ventilation (increase Respiratory Rate, Tidal Volume or both, reduce PEEP).
The final part of the tutorial looks at Non Invasive Ventilation (NIV), and I explain how, in general we only use 2 modes on standalone devices – CPAP and Spontaneous Timed (S/T). The latter is similar to PSV with a backup rate, but I point out that instead of PEEP+PS the breath is EPAP + IPAP and IPAP is not built upon IPAP, as is the case with PSV. If one is delivering NIV on an ICU ventilator, then “leak” adjustment or “leak sync” should be used.
I am now going to move on, in the Introduction to Critical Care course, to a systems based assessment of the patient where you are expected to compile measurements and observations from the clinical information system, radiologic system and monitors to construct an overview of the patient’s status. This is the crux of intensive care medicine and it is not easy. I am going to visit each system sequentially, and some systems will have multiple tutorials. By the end of this process, you will have compiled all of the data, assessed and processed it, and be ready for the big presentation.
The first tutorial in this part is an overview of patient assessment. It is relatively short but essential.
The Second tutorial in this sequence is on at neurological assessment in the ICU. It contains a discussion about the Glasgow Coma Scale, The Richmond Agitation Sedation Scale and CAM-ICU. I also cover the assessment of suffering (PAID) in critical care.
You will need to assess the patients neurologic status, whether or not they appear to be suffering and what interventions, both environmental and pharmacological, that you are administering to help them.
Since the 1920s it has been known that administration of chloride rich intravenous fluids, characterized by a reduced Sodium to Chloride strong ion difference (SID), causes a progressive metabolic acidosis. This iatrogenic hyperchloremic acidosis was particularly problematic in the era before lactate and ketone measurement was widely available, prolonging critical care stay and resulting in, often, unnecessary tests and therapies. During the 2000s a body of literature emerged supporting the hypothesis that hyperchloremia, defined as a plasma chloride of greater than 110mmol/l may be harmful. In a series of retrospective analyses, hyperchloremia was associated with increased mortality across a spectrum of disorders, including surgery and critical illness. Hyperchloremia was also associated with increased risk of kidney injury and the requirement for renal replacement therapy. There was also some data that hyperchloremia may be associated with reduced splanchnic blood flow.
A series of papers that looked at isotonic saline solution (ISS – 0.9% NaCl ), often referred to as “normal” saline, versus Plasmalyte 148 (PL) in Diabetic Ketoacidosis (DKA), demonstrated that ISS was associated with prolonged duration of stay in critical care, usually associated with persistent metabolic (hyperchloremic) acidosis. No studies to date have demonstrated superiority of ISS to PL Four major clinical trials – SALT ED, SMART, BaSics and PLUS– were conducted to compare outcomes of acute and critically ill patients randomized to either balanced salt solutions (sodium lactate products – Hartmann’s, Lactated Ringers or PL) or ISS. The first 2 studies demonstrated that ISS was associated with renal dysfunction and worse outcomes with sepsis. The BaSics and PLUS trials, in their initial reporting, showed no outcome differences. However, these trials were “catch” all ICU studies, including perioperative patients, patients pre-resuscitated with ISS, and, overall very little fluid was administered. The studies were grossly underpowered to detect outcome differences. However, subsequent systematic reviews and meta-analyses that included these data, and subgroup analyses of high risk patients, determined that fluid resuscitation with ISS was associated with worse 30 mortality, particularly in sepsis, and worse renal outcomes.
It is my view that, based on decades of research and experience, “normal” Saline (ISS) should not be used as a first line agent for fluid resuscitation in critical illness. I believe that the current international guidelines for the management of DKA are flawed in that they continue to recommend the administration of an agent that may well be toxic to patients, particularly when alternatives are easily available. Watch the video and make up your own mind.
This tutorial looks at an emerging problem in medicine – iatrogenically induced eugylcemic ketoacidosis, associated with the use of SGLT2 (sodium glucose cotransporter 2) inhibitor drugs, also known as Flozins.
There is a global pandemic of metabolic disease caused by escalating ingestion of carbohydrate rich ultra processed food. This results in central obesity, hepatic steatosis (fatty liver) and insulin resistance: together these findings are labelled the “Metabolic Syndrome” (MetS). MetS is associated with systemic inflammation and atherogenesis. In many cases it progresses to Type 2 Diabetes (T2D), the majority of treatments for which increase adiposity and escalate insulin resistance. SLGT2 inhibitors are a relatively new class of drug that work by increasing excretion of ingested glucose by blocking the Sodium-Glucose symporter channel in the proximal tubule of the nephron. The result is mild natiuresis and glycosuria. These agents have been proven effective in the management of T2D and are emerging as effective treatments for other diseases such as congestive cardiac failure and nephropathy. As the name of each of these medications involves the suffix -flozin – they are commonly termed “Flozin” drugs.
One of the major problem with the use of Flozins in the community is failure to discontinue the drug when fasting or not consuming calories. Glucose will continue to be wasted, often generated by gluconeogensis, suppressing insulin secretion, resulting in lipolysis and ketosis. As blood glucose is low there is insufficient insulin present to prevent ketoacidosis. This is one of the causes of euglycemic diabetic ketoacidosis (EDKA). EDKA is associated with both ketoacidosis and hyperchloremic acidosis.
The treatment of EDKA is dextrose (to restore the Kreb’s cycle and suppress ketosis) and insulin – to put some control on the metabolic system. The patient may require a couple of liters of resuscitation fluid – preferably sodium lactate solution (Hartmanns or LR). The ketosis resolves rapidly, but the acidosis resolves slowly because it is principally driven by hyperchloremia. Patients who are being treated with SGLT2 inhibitors that are scheduled for surgery should stop taking these drugs 3 days pre-op. If they are continued inadvertently or surgery is emergent, then a dextrose infusion should be considered and ketones checked routinely.
Tutorials 2 and 3 in the Introduction the Critical Care Series.
The first is titled “What is Critical Illness” and it covers the concept of Physiologic Reserve.
The third tutorial looks at the problem that many healthcare providers encounter: how do I identify the critically ill patient. In this tutorial I discuss the type of scenarios in which you might be called to the patients bedside. I principally discuss Early Warning Scores (EWS) and why I think they are helpful. I also mention the SOFA score.
Critical care medicine is the multidisciplinary healthcare specialty that cares for patients with acute, life-threatening illness or injury (SCCM definition).
Critical Care Medicine is a term used in the North America to describe the practice of medicine in intensive care units (ICU). Elsewhere it is known as Intensive Care Medicine (ICM); in Great Britain, ICUs are often referred to Intensive Therapy Units (ITU). A specialist who practices intensive care medicine is known as an intensivist, and has usually been trained and board certified in anesthesiology, surgery, internal medicine or pediatrics.
Critical Care Medicine is a relatively modern specialty; the first intensive care units opened in Europe in the late 1950s and rapidly spread to North America. Certification of training in this field did not occur in the United States until 1986. By the late 1990s, there were approximately 5000 intensive care units in the USA. For many years intensive care was something of a “free for all” struggle between various interest groups, with the patient often caught in the middle. This arose from the mistaken view of many physicians that intensive care patients were merely sicker versions of the patients that they already looked after on the wards. An open ICU model evolved, with the primary physician making the decisions and a support team of specialists acting as consultants. It has since been shown that the presence of a properly trained intensive care physician in the unit significantly reduces morbidity, mortality and cost. Modern critical care units tend to be “semi closed” with a multidisciplinary team, led by an intensivist, managing most aspects of the patient’s care. Limited external consultations take place, aside from microbiology/infectious diseases, radiology, or where specific specialist input may advance patient care (hematology, cardiology etc).
Three factors differentiate intensive care from other wards in hospitals: 1) a very high nurse to patient ratio, 2) the availability of invasive monitoring, 3) the use of mechanical and pharmacological life sustaining therapies (mechanical ventilation, vasopressors, continuous dialysis).
LEVELS OF CARE
As critical care units are specialized hospital wards, it is worthwhile to elaborate on how we label units. There are many ways to do this but the “level of care” paradigm is the most useful.
Level 0 = a standard ward with a nurse to patient ratio (NPR) of 1:6 or higher
Level 1 = advanced ward based care with a NPR of 1:3 or 1:4 – e.g. an extended recovery unit (PACU)
Level 2 = High Dependency Care – the patient usually has single organ failure or requires intensive monitoring, electrolyte replacement or extensive postoperative care (e.g. HDU). NP ratio is 1:2 or 1:3
Level 3 = Intensive Care. The patient may have multi organ failure, requires invasive ventilation and or continuous kidney replacement therapy, or is comatose or requiring aggressive resuscitation. The patient requires 1:1 NPR (UK Ire) – 1:2 USA.
There are several critical care units in any hospital, but they are not always labelled as such:
•Operating rooms and Recovery
•ED Resuscitation areas
•Labour Ward
•Coronary Care
•PACU (extended postoperative recovery)
•High Dependency (HDU)
•Intensive Care (ICU)
The Critical Illness Paradigm
Critical illness is a very specific series of disease syndromes that arise from an enormous spectrum of diseases
A wide variety of disease processes are treated with a limited number of interventions, in an intensive nursing environment.
Many doctors and nurses have a very poor understanding of what constitutes an intensive care patient: they are not merely standard medical or surgical patients, sicker than normal, perhaps plugged into ventilators. All intensive care patients fit into one of the following categories:
Patients admitted to intensive care for intensive monitoring, in anticipation of possible aggressive interventions: this is the coronary care model.
Patients admitted to units which act as extensions of the post-operative recovery room, allowing abnormal perioperative physiology to reverse, with or without modulation of the normal stress response. Post operative cardiac care is an example of this model.
Patients who require very intense nursing care, which would not be available elsewhere: an example of this is a burns unit.
Patients who do not necessarily require life sustaining treatments, but whose physiology is taken under control in order to prevent organ injury: neurosurgical critical care.
Patients who have minimal physiologic reserve, and who undergo acute potentially reversible injury, requiring life support until the abnormalities have been reversed and reserve restored: this is the archetypical medical intensive care patient (COPD with pneumonia requiring mechanical ventilation).
Patients who undergo a massive disruption to their physiology, due to an overwhelming stress response to injury, or inadequate compensation to the response: this is the patient frequently seen in surgical intensive care units – major trauma or sepsis such as pancreatitis.
It is important that you are able to differentiate between the types of patients that you look after in ICU: for routine post operative surgical patients fluid balance, analgesia and heart rate control may be the over-riding priorities, rather than feeding, for example. It is also important to realize that patients admitted under one category may enter another: a patient following coronary bypass surgery may develop severe sepsis.
It is important to differentiate patients who are in critical care units from those with “critical illness,” which is characterized by acute loss of physiologic reserve.
The patients in groups 5 and 6 have “critical illness”: their admission to ICU has followed an injury which has depleted endogenous reserves, and death is inevitable without life supporting interventions. These patients do not follow predictable courses of illness, such as “the ebb and flow paradigm”, originally described by Cuthbertson. In many cases the course of illness is prolonged, and the underlying causes difficult to discern. Indeed there appears to be great interpatient variability – two patients with the exact same injury may follow different paths: one may follow the standard stress response – acute compensation, followed by hypermetabolism and catabolism and, after 4 to 7 days, resolution with fluid mobilization and anabolism. The second patient may rapidly develop multi organ failure and remain in intensive care for a prolonged period of time. We do not know why this occurs, but suspect a hereditary element. To look at this another way, the standard stress response has evolved as the body’s mechanism to save itself and deal with major injury: the greater the injury, the greater the response. Conversely, an overwhelming response, which will lead to death without life support, cannot be considered normal.
Medical versus Surgical Intensive Care Units
Critical illness should not be compartmentalized into medical and surgical, the problems experienced by critically ill patients and the treatments given are essentially the same, although the causes may differ.
Critical illness is a very specific series of disease syndromes that arise from an enormous spectrum of causes. There is no such thing as a medical intensive care patient or a surgical critically ill patient; the syndromes are the same regardless of the origin, although the approach taken may differ amongst patient populations depending on age and chronic health issues. A good intensivist can seamlessly move from medical to surgical units, divisions set up for convenience. In many institutions, internationally, there is only one (mixed) unit, with no distinction between medical and surgical patients. Critical illness is a paradigm where patients are afflicted by syndromes such as ARDS, sepsis, kidneydysfunction, hemodynamic insufficiency, neuroendocrine insufficiency and exhaustion.
A wide variety of disease processes are treated with a limited number of interventions, in an intensive nursing environment.
The Multidisciplinary Team
The intensive care unit is not merely a room or series of room filled with patients attached to interventional technology, it is the home of an organization: the intensive care team. This team – doctors, nurses, therapists, chaplains and other support staff, builds an environment for healing, under the umbrellas of medicine, care and compassion and unit management.
A high quality Critical Care Team will contain the following members:
Critical Care Consultant (Intensivist) – ICU Clinical Nurse Manager Critical Care Medical Team Critical Care Nurses ICU Pharmacists ICU Physiotherapy Team IICU Occupational Therapists ICU Dieticians ICU Speech and Language Therapists Healthcare Assistants Administrative Assistants Chaplain Psychologist
Medicine, Care, Compassion and Organization
Critical Care is about medicine, care, compassion and organization.
The best intensive care units are the ones with the most effective management structures.
Patients are admitted to intensive care, for the most part, with one or more of the following problems: hemodynamic insufficiency, respiratory failure, abnormalities of fluid and electrolytes, sepsis and coma.
I frequently refer to the seven Cs of critical care:
Compassion
Communication (with patient and family).
Consideration (to patients, relatives and colleagues) and avoidance of Conflict.
Comfort: prevention of suffering
Carefulness (avoidance of injury)
Consistency
Closure (ethics and withdrawal of care).
It is not possible to build an effective critical care practice without high quality management which addresses the following issues:
1. Environment (patients, staff and visitors). 2. Organization Structure (multidisciplinary). 3. Teamwork. 4. Gatekeeping (appropriate bed usage). 5. Evidence based practice and cost effectiveness. 6. Continuous Education. 7. Audit with transparency.
Every good intensive care unit will have one or more medical and nursing directors, perhaps a business manager and very strong pharmacy and radiology, infectious disease/microbiology backup.
Critical care is characterized by a very high doctor and nurse to patient ratio. In critical care nurses are present at the bedside 24 hours per day. Not all of that time is spend “doing stuff” – hence, I prefer the moniker “Intensive Care” to “Intensive Therapy.”
In the early 1970s much of the world adopted the System International (SI) approach to scientific measurement. Unfortunately, the remainder of the world ignored it. This means that, today, we have different units presented in the scientific literature depending on the location of the source of the publication.
The USA is the most notable non SI country and this presents a problem in that the majority of English language textbooks and journals in medicine as well as a lot of the international guidelines and clinical pathways are derived in the US. In critical care this is important – as blood gasses are reported in mmHg in the USA (and most of the literature) and in kPa elsewhere – notably in Europe.
In many of my tutorials I have reported clinical “rules” such as the PaO2/FiO2 ratio, the Alveolar Gas Equation and the majority of the calculations in acid base – in mmHg. This series of two tutorials serve to right the balance. However there is a twist.
In this first tutorial I am not just rehashing the approach to oxygenation by swapping out mmHg for kPa. In fact, the use of kPa to measure and monitor oxygenation provides us with a significant helping hand. Effectively, as atmospheric gas is effectively 100kPa and Oxygen exerts 21% of that – Dalton’s law – then it is clear that the partial pressure of inspired oxygen (PiO2) is 21kPa. Oxygen is poorly soluble in blood and water – the solubility co-efficient is 0.225 – meaning that the quantity of oxygen dissolved in blood is the PaO2 x 0.225 kPa. Oxygen follows Henry’s law – meaning that solubility is related to temperature (37 degrees C) and pressure – the PiO2. In the best case scenario the PaO2 – the partial pressure of oxygen in arterial blood is 13kPa. That means that the gradient between PiO2 and PaO2 is, at a minimal, 8kPa. The greater the stretch between the two the larger the lung injury or ventilation perfusion mismatch.
The oxygen content of blood is 1.34 x Hb x SaO2/100 + (PaO2 x 0.225). I explore the impact of different FiO2s and ambient pressure on the blood oxygen content. Although dissolved oxygen is very low breathing air – the use of supplemental oxygen may dramatically increase it – particularly in hyperbaric conditions.
Finally I address the issue of PaO2/FiO2 as a way of quantifying oxygenation. The PF ratio, as we call it, is a significant component of the ARDS definition. A PF ratio of 200 in mmHg is equivalent to 25 in kPa and a ratio of 100 in mmHg is equivalent to 12.5 in kPa. An easier way to look at this, though, is to divide the PiO2 by the PaO2 – the numbers look similar but you now have a proportion in kPa. That PF ratio of 25 in kPa resolves to 0.25 meaning that only 25% of inspired oxygen is reaching the pulmonary veins (PaO2). Likewise a PF ratio of 12.5 in kPa (100 in mmHg) resolves to 0.125 – which means that only 1/8th of the inspired oxygen is delivered to arterial blood. I think that this is a really good way of assessing oxygenation – and a way of clarifying hypoxemia in your brain.
Proportional assist ventilation has been around in various shapes and forms since the late 1990s. The most advanced current iteration – PAV+ – is unique to Puritan Bennett ventilators. It is a closed loop mode of ventilation. That means that the ventilator dynamically changes the level of assistance that the patient receives in response to patient effort. PAV+ is neither volume controlled nor pressure controlled but is patient (and operator) controlled. The operator adjusts the percentage support that the ventilator delivers to the patient. The patient breathes – triggering the ventilator – and the ventilator amplifies the patient’s breath. Consequently the more work that the patient does to generate muscular effort the more work the ventilator performs to match the patient’s workload.
It has been known for some time that the diaphragm becomes both atrophic and dysfunctional in acute critical illness, in particular due to disuse during control of mechanical ventilation. In most assisted modes, all the patient needs to do is trigger the ventilator. Patient workload may be inversely proportional to ventilator workload. Frequently the patient’s diaphragm and ventilator are out of synchrony.
PAV+ is patient triggered and flow cycled so it should be seen as a form of pressure support ventilation. PAV+ contrasts with standard pressure support in that the degree of support changes from breath to breath and indeed within breath depending on patient effort. Pressure support delivers a fixed airway pressure for every single breath irrespective of patient effort. Consequently if we map patient effort to ventilator workload there is only one point where the two will intersect. Conversely in proportional assist ventilation the workload of the ventilator and the workload of the patient increase and decrease linearly.
PAV+ works by utilizing very high quality flow and pressure sensors. The ventilator determines when the patient initiates the breath and when the breath is completed. Having instructed the ventilator what proportion of work of breathing that the ventilator should perform, one observes, using a work of breathing bar, if the patient is doing satisfactory work or whether they need to increase or decrease their workload. The work of breathing (WOB) is determined by the ventilator by measuring compliance, resistance and intrinsic peep dynamically every 9 to 12 breaths. As such a Green Zone between 0.3 and 0.7 joules per litre is indicative of ideal work of breathing for the patient; I call this the “sweet spot.” As long as the patient’s WOB resides within the sweet spot of the toolbar the bedside clinician can be satisfied that the patient is both comfortable and safe.
As the tidal volume relates to the patient’s neural activity that results in diaphragmatic power one should not be unduly concerned about high or low tidal volumes in this mode.
If one wishes to put a patient on proportional assist ventilation it is imperative that one determines if the patient is breathing spontaneously and taking an adequate minute ventilation prior to using this mode. The reason for this is that there is no backup rate in PAV+. Usually one starts with 70% support: that means 70% of the work of breathing is performed by the ventilator on 30% by the patient. After a couple of minutes, once one has observed the work of breathing bar, one can make adjustments either to increase the workload of the ventilator or to reduce it by keeping the patient within that Green Zone sweet spot. Generally failure of the patient to settle on this mode is manifest by a respiratory rate of more than 35. Once the patient has been on 20% support for an hour or more and is awake, obeying commands, protecting their airway, and not being suctioned frequently then the patient can be extubated.
Studies that have looked at PAV+ versus pressure support have indicated that weaning is more rapid with PAV+.
This tutorial looks at hypernatremia and hyperosmolar syndrome. Hypernatremia is usually caused by three things: 1) Profound dehydration, 2) Too much sodium intake – most of the time this is due to over-resuscitation with isotonic fluids, 3) Central or Nephrogenic Diabetes Insipidis. I explain how to calculate water deficit and water replacement and how to evaluate and treat patients with diabetes insipidus. @ccmtutorials
I often wonder if the obsession amongst physicians regarding the prevention of Osmotic Demyelination Syndrome (ODS or Central Pontine Myelinolysis – CPM) results in adverse patient outcomes – for example a greater incidence of iatrogenic complications, prolonged length of stay etc.
In this discussion, I look at the history of ODS/CPM, how it became identified with rapid correction of hyponatremia and what patients are at particular risk of this disorder. In the second part of the discussion I address the re-ignited controversy about Sodium/Osmolality correction subsequent to the publication of a major study in NEJM Evidence in 2023.
Ultimately each clinician must make up their own minds on the evidence that is available. It appears to me that there is little or no risk of ODS/CPM in patients with acute hyponatremia, symptomatic or not, and those with a plasma sodium of greater than 120mmol/L. Patients with Sodium levels below 105mmol/L, alcoholics or cirrhotics and malnourished patient appear to be at very high risk. Finally attention should be paid not only to the speed of correction, but where the plasma sodium levels ends up. In many studies – ODS/CMP is a late diagnosis and patients, at the time of diagnosis are hypernatremic (greater than 145mmol/l) – although the rise in Sodium/Osmolality may appear slow over days or weeks.
ORtoICU 