ABG Analysis Case 3

Answer the questions at the bottom of the page when all data has been reviewed

History & Examination

We met this patient in Diabetic Emergencies Case 2. An insulin dependent diabetic with multiple end-organ complications, she was admitted with DKA. She mades a satisfactory recovery in the days following admission. She was taken off her sliding scale of insulin and continued on her regular subcutaneous insulin. However, her blood glucose proved difficult to control. On the morning of the 14th December, 4 days after her admission, she experienced a generalised tonic clonic seizure on the ward. A glucometer analysis confirms the presence of hypoglycemia which is managed appropriately. In the postictal phase she is somewhat disorientated.

Several hours following this episode, she spikes a fever (see Observation chart) and the Senior House Officer (SHO) on call is asked to review her. The patient is maintaining her SpO2 on oxygen. She is orientated and alert. Bowel sounds are normal. The SHO notes inspiratory crepitations at the right base which are new since she last examined the patient’s chest. A CXR shows minor patchy consolidation in the right lung field. The SHO commences IV antibiotics and restarts the insulin sliding scale.  In the light of the fever, she also, of course, sends blood cultures to the lab. The patient is allowed to drink and IV fluids are not commenced. There is no obvious major problem and the SHO has a moment, so, they talk about plans for the coming weeks. The SHO and the patient then discuss the need for an ABG. In the end it is decided to spare the patient this painful investigation. The SHO carries out the far less painful venous blood gas (VBG). She and the patient agree that if there is any cause for concern on the VBG, an ABG will be necessary. The result of the VBG is shown below.


The SHO returns to review the patient at 6 AM on the morning of the 15th December. The patient has had a difficult night. She has not slept. She is groggy and seems disorientated. The nursing staff report that there have been no further seizures. The SHO checks her BM (blood sugar) and confirms the absence of hypoglycemia. In fact her BMs have been running a little high overnight. Examination of her chest reveals marked bilateral inspiratory crepitations throughout both lung fields. Her observations are shown on the observations chart below and are a cause of great concern. The SHO carries out an ABG, requests a CXR (the results of which are shown below) and calls her registrar. The registrar arranges transfer to ICU.


Obs 2

Bernadette b

BK complication.Csi 001


Q1. What is the diagnosis at the time of the initial spike in temperature?

Q2. What do you think of the decision to do a VBG on the morning of the index event?

Q3. Why is the SHO so concerned on the morning of the 15th December (Dec)?

Q4. What abnormalities are present on the ABG, taken on the morning of the 15th Dec?

Q5. What abnormalities are present on her CXR on the 15th Dec?

Q6. What clinical syndrome has developed at the time of her CXR on the 15th Dec?

Q7. What are the key points in management of this syndrome?


Q1. What is the diagnosis at the time of the initial spike in temperature?


She may well have developed aspiration pneumonia. She has known gastroparesis and will have been at risk of aspiration during the seizure. Aspiration of gastric contents usually results in patchy consolidation evident on CXR. These X ray changes develop about two hours after the aspiration event. Aspiration tends to affect the right lung more commonly than the left due to the anatomy of the main bronchi.

Q2 What do you think of the decision to do a VBG on the morning of the index event?

She is requiring oxygen but is maintaining her SpO2 and, crucially, is alert. We are concerned about her metabolic state and the currently available evidence suggests that a VBG is adequate in terms of assessment of pH disturbances. There is no reason to suspect that hypercapnia is present, so we can rely on the SpO2. ABGs are painful!

Q3. Why is the SHO so concerned on the morning of the 15th December (Dec)?

She may be developing type II respiratory failure.

Don’t be fooled by the apparently normal SpO2 at this point in the story. Compared to the time point discussed in Q2, the difference in our approach on this occasion is the clinical context. She is maintaining her SpO2 in the normal range but the level of oxygen therapy required to achieve this has risen from 2L to 6L of oxygen per minute overnight. Her oxygen requirements are rising. This could be due to worsening of her lung condition (which was undoubtedly true, see below) but note also that her respiratory rate is falling. This suggests that she could be tiring resulting in reduced activity in the muscles of respiration. This places her at risk of hypoventilation with type II respiratory failure (CO2 retention). In this situation, respiratory arrest may be imminent. We now need to act quickly and an ABG will be part of her management.

Q4. What abnormalities are present on the ABG, taken on the morning of the 15th Dec?

She has indeed developed type II respiratory failure!

Bernadette b

This ABG result is very worrying. There is clear evidence of an acidosis (low pH). The level of respiratory acid (CO2) is elevated. This is a respiratory acidosis. This is not a metabolic acidosis – there has been no fall in her level of metabolic base (bicarbonate). This respiratory acidosis is acute. In the presence of a longstanding respiratory acidosis, the body compensates by raising the level of metabolic base (HCO3-). It takes about 24 to 48 hours for the kidneys and liver to mediate this effect and so in an acute respiratory acidosis (present for hours or less) the bicarbonate concentration remains in the normal range, as in our case.

Taking things to a slightly more advanced level. We know that as a type I diabetic she is prone to DKA. Could this be a chronic CO2 retention with failure of the bicarbonate to rise because an underlying co-existent metabolic acidosis is pushing down the level of this metabolic base? Well, no. Clinically, as outlined in the answer to question 3, the findings suggest fatigue with, therefore, an acute onset of hypoventilation. Biochemically, the ‘base excess’ is normal. The base excess tells us if there is an excess of acid or base in the extracellular fluid apart from carbonic acid related to elevated CO2. Clearly, with a normal base excess there is no excess acid in the system other than that related to elevated CO2. This is a pure respiratory acidosis (ie due purely to elevated CO2). There is no metabolic component. This is one of a few situations were understanding the ‘base excess’ is useful. You will also note that on the original VBG, there was no evidence of elevated CO2 levels. As VBG CO2 levels tend to be higher than ABG CO2 levels most physicians would take this as strong evidence of the absence of hypercapnia at the time the VBG was taken.

There is even more to worry about on this ABG. Although she has a normal SpO2 on the morning of the 15th December, she requires 6L of oxygen to maintain it! Analysing the ABG we find that her A-a gradient is grossly elevated at 145 mmHg (19 kPa). Her predicted A-a gradient is about 30 mmHg (4 kPa).

The likely sequence of events here is that the lung pathology triggered by aspiration resulted in a raised A-a gradient. There is evidence clinically and radiologically (see below) that the underlying lung pathology has worsened considerably. The patient’s respiratory effort was greatly increased overnight to maintain her paO2. She has a high degree of co-morbidity and is now becoming tired and failing to ventilate the alveoli. She is now developing type II respiratory failure. This move from a type I pattern to type II respiratory failure is life-threatening and must never be missed.

Q6. What clinical syndrome has developed at the time of her CXR on the 15th Dec?

Acute respiratory distress syndrome (ARDS).

One of the most interesting and deadly conditions in modern medicine. A difficult condition to define! In this case, the development of non-cardiogenic pulmonary edema secondary to severe lung inflammation (direct ARDS). We go through the history and nature of this condition in the article below.

Q7. What are the key points in management of this syndrome?

Supportive management

No specific treatment exists for ARDS (see article wich follows this case). Steroids have no role. The principle intervention remains treatment of the underlying cause. There is a logic in ‘running the patient dry’ with IV diuretics on ICU with appropriate monitoring. Even with intubation and management of the underlying triggering condition, ARDS may evolve over a period of weeks. Fibrotic changes within the lungs may render ventilation increasingly problematic as time passes. The mortality rate of ARDS remains high.

Subsequent Course

cappella di magi

There was a gradual deterioration in her condition over the days following admission to ICU. Mechanical ventilation became increasingly difficult. With worsening renal function and deterioration in her lung condition and in view of her pre-morbid state, a decision was made not to proceed to dialysis and to withdraw ventilatory support. She died shortly after being taken off the ventilator with her mother and sisters by her side.

The Senior Registrar of the medical firm arranged for all medical and nursing team members involved in her care to attend her funeral held on the weekend following her death. Standing in the middle of the huge West London congregation and in perfect unison with the local choir, the Senior House Officer who had looked after the patient during her final illness sang the beautiful Latin Hymns like an angel.


Acute Respiratory Distress Syndrome: A Tale of Two Wars

The Great War: ‘Shock Lung’

Ru de bois II resize

It was known in Sir William Osler’s time that severe inflammation in the lung from any cause may give rise to a clinical syndrome characterised by generalised pulmonary edema. In a high proportion of cases, irrespective of the initiating factor, the clinical course proved relentless with progression to respiratory failure and death. This syndrome gained infamy during the Great War.

The first major deployment of chemical weapons on the battlefield occurred on the Eastern and Western fronts during World War I. Inhalation of irritant poison gas resulted in the deaths of many soldiers from respiratory failure, a condition referred to at the time as ‘shock lung’. The physical and psychological suffering of the soldiers was immortalised by Wilfred Owens description of a gas attack on the Western front.

owen 3 a

Phosgene gas deployed at Fromelle near the Somme was believed to be responsible for many Australian and British deaths (centre image, 1916). Pictured right: 2nd Lt Wilfred Owen MC.

The initial agent deployed, chlorine gas, was highly irritant and exposure induced violent bouts of coughing limiting the dose inhaled by soldiers. Phosgene gas was developed as an alternative. Phosgene was less irritant, thereby ensuring soldiers were unaware of the exposure and inhaled a larger dose before using their masks. The inhaled agent induced an inflammatory response in the lung and, often after a latent symptom-free period of two to three days, was followed by death from intractable pulmonary oedema.

The Vietnam War: ‘Da Nang Lung’

da nang resize

Men of the 101st airborne call in helicopters to evacuate their wounded comrades. Medical evacuation from the battlefield reached its zenith in the Vietnam war with unexpected consequences.

The first major incursion into Vietnam by the US marines took place in the coastal region of Da Nang and, after a time, resulted in heavy casualties, wounded and dead. Following on from the lessons of the Korean war, the US military were determined that when soldiers were wounded on the battlefield they would be evacuated as quickly as possible to receive immediate medical attention, including fluid resuscitation for blood loss. To their astonishment, this approach seemed to worsen outcome. Many young men died in the days and weeks post evacuation from intractable pulmonary edema and respiratory failure. In many cases, the initial wounds did not involve the lungs but distant parts of the body. The pulmonary condition killing these men (‘Da Nang lung’) was so similar to the delayed ‘shock lung’ of phosgene gas attacks in World War I, that for some time the military were convinced that the Viet Cong were using a phosgene gas-like agent in the conflict. In reality, generalised pulmonary edema was occurring secondary to injuries distant from but linked to the lungs by a mechanism unknown. The tendency to develop pulmonary edema may have been aggravated by aggressive fluid replacement.

Acute Respiratory Distress Syndrome (ARDS)

It seems that in susceptible individuals, acute severe inflammation of the lung (direct ARDS), or inflammation in organs distant from the lung (indirect ARDS), can result in generalised increased permeability of the alveolar capillary membrane to fluids. To understand ARDS we need to ask a question. Under normal circumstances, how are the air filled alveoli kept free of fluid in lungs which are themselves full of fluid?

Alveolus FINAL NOW 1

(Image above) The formation of interstitial fluid in the lung from pulmonary capillaries is determined by the balance between Starlings forces (blue arrows) acting on the fluid in the capillaries. Under normal circumstances, the movement of fluid between the interstitial compartment and the alveoli is tightly regulated. Sodium ions enter the alveolar epithelial cells from fluid in the alveoli facilitated by sodium channels present in the apical membrane of the cells. Energy dependent Na+/K+ pumps (blue spheres) generate an osmotic gradient by extruding sodium ions from the cytosol of the alveolar epithelial cells into the interstitium. Possibly facilitated by aquaporins, water moves along the osmotic gradient generated by this process (black arrows). This results in the maintenance of a very thin layer of fluid (white arrow) on the alveolar luminal side of the alveolar epithelium.

Alveolus FINAL NOW 2

(Image above) Recent evidence suggests that cytokines released from macrophage in response to viral pneumonitis can inhibit the assembly and insertion of the energy-dependent Na+/K+ pump into the alveolar cell membrane (Peteranderl et al, JCI 2016, 126(4) p1566). Loss of the osmotic gradient discussed above would be expected to lead to fluid accumulation in the alveoli. There is experimental evidence that phosgene gas increases alveolar capillary permeability by a similar mechanism.

Intuitively, one would assume that direct destruction of alveolar epithelium with loss of all semblance of a permeability barrier must play a role in cases of direct ARDS. For example, weaponised chlorine gas reacted with water in the tissues to form hydrochloric acid and presumably destroyed elements of the barrier instantly. However, the cytokine hypothesis as outlined above is attractive. It would offer a mechanism by which distant severe trauma/inflammation (distant from the lungs) can be linked to the development of indirect ARDS.

ARDS Clinical Definition

Cambridge_-_Bridge_of_Sighs (1)

Like civilisation, ARDS is difficult to define but most of us know it when we see it.

Recently, the definition of ARDS has changed. The ‘Berlin’ definition (a European initiative endorsed by the American Thoracic Society), captures the main features of this historic syndrome.

The presence of ARDS is defined by new or worsening respiratory symptoms 1) within seven days of a recognised clinical trigger* with 2) bilateral opacities on chest X ray not fully explained by effusions, lobar lung collapse or nodules and 3) the respiratory distress cannot be fully explained by cardiac failure or fluid overload (in the best estimation of the attending physician).

Ventilatory requirements retain a role in the new definition. To diagnose ARDS we must show that a certain level of PEEP (positive end expiratory pressure) is required to generate adequate oxygenation. the degree of difficulty in oxygenating the patient is used as a measure of the severity of the condition.

ARDS sevjpg

(mmHg = kPa x 7.5) *PEEP (positive end expiratory pressure) or CPAP (continuous positive airway pressure) in mild cases.

Exclusion of heart failure as a cause of respiratory distress remains a central part of the diagnosis of ARDS. In the older definition this required placement of a Swann Ganz catheter with direct measurement of pulmonary capillary wedge pressure. This is no longer required. Clinical judgement, usually complimented by echocardiography, is considered sufficient. Also, the term acute lung injury (ALI) is now defunct. The new definition of ARDS retains a central role for the chest X ray in diagnosis.


Compare a normal CXR (A) to the ‘froth corrupted lung’ of ARDS (B). In (B) a pattern of shadowing consistent with alveolar consolidation is seen. Multiple ‘fluffy’ areas of shadowing with soft edges are present. The process appears to fan out from the hilum. Air bronchograms (white arrow) can be seen indicating that the airways are spared (filled with air rather than fluid). There may be evidence of acinar shadowing (black arrow). One cause of such an alveolar pattern of consolidation is alveolar edema.

*Some recognised triggers of ARDS

ARDS causes 2

DIC: disseminated intravascular coagulation

ARDS treatment

There is no specific treatment for ARDS. As a key feature of the syndrome is fluid accumulation in the alveoli, intensivists tend to ‘run the patient dry’ using inotropic support as necessary. When possible we treat the presumed underlying trigger of the condition aggressively. Careful attention is paid to the technique of ventilatory support employed in the acute phase as there is some evidence that excessive airway pressures generated during mechanical ventilation may worsen the long term outcome. Perhaps surprisingly, anti-inflammatory agents including steroids are of no benefit. Recovery is unpredictable and, after supportive measures have been instigated, is largely in the lap of the gods. A high proportion of patients will die either of respiratory failure in the acute phase or progress to intractable pulmonary fibrosis with increasing loss of lung compliance rendering mechanical ventilation difficult. In those who survive, the condition may resolve completely or evolve into a debilitating chronic restrictive lung disease characterised by pulmonary fibrosis.