College Answer

•    Causes can be divided into increased lactate production (including enhanced pyruvate production, reduced pyruvate conversion to CO2 & water or glucose, or preferential conversion of pyruvate to lactate), and diminished lactate utilisation.
•    Most causes in the critically ill are due to the many causes of tissue hypoperfusion [Type A] (resulting in increased production and decreased utilisation), or decreased utilisation due to liver disease (especially with use of lactate containing fluids in renal replacement therapy). Other common causes include seizures, beta-2 adrenergic agonists (eg. adrenaline and salbutamol), metformin (uncertain mechanism) and post-cardiac surgery. Consider also d- lactic acidosis associated with the short bowel syndrome.
•    Principles of management include correcting hypoperfusion (fluids, inotropes, vasopressors), and if possible, correction of underlying disorder (treat seizures, shivering, glucose abnormalities, etc.) and removal of offending drugs (including metformin, adrenaline, renal replacement fluid).

Discussion

Causes of lactic acidosis  are discussed at length in another chapter.

In brief, one may classify it according to several classification systems, of which the Cohen-Woods classification is for some reason the more popular.

The Cohen-Woods classification is as follows:

• Type A (lactic acidosis due to impaired tissue perfusion)
• Type B (lactic acidosis with adequate tissue perfusion).
  • Type B1 (lactic acidosis a result of “disease states”)
  • Type B2 (drug-related)
  • Type B3 ( the consequence of inborn errors of metabolism)

Causes of lactic acidosis can be better classified according to the lesser-known (but biochemically superior) Phuypers and Pierce system:

The approach to the management of lactic acidosis can be divided into two main approaches:

• Manage the underlying cause of the lactic acidosis - and the precise management strategy obviously depends on the aetiology
• Increase the clearance of lactate artificially using CVVHDF (which is probably not a very effective strategy, and represents immature "lactate-centric" management).

College Answer

A metabolic acidosis is a process which, if uncorrected, would lead to an acidaemia.  It is usually associated with a low bicarbonate concentration (or total CO2), but an acidosis may be masked by a co-existing metabolic alkalosis.  A simple classification is to categorise acidosis into accumulation of acids (measured, i.e. chloride [hyperchloraemic metabolic acidosis] or unmeasured [increased anion gap metabolic acidosis]), or renal or gastrointestinal loss of bicarbonate (with absorption of chloride, resulting in hyperchloraemic metabolic acidosis). The anion gap (Na + K – Cl – HCO3) is usually determined primarily by negatively charged plasma proteins and has a range of approximately 10 to 16 mmol/L.  This will be decreased by about 2.5 mmol/L for every decrease inalbumin by 10 g/L.  An increased anion gap (which can occur in the absence of a low bicarbonate concentration) may be due to a fall in unmeasured cations (Ca, Mg), or more commonly to the presence of unmeasured anions (e.g. lactate [d- or l-], ketoacids, formate [methanol], glycolate and oxalate [ethylene glycol].  Some of these can be specifically measured.  Calculation of an osmolar gap may also help as a screening test for methanol or ethylene glycol intoxication once alcohol has been excluded (calculated osmolality = 2*Na + Glucose + Urea + ethanol/4.6).   Urinary pH (inappropriately alkaline for an acidaemia) and electrolytes may facilitate eliciting the specific cause of the renal bicarbonate loss (e.g. renal tubular acidosis).

Discussion

The college answer - I assume - outlines in prose the features which the examiners would have been looking for. The answer, in my opinion, calls for a systematic and structural response. The candidate ought to demonstrate that their approach to the evaluation of metabolic acidosis is not a disorganised bafflement of various physiological principles and biochemical tests.

Thus:

• Calculate the anion gap
  • (Na⁺ + K⁺) - (Cl^- + HCO₃^-)
  • The anion gap achieves a diagnostic classification of metabolic acidosis
  • A raised anion gap identifies unmeasured anions (eg. lactate, ketones, metabolic byproducts of toxic alcohols) as potential causes of the acidosis
  • A normal anion gap identifies changes in chloride and bicarbonate as causes of the metabolic acidosis
• Calculate the delta ratio
  • (change in antion gap) / (change in bicarbonate)
  • The delta ratio quantifies the contribution to the acidosis of unmeasured anions and the chloride-bicarbonate balance
  • This way, mixed acid-base disorders may be revealed
• For normal anion gap metabolic acidosis, measure urinary electrolytes and calculate the urinary anion gap
  • Urinary anion gap can be used to differentiate between gastrointestinal and renal causes for the normal anion gap metabolic acidosis.
  • A high urinary anion gap suggests that there is a renal cause for the acidosis (i.e. that chloride and ammonium excretion is inappropriately low)
  • A low or negative urinary anion gap suggests that there is an appropriate attempt to excrete chloride and ammonium, and that the cause of acidosis is extrarenal.
• For high anion gap metabolic acidosis, measure the serum osmolality and calculate the osmolar gap
  • The osmolar gap will reveal whether there are osmotically active substances in the bloodstream which are not measured as a part of the normal bicohemistry. These may be responsible for the raised anion gap metabolic acidosis.
  • The osmolar gap may narrow the list of differentials

Given that one cannot predict the opinions of the examiners regarding Stewart's physicochemical approach to acid-base disorders, in an answer like this one, the candidate should probably give it a miss.

College Answer

Evaluation of causes of metabolic alkalosis requires a systematic approach involving history, examination and some specific investigations. Categories of aetiology include loss
of hydrogen ions (gastrointestinal, renal), intracellular shift of hydrogen ions, administration of alkali, and contraction alkalosis. History and examination will reveal, documented fluid losses (vomiting & gastric losses, laxative induced diarrhoea), volume depletion (loss of bicarbonate free fluids), administered drugs (mineralocorticoids, diuretics, and antacids in renal failure), alkali (bicarbonate, lactate, citrate etc) and recent hypercapnia. Investigations may reveal hypokalemia (with hydrogen shifting into cells), hypochloremia and urinary findings may include excessive potassium excretion (reabsorbing hydrogen), alkaline pH (increased bicarbonate) and inappropriately elevated chloride excretion (diuretic therapy, hypokalaemia).

Discussion

This question closely resembles Question 7  from the first paper of 2008.

College Answer

The  extent of  respiratory compensation for  a  metabolic disorder is  determined by the balance between the abnormality in the pH (hence the drive to change), and how hard it is to get  there  (eg.  work  of  respiratory muscles in  hyperventilation).   A  knowledge of  the expected degree of compensation for a given acid base status is necessary to determine the presence of an additional respiratory disorder.  Two traditional methods have been used: use of formulae, and the use of a standardized diagram.

The direction of change in the CO2 should be to normalise the pH for the underlying disorder.  A normal pH indicates an additional process is present.  The commonest relevant formulae to estimate the PaCO2 in use are:

For a metabolic acidosis, the expected PaCO2 = the last two digits of the pH (+/- 2-5 mmHg; from pH 7.1 to 7.6; Narins 1980), or the expected PaCO2 = (1.5 X measured bicarbonate) + 8 (+/- 2).  The measured bicarbonate, not the standard bicarbonate, must be used.   Other approachs include: expected change in PaCO2 = Standard Base Excess (Schlichtig R et al Crit Care Med 1998); 1.2 mmHg fall in PaCO2 for each 1 mmol/L reduction in HCO3.

For a metabolic alkalosis, the same equation is used, though the reliability may be less than with a metabolic acidosis.  Expected PaCO2 = the last two digits of the pH (+/- 2-5; from pH 7.1 to 7.6). Other approachs: change in PaCO2 = 0.6 X Standard Base Excess (Schlichtig R et al Crit Care Med 1998); 0.7 mmHg rise in PaCO2 for each 1 mmol/L increase in HCO3.

Discussion

This question regards the routine bedside tests for adequacy of compensation.

Specifically it is the magnitude of change of CO2 in response to the changes in pH.

One recalls several rules. None are especially precise. This whole issue is discussed at great length in the chapter on the assessment of compensation by the Boston and Copenhagen methods.

However, this question specifically asks for Winter's Rule:

1) In metabolic acidosis, PCO₂ = (1.5 x HCO₃) + 8 ... within a range of plus-minus 2mmHg

2) In metabolic alkalosis, PCO₂ = (0.7 x HCO₃) + 20 ... within a range of plus-minus 5mmHg

College Answer

Severe acidemia due to a mixed severe respiratory and metabolic (lactate) acidosis. Apparently adequate oxygenation (very low A-a gradient). The elevation in lactate may have been due to tissue hypoxia but the PaO2 at present might appear adequate. She is likely to have received intravenous adrenaline and possibly salbutamol, both can cause lactic acidosis (due to inhibition of pyruvate dehydrogenase (Day NP et al Lancet 1996;348:219-223) and possibly other effects on glycolytic enzymes, accelerating glycolysis).
Measures to increase alveolar ventilation are urgently required. Evaluation of airway, breathing and circulation (ABC) is the first response. Urgent chest auscultation and observation should be performed. Capnography (+/- laryngoscopy) should be used to confirm intra-tracheal position of ETT. To achieve a PO2 of 90 with almost zero A-a gradient, the tube must be in the trachea; for the same reason, endobronchial intubation is unlikely but should be excluded. Consider pneumothorax
– as this might compromise alveolar ventilation. Any leak should be identified (if the endotracheal tube is un-cuffed it could be replaced with a larger tube or a cuffed tube). Check to make sure the ventilator tubing is not leaking/unattached. The tidal volume and respiratory rate must be checked together with the peak pressure limit (if the patient is being ventilated with a mechanical ventilator). The chest excursion should be observed to ensure that there is a rise during inhalation. If excursion appears inadequate it is appropriate to try an increase in the volume or pressure and observe
whether the ventilation improved. The extent of gas trapping needs to be evaluated. If there is significant gas trapping then a reduction of respiratory rate is required (to prolong the expiratory time), however given the current PaCO2 this may require an increase in tidal volume. An alternative strategy would be to plan for regular 30 second disconnections every 5 minutes or so (particularly if there was associated hypotension).

Discussion

This is not a pure ABG interpretation question; it required some thinking about the management of severe asthma.

But first, lets interpret the ABG.

1. The A-a gradient is normal:
   (0.5 x 713) - (212 x 1.25) = 91.5
   Thus, A-a = (91.5 -90) = 1.5mmHg
2. There is acidaemia
3. The PaCO₂ is contributing to the acidosis
4. The SBE is not given, but the bicarbonate is normal, suggesting that there is either no metabolic acidosis, or that the metabolic acidosis coexists with a metabolic alkalosis. The presence of a raised lactate (12mmol/L) suggests that the latter is correct. 
5. This question gives a history of asthma and cardiac arrest which suggests that the respiratory acidosis is acute. The expected HCO₃^- in this setting would be 41.2:
   bicarbonate compensation = (212-40)/10 multiplied by 1 = 17.2 mmol/L
   Thus, there is also a metabolic acidosis present here.

The lactate can be left to its own devices. Given that you would be using vast amounts of salbutamol here, it can be expected to persist for some time.

The PaCO₂ requires immediate attention.

The ventilation strategies in status asthmaticus are discussed elsewhere.

A systematic approach would resemble the following:

Airway:

• Ensure the ETT is not kinked or blocked with secretions, and is of a satisfactory internal diameter

Breathing:

• Assess the expiratory flow using ventilator waveforms and capnometry
• Adjust the I:E ratio to allow for prolonged expiration
• Adjust the repiratory rate to allow for prolonged expiration
• Consider using a low PEEP.

Circulation

• Assess the extent of dynamic hyperinflation
• Assess intravascular volume and aim for a higher CVP, to counteract the preload-limiting effects of dynamic hyperinflation
• Consider regular disconnections of the ETT, and/or manual decompression

College Answer

Results demonstrate an increased anion gap metabolic acidosis with mild hyperglycaemia and hyperosmolar hyponatraemia. The marked anion gap (39.5) is not solely explained by the lactate level, and ketones or renal failure are not present. There is also a large osmolar gap (calculated osmolality = 340 –261=81), which suggests an additional agent/toxin is causing acidosis and having osmotic effect.
Differential diagnosis includes:
•    Methanol: History (“hootch” consumption); measurement of methanol and formate levels often takes time, but negative ethanol may be useful. The lack of renal dysfunction does not exclude methanol.
•    Ethylene glycol: History (?suicidal intent), plasma ionised calcium, oxalate crystalluria, Woods lamp examination for fluorescence; unlikely given normal renal function
•    Alcoholic ketoacidosis: measure ethanol and plasma beta- hydroxybutyrate. Negative urinary ketones (acetoacetate) might reflect a low redox state, with most of the keto-anion being in the form of beta-hydroxybutyrate, and with a huge amount of acetone and glycerol (plus some ethyl alcohol) causing the osmolar gap. The enormous anion gap would then be a mixture of lactate, beta-hydroxybutyrate and some acetate from ethanol metabolism.
•    Pyroglutamic acidosis: History of paracetamol ingestion in the face of liver dysfunction.
Measure pyrogluamic acid levels. Less likely as large osmolar gap.
•    Salicylic acid: History (?suicidal intent), measure salicylate levels, may have respiratory alkalosis. Less likely as large osmolar gap.
Hypoadrenalism could be considered (hyponatraemia, hyperkalaemia, metabolic acidosis), but on its own does not explain the osmolar gap and the non-lactate component of the anion gap.
DKA is on the differential, but the osmolar gap (acetone, glycerol) is higher than it usually gets, unconsciousness is not normally a feature and needs to be explained separately, and the lactate is unusually high.
Factitious causes of hyponatraemia (hyperlipidaemia etc) might have been mentioned. However, although they will artefactually raise the osmolar gap, they do not increase the anion gap, and they don’t cause acidosis or unconsciousness.

Discussion

Let us dissect these results systematically.

1. The A-a gradient cannot be calculated
2. There is no pH reported
3. The PaCO2 is not reported
4. The SBE is not reported, but the bicabonate is less than 5mmol/L, suggesting a severe metabolic acidosis
5. The respiratory compensation cannot be assessed
6. The anion gap is (119) - (80 + 5) = 34, or 39.5 when calculated with potassium
   The delta ratio, assuming a normal anion gap is 12 and a normal bicarbonate is 24, would therefore be (34 - 12) / (24 - 5) = 1.15. This delta ratio suggests that there is a pure high anion gap metabolic acidosis here.
7. The osmolar gap is raised:
   340 - [ (2 × 119) + 13 + 10] = 79

The lactate cannot account for this increase in anion gap, much less the increase in osmolar gap.

Differential diagnoses of a raised osmolar gap in metabolic acidosis include the following:

• Methanol and other toxic alcohol poisoning (the college examiners seriously called it "hootch", which as far as I can tell is crude American slang from the late 1890s)
• Ethanol and alcoholic ketoacidosis (most of the anions might be in the form of β-hydroxybutyrate, which is not measured wih the urinary ketone assay)
• Pyroglutamic acidosis
• Salicylate toxicity

College Answer

Candidates were expected to outline each of the important abnormalities and how they should influence her management. One approach would be the following:

•    The biochemistry supports a diagnosis of hyperosmolar hyperglycaemic syndrome, with post-ictal lactic acidosis. There is also a likely component of ketoacidosis.

•    The high urea / creatinine ratio confirms significant dehydration – most people with this condition are at least 10% dehydrated by body weight – usually more. She will require steady rehydration over 12 to 24 hours.

•    The sodium adjusted to normoglycaemia is about 153 mmol/L. This means that, after intravascular volume is restored with isotonic fluids, rehydration should be conducted with relatively hypotonic fluids eg 0.45% saline initially.

•    The metabolic acidosis is uncompensated. The minute ventilation should be increased (if this can be done safely), to replicate appropriate respiratory compensation.

•    There is a raised anion gap (34.5 mEq/L without K). The degree of lactate elevation is insufficient to explain the anion gap, indicating probable co-existing ketoacidosis. Beta- hydroxybutyrate should therefore be measured. An insulin infusion (0.1 U/kg/hr) is required for gentle glucose correction and reversal of ketoacidosis.

•    The plasma potassium is already reduced, indicating a severe deficit and the need for immediate replacement in the absence of anuria. She is likely to need 10 - 20 mmol/hr or more for many hours, since the total deficit will exceed 6 mmol/kg.

•    Blood gases, Na, K and glucose will need frequent (hourly) measurement initially. Rapid reductions in osmolality should be avoided (using sodium adjusted to normoglycaemia as a surrogate).

The A-a gradient is significantly elevated at 380 mm Hg. This could indicate aspiration pneumonitis, pneumonia, or segmental collapse – even endobronchial intubation. These possibilities should be looked for clinically and on CXR.

Discussion

Let us dissect these results systematically.

1. The A-a gradient is high:
   PAO₂ = (713) - (43 × 1.25) = 659.25
   Thus, A-a = ( 659.25 - 280) = 379.25mmHg.
2. There is acidaemia
3. The PaCO₂ is not compensatory
4. The SBE is -16.8, suggesting a severe metabolic acidosis
5. The respiratory compensation is inadequate - the expected PaCO₂(11.5 × 1.5) + 8 = 25.25mmHg, and thus there is also a respiratory acidosis
6. The anion gap is (128) - (82 + 11.5) = 34.5, or 37.6 when calculated with potassium.
   The delta ratio, assuming a normal anion gap is 12 and a normal bicarbonate is 24, would therefore be (34.5 - 12) / (24 - 11.5) = 1.8. 
   This delta ratio suggests that there is a pure high anion gap metabolic acidosis here, with perhaps just a hint of pre-existing metabolic alkalosis.
    

The lactate is 9.2, which does not account for much of the anion gap, suggesting that other unmeasured anions are present.

If one were to ignore the HONK, one would leave it at that.

However, most people would notice that the glucose is 79.

This is slightly abnormal. And it influences the sodium value.

The corrected sodium is 128 + (glucose/4), or 147.8 mmol/L.

Because the college do not show their working, it is impossible to say how they managed to get a sodium value of 153. If one were to use the standard formula, one gets 147.75mmol/L.

One may also use a slightly less standard formula, where for every 5.6mmol/L of glucose, the sodium decreases by 1.6mmol/L for the first 25mmol of glucose, and by 2.4mmol/L for hyperglycaemia above 25mmol/L. In such a case, the corrected sodium is 158:

= measured Na⁺ plus (25 / 5.6) × 1.6, plus [(79-25)/5.6]× 2.4

In any case, there is hypernatremia.

Anyway. The college question asks how any of this information would influence your management.

I will go through the abnormalities systematically.

• Acidosis
  • This is a combination of several contributing aetiologies:
    • uremic acidosis
    • post-ictal lactic acidosis
    • ketoacidosis
    • respiratory acidosis
  • The ketoacidosis should be confirmed with a blood ketone (hydroxybutyrate) level.
  • Management would consist of the following meaasures:
    • careful rehydration
    • insulin therapy (the college recommends 0.1 unit/kg/hr)
    • increased minute ventilation
• Hypokalemia
  • According to HONK literature, the total deficit will likely exceed 5-15mmol/kg. Aggressive replacement is in order.
• A-a gradient
  • For whatever reason, this patient is relatively hypoxic.
  • This should be investigated clinically, and with a CXR.

College Answer

The major abnormalities are:

a)  Metabolic alkalosis with respiratory compensation

b)  Hypokalemia and hypochloremia

c)  Normal anion gap

d)  An increased apparent strong ion difference [(Na + K) – Cl] = 47

Possible causes in this patient include

a)  Diuretic therapy

b)  Steroid therapy

c)  Vomiting from gastric outlet obstruction

d)  Post hypercapnic alkalosis

Outline the management of the metabolic and acid-base disturbance.

1) Normal saline administration

2) K supplements

3) Acetazolamide

Discussion

Let us dissect these results systematically.

1. The A-a gradient is high:
   PAO2 = (0.4 × 713) - (47 × 1.25) = 226.45
   Thus, A-a = ( 226.45 - 58) = 168.mmHg.
   But you did not need to know the A-a gradient to recognise that the patient is hypoxic.
2. There is alkalaemia
3. The PaCO2 is compensatory
4. The SBE is 10.2, suggesting a metabolic alkalosis
5. The respiratory compensation is adequate - the expected PaCO₂( 0.7 × 34.8) + 20 = 44.4mmHg, and this falls within the +/- 5mmHg error range of this blood gas rule. If we were to use the Copenhagen rules, the expected PaCO₂ would be around 50mmHg (40+ SBE).

Thus, this is a reasonably well compensated metabolic alkalosis.

There is also hypochloraemia and hypokalaemia.

The patient has chronic cardiac and respiratory disease, and has been vomiting.

Thus, the broad list of potential causes of metabolic alkalosis can be narrowed to the following:

• Diuretics (for CCF)
• Corticosteroids (for COPD)
• Gastric secretion loss though vomiting
• β-lactam use (for exacerbation of COPD)
• Post-hypercapneic alkalosis is possible, but less likely (because this metabolic alkalosis is well-compensated, and with post-hypercapneic alkalosis the respiratory compensation is by definition inadequate, i.e. the calculated "expected" PaCO₂ should be higher than the measured PaCO₂)

The college then go on to suggest three uninspired management options. For instance, they want to replace potassium, which is a sensible reaction to a low potassium, but not an elegant solution to any underlying problem. Similarly, acetazolamide and sodium chloride may seem essentially cosmetic measures, as they only serve to improve the appearance of the blood gas. 

Management of metabolic alkalosis is discussed in greater detail elsewhere. Most of the detail can be found in the following literature sources:

• A good old article on metabolic alkalosis in ICU  by Webster and Kulkarni (1999)
• A more recent article on extreme metabolic alkalosis (Tripathi, 2009).  
• An even more recent (2015) UpToDate article on this topic.

In brief:

• Address the underlying problem:
  • Stop the diuretics, treat the hyperaldosteronism, etc etc.
• Correct volume depletion - ideally with normal saline
• Correct hypokalemia - ideally with potassium chloride
• Alternative strategies if the patient is fluid-overloaded:
  • Acetazolamide
  • Hydrochloric acid (HCl)
  • Infusions of cationic amino acids such as lysine and arginine
  • Correction of hypoalbuminaemia

College Answer

a) Base excess is defined as the amount of strong acid or base required to titrate pH of an in vitro sample of blood back to 7.40 at 37C at a PCO2 of 40 mm Hg.

b) Negative base excess without anion gap elevation

1)  Dilutional acidosis from saline resuscitation
2)  Renal tubular acidosis
3)  Ureterosigmoidostomy
4)  Acetazolamide therapy

c)  Lactic acidosis in a patient with pre-existing metabolic alkalosis

Discussion

Base excess definition

• Dose of acid or base required to return the pH of a blood sample
• Measured at standard conditions: 37°C and 40mmHg PaCO₂
• Thus, isolates the metabolic disturbance from the respiratory

Standard base excess

• Dose of acid or base required to return the pH of an anaemic blood sample
• Calculated for a Hb of 50g/L
• Haemoglobin buffers both the intravascular and the extravascular fluid
• Thus, SBE assesses the buffering of the whole extracellular fluid, not just the haemoglobin-rich intravascular fluid

"Base Excess" is the amount of acid or base required to titrate a blood sample (of whole blood) to a pH of 7.40, at standard temperature and pressure, with a standard PaCO₂ of 40mmHg.

The "Standard Base Excess" is different because it uses extracellular fluid rather than whole blood. (of course, you dont sample the extracellular fluid - the ABG machine calculates the SBE for anaemic blood, with a Hb of 50g/L). The argument for this is the buffering ability of haemoglobin. It would be inappropriate to extrapolate whole blood findings to the total extracellular fluid, because though circulating haemoglobin buffers all extracellular fluid, it does so from the intravascular compartment to which it is confined.

An example of a base deficit in the presence of a normal anion gap is essentially any cause of normal anion gap metabolic acidosis- take your pick.

Examples of a normal base excess in the presence of a raised anion gap would include any situation where the high anion gap metabolic acidosis occurs in the setting of a chronic metabolic alkalosis. A favourite example is a raised lactate in the diuretic-using CCF patient, but one could just as easily use the case of the torrentially vomiting methanol drinker. LITFL also mention salicylate toxicity and HAGMA masked by uncorrected hypoalbuminaemia.

College Answer

a} List the acid-base abnormalities

High anion gap metabolic acidosis with raised lactate Metabolic alkalosis (Delta BE <Delta AG) Respiratory alkalosis

b)  What are the causes of elevated plasma lactate in sepsis? 
1)  Circulatory failure due to hypotension and hypoxia
2)  Microvascular shunting and mitochondrial failure (cytopathic hypoxia)
3)  Use of adrenaline as an inotrope
4)  Inhibition of pyruvate dehydrogenase (PDH) by endotoxin.

c)  Name 3 drugs which result in plasma hyperlactaemia
-     Catecholamines
-    Metformin I Phenformin 
-      Alcohols
-    Cyanide, nitroprusside
-     Salicylates

Discussion

This question is frequently repeated. Notable duplicates include the following:

• Question 6.4 from the first paper of 2013 (a detailed discussion is carried out here)
• Question 22.2 from the second paper of 2011
• Question 6.4 from the first paper of 2011
• Question 24 from the first paper of 2010
• Question 18 from the first paper of 2007

College Answer

Definition AG -a derived variable for the evaluation of metabolic acidosis to 
determine the presence of unmeasured anions.

AG = [(Na + K)-(Cl + HC03)], normal reference range: 8-12 meq/ L

Utility: A raised AG is seen with elevated lactate, ketoacidosis, salicylates, alcohol 
poisonings, and pyroglutamate

Advantages of the anion gap: 
a) A simple measure to quantify and evaluate acid-base disturbance 
b) Can be· easily done at the bedside

Limitations: 
1) Reduced unmeasured anions such as hypoalbuminemia (frequently seen in critical 
illness) will reduce the AG and may mask an elevated AG 
2) UnmeasUred cations such as elevated Li and hyperglobulinemia will reduce AG. 
3) Hypercalcemia and hypermagnesemia will also reduce the AG. 
4) Calculation of AG involves measurement of electrolytes and therefore depends on 
the accuracy of the measurement process.

To overcome the effects of the hypoalbuminemia on the AG, the corrected AG can be 
used which is AG + (0.25 * (40-albumin) expressed in G/L

Discussion

The article by Kraut and Nicolaos is an excellent longform answer to this question.

The question itself is not even a "critically evaluate" type of question. It asks simply for the advantages and limitations.

Thus, there they are.

Advantages of the anion gap

• Easy to calculate: (Na⁺ + K⁺) - (Cl^- + HCO₃^-)
• Offers a simple stratification of acid-base disorders into HAGMA and NAGMA

Limitations of the anion gap

• Subject to laboratory error in the measurement of any of the constituents
• Spurious sodium results can alter the anion gap (eg. in "pseudohyponatremia")
• Spurious chloride results can alter the anion gap (eg. bromide and iodide can be mistaken for chloride in the laboratory)
• Unmeasured or uncounted cation excess can alter the anion gap (eg. lithium, calcium, magnesium)
• Strongly cationic drugs can decrease the anion gap, or even make it negative (eg. polymyxin B)
• Modified by hypoalbuminaemia, and needs to be corrected (there is a 1mmol/L decrease in the "normal" anion gap for every 4g/L of albumin deficit below 40g/L)
• Modified by hyperprotinaemia, eg. in some sort of myeloma-like illness (an excess of charged proteins changes the anion gap; it is impossible to say in which direction, as the proteins may be anionic or cationic)

A discussion of the anion gap is available locally in two forms: as a quick revision summary and as a massive rambling digression.

College Answer

1.  Compensated respiratory alkalosis, or else a respiratory alkalosis superimposed on

a normal anion gap metabolic acidosis.

2.  Most likely rhabdomyolysis secondary to severe hypokalaemia. We need to examine muscle compartments, and measure the plasma CK.

3.  The mild metabolic acidosis may just be compensatory for her respiratory alkalosis. However, it could be a primary process (with a separate superimposed respiratory alkalosis from anxiety for example) in which case it is more consistent with enteric potassium loss, or else a proximal (Type 2) renal tubular acidosis. However urinary potassium loss of 18 mmol / day is low, and close to the obligatory minimum. Therefore bowel loss is more likely, or else transcellular potassium shifts (periodic paralysis). The persistent hypokalaemia after 24 hours of aggressive replacement makes periodic paralysis unlikely, as is the fact that this is the first episode. Apart from that, the absence of metabolic alkalosis and urinary potassium wasting are against Cushing’s and Conn’s syndromes as well as diuretic abuse, and the absence of hypertension is also against Cushing’s and Conn’s syndrome. Therefore on balance and with the history of diarrhoea, bowel loss is the most likely.

Bowel loss (eg villous adenoma, aperient abuse)

Diuretic abuse

Conn’s syndrome

Cushing’s syndrome

Periodic paralysis.

Renal tubular acidosis

Discussion

Let us dissect these results systematically.

1. The A-a gradient cannot be calculated
2. There is no acidaemia
3. The PaCO₂ suggests a compensatory respiratory alkalosis
4. The SBE is -6, suggesting a metabolic acidosis
5. The respiratory compensation is excessive; the expected PaCO2(17 × 1.5) + 8 = 33.5mmHg, and thus there is definitely also a respiratory alkalosis
6. The anion gap is normal:
   (144) - (119 + 17) = 8, or 10.9 when calculated with potassium

The urinary potassium loss is also normal, suggesting that gastrointestinal potassium losses must account for this hypokalemia.

Thus: this is a normal anion gap metabolic acidosis with respiratory alkalosis.

Let us leave the respiratory alkalosis well alone for the moment.

There are several differentials  for NAGMA.

The college asks for an explanation of the raised enzyme levels. The LDH and AST are not exclusively hepatic enzymes; they can also leak from damaged muscle, and the lady has presented with weakness, which might suggest that she is suffering from hypokalemia-associated rhabdomyolysis (though unless I am grossly mistaken, damaged muscle should also leak potassium, resulting in hyperkalemia). Certainly a CK an urinary/serum myoglobin levels will set this matter straight.

Differential diagnosis of hypokalemia is discussed at lengths elsewhere.

Below is a table of differentials which is organised according to the findings of investigations.

This lady falls into the "low urinary potassium" category, which suggests that the potassium losses are occurring though the gut. Familial hypokalemic periodic paralysis is not out of the question, but this tends to start in adolescence so its unlikely.

The absence of alkalosis and hypertension steers the candidate away from thinking about the derangements of the renin-angiotensin-aldosterone axis. There is a normal anion gap metabolic acidosis present, which could be contributing to the picture. Of the possibilities, one may focus most on Type 1 and Type 2  renal tubular acidosis, seeing as Type 4 renal tubular acidosis tends to cause hyperkalemia instead. Of course, in all forms of hypokalemic RTA, the renal potassium losses are significant, and one would not have such a tiny 24-hour urinary K⁺ level. And on top of that the hypokalemic forms of RTA tend to present with profound acidosis - the pH would be very abnormal and the HCO₃^- would be very low.

College Answer

a) Describe the acid-base status.
Raised anion gap metabolic acidosis with appropriate respiratory compensation

b) Does he need continuation of insulin therapy over the next 6 hours? Give your reasoning.

The patient will require insulin therapy for the next few hours as the anion gap is raised, indicating ongoing ketoacidosis.

Discussion

Let us dissect these results systematically.
 

1. The A-a gradient is normal.
   PAO2 = (0.21 × 713) - (30 × 1.25) = 112.2
   Thus, A-a = ( 112.2 - 100) = 12.2mmHg.
2. There is acidaemia
3. The PaCO2 is compensatory
4. The SBE is -9.9, suggesting a metabolic acidosis
5. The respiratory compensation is adequate - the expected PaCO2(15.2 × 1.5) + 8 = 30.8mmHg
6. The anion gap is (136) - (105 + 15) = 16, or 19.5 when calculated with potassium
   The delta ratio, assuming a normal anion gap is 12 and a normal bicarbonate is 24, would therefore be (16 - 12) / (24 - 15) = 0.44. This delta ratio suggests that there is a mixed high anion gap and normal anion gap metabolic acidosis

This questions would seem identical to Question 7.1 from the first paper of 2009, but the examiners have changed the chloride and glucose. In 2009, this 16 year old patient has a normal anion gap and slightly raised glucose, forcing us to conclude that the ketoacidosis has resolved. No mention is made of insulin.

This time, the ketoacidosis persists. One would be tempted to continue the insulin/dextrose infusion until the anion gap is normal, and only the NAGMA remains.

College Answer

Drugs:

1) RTA 1 or 2
2) Ampho B,
3) Acetazolamide
4) GI losses

Discussion

Let us dissect this set of results.

There is some sort of acidosis. There is no SBE, but the bicarbonate is 18. The anion gap is (143) - (117 + 18) = 8, or 10.6 when calculated with potassium

Thus, it is a normal anion gap metabolic acidosis.

What could possibly cause a normal anion gap metabolic acidosis?

In the absence of history, one falls back on broad differentials:

• Pancreatic secretion loss
• Acetazolamide, or renal tubular acidosis Type 2
• Normal saline intoxication
• Diarrhoea
• Aldosterone antagonists
• Renal tubular acidosis Type 1 (distal)
• Ureteric diversion
• Small bowel fistula
• Hyperalimentation (TPN)

Those offered by the college are frankly bizarre. I cannot imagine how a candidate could have scored full marks with that. Acetazolamide use is a pharmacological form of Type 2 RTA, and "Ampho B" probably refers to amphotericin-induced distal (Type 2) renal tubular acidosis (and I cannot imagine how even the most time-constrained examiner might have been so rushed that they actually failed to finish the word "amphotericin" in their model answer).

Thus, the three non-overlapping causes I would have given would be the following:

• Normal saline intoxication
• Renal tubular acidosis (Type 1 or 2) - not 4, as the K⁺ is low
• Diarrhoea or other gastrointestinal losses

College Answer

Evaluation of causes of metabolic alkalosis requires a systematic approach involving history, examination and some specific investigations. Categories of aetiology include
• loss of hydrogen ions (gastrointestinal, renal)

• intracellular shift of hydrogen ions

• administration of alkali

• contraction alkalosis

•    History and examination will reveal, documented fluid losses

-(vomiting & gastric losses, laxative induced diarrhoea),

-signs of volume depletion (loss of bicarbonate free fluids),

- administered drugs (mineralocorticoids, diuretics , and antacids in renal failure) , alkali (bicarbonate, lactate, citrate etc)

• and recent hypercapnia.

• Blood investigations may reveal hypokalemia (with hydrogen shifting into cells), hypochloremia

•    Urinary findings may include excessive potassium excretion (reabsorbing hydrogen), alkaline pH (increased bicarbonate) and inappropriately elevated chloride excretion (diuretic therapy, hypokalaemia).

•          Using Stewart’s physicochemical approach, an isolated increase in Strong ion difference (SID) seen with the use of solutions such as plasmalyte or NaHCO3 or a reduction in ATOT seen with hypoalbuminemia can lead to metabolic alkalosis

Discussion

The diagnostic approach to metabolic alkalosis is discussed in more detail elsewhere.

Instead of a text excess, I will reproduce the diagnostic flowchart which was suggested by a Medscape article, and which seemed like a nice way of remembering this process.

This flowchart can be converted into the form of a point-form answer:

• History and examination
  • Recent exogenous bicarbonate administration
  • Milk-alkali syndrome
  • Hypercalcemia
  • β-lactam use
  • Cystic fibrosis
• Low urinary chloride
  • Recent diuretic therapy
  • Gastric losses via NG suction or vomiting,
  • Villous adenoma of colon
  • Post-hypercapneic (compensatory) alkalosis
• High urinary chloride and normal blood pressure
  • Ongoing diuretic therapy (must be effective, the blood pressure is normal...)
  • Bartter syndrome
  • Gitelman’s syndrome
  • Hypokalemia
  • Hypomagnesaemia
• High urinary chloride, hypertension and high renin activity
  • Ongoing diuretic therapy
  • Renal artery stenosis
  • Renin-secreting tumour
  • Malignant hypertension
• High urinary chloride, hypertension and normal renin-aldosterone axis
  • Cushing syndrome
  • Corticosteroid use
  • 17-hydroxylase deficiency
  • Liddle’s syndrome
  • Licorice overindulgeance
• High urinary chloride, hypertension and hyperaldosteronism without high renin levels
  • Adrenal adenoma
  • Adrenal hyperplasia
  • Aldosterone synthase hyperactivity

Thus, the following is a list of the necessary details one will need to determine from the patients history, physical examination, and biochemistry values:

Background history

• History of congential adrenal hypoplasia
• History of cystic fibrosis
• History of CCF (suggesting chronic exposure to diuretics)
• History of uncontrolled hypertension (malignant hypertension or renal artery stenosis)

Recent history

• Recent antacic consumption
• Recent use of calcium supplements
• β-lactam antibiotic use
• Massive abuse of licorice
• History of diarrhoea (villous adenoma) or vomiting (chloride loss)
• History of recent hypercapneic respiratory failure

Examination

• Clinically, findings consistent with severe hypertension (eg. retinal changes)
• Renal artery stenosis bruit
• Peripheral oedema (suggesting chronic exposure to diuretics)

Biochemistry

• Serum potassium
• Serum magnesium
• Urinary chloride
• Serum renin levels
• Serum aldosterone levels

College Answer

a. Describe the acid-base abnormality

Respiratory and metabolic alkalosis

b. List 3 likely causes of each acid-base abnormality in this patient. Respiratory alkalosis: Hyperventilation – spontaneous or IPPV induced, septic
encephalopathy, pneumonia
Metabolic alkalosis: Diuretics, volume contraction, upper GI losses, steroids.

Discussion

Let us dissect these results systematically.

1. The A-a gradient is normal.
   PAO2 = (0.4 × 713) - (28 × 1.25) = 250.2
   Thus, A-a = ( 250.2 - 189) = 61.2mmHg.
2. There is alkalaemia
3. The PaCO2 is contributing to the alkalaemia
4. The SBE is not supplied, but the bicarbonate is 29, suggesting that there is a metabolic alkalosis.
5. If the metabolic alkalosis is the primary disorder, then there is no respiratory compensation- the expected PaCO₂(29 × 0.7) + 20 = 40.3mmHg. Alternatively, if we assume that acute respiratory alkalosis is the primary disorder, then the expected bicarbonate is 26.4mmol/L: 
   (measured bicarbonate + 2mmol/L per every 10mmHg decrease in PaCO₂). Thus, there are two coexisting disorders - a metabolic alkalosis and a respiratory alkalosis - irrespective of how you look at it.

What the hell is driving this derangement in this septic patient?

The disturbances are not connected. Each likely has a separate cause.

The respiratory alkalosis can be driven by spontaneous hyperventilation, or by excessive mechanical ventilation (i.e. somebody has increased the pressure support level to a point where the patient is generating unnecessarily massive tidal volumes). Alternative explanations include pain, encephalopathy (of whatever cause) and pneumonia.

The  causes of metabolic alkalosis are numerous; those relevant to this question can include the following:

• Gastric losses by vomiting or drainage
• Diuretics: loop diuretics or thiazides
• Diarrhoea 
• Posthypercapneic state (hypercapnea recently reversed by mechanical ventilation)
• Recent use of corticosteroids
• Hypoalbuminaemia

College Answer

3.2a. The initial acid-base disturbance is a mixed metabolic and respiratory acidosis with a raised anion gap.

3.2b. Seizures
Resuscitated cardiac arrest
Near drowning
Near hanging

Discussion

This question is identical to Question 15.2 from the second paper of 2009.

College Answer

3.3
a) Diagnoses: 3 ischaemic bowel, 2 metformin induced lactic acidosis, thiamine deficiency, pancreatitis.

b) Two of the following investigations: Diagnostic laparoscopy or laparotomy, CT abdomen, red cell transketolase, lipase

Discussion

This question is identical to Question 15.3 from the second paper of 2009.

College Answer

pH  –                glass electrode,   The specimen is put in a capillary tube surrounded by buffer solution. The tube is made of pH sensitive glass across which a potential difference is generated,           Measures potential difference across the electrodes

PCO2                    modified glass electrode , comprises a glass pH electrode which is in contact with a thin film of NaHCO3 solution. This is separated from the specimen by a membrane that is permeable to CO2 – CO2 diffuses from specimen into the HCO3 solution where it dissociates with a change in pH which is measured by the electrode. Measures potential difference across the electrodes

PO2 –               Clark electrode or polarographic electrode,   polarographic or measures current generated (an amperometric system) or the current flow across the Clark electrode is determined by the PO2 of the solution

Discussion

The scientific principles behind the measurements which magically occur behind the shiny glass of the ABG machine are discussed in greater detail elsewhere. 

In brief:

• pH is measured by a glass electrode
  • The potential difference across the electrode is proportional to the pH difference, and this can be measured.
• PaCO₂ is measured by a modified glass electrode
  • The electrode contains some sodium bicarbonate, which reacts with the CO₂; the reaction changes the pH in the electrode, which corresponds to a change in potential difference, and this is measured. The CO₂ is then inferred from the change in pH.
• PO₂ is measured by a Clark electrode or polarographic electrode
  • O₂ in an aqueous buffer is reduced to OH^- ions with the application of current (600-800mV); this causes a current to flow between two submerged electrodes. Increasing the voltage across this system also increases the current - up to a plateau. The plateau level depends upon, and is proportional to, the concentration of oxygen.

College Answer

a) Describe the acid base abnormality

This is a normal anion gap metabolic acidosis.

b) List three causes of the acid-base  abnormality.

The major causes are:
o   Loss of bicarbonate (eg diarrhoea, pancreatic biliary drainage and urinary diversions (ureterosigmoidostomy)
o   Renal tubular acidosis (RTA).
o   Other: saline loading, TPN and cholestyramine use. 

c)  What is the most likely cause of the abnormality in this patient?

Type 4 RTA.

Discussion

Let us dissect these results systematically.

1. The A-a gradient cannot be calculated, as no FiO₂ is supplied to us
2. There is acidaemia
3. The PaCO₂ is compensatory
4. The SBE is not supplied, but the bicarbonate is 10, suggesting a severe metabolic acidosis
5. The respiratory compensation is adequate and lurks around the natural limits of human respiratory compensation for metabolic acidosis- the expected PaCO₂(10 × 1.5) + 8 = 23mmHg
6. The anion gap is (135) - (120 + 10) = 5, or 10.7 when calculated with potassium.
7. Urinary pH and electrolytes are not supplied, but they would be interesting.

Three causes of normal anion gap metabolic acidosis?

The mnemonic PANDA RUSH comes to mind, even though it is not very good.

• Pancreatico-duodenal fistula
• Acetazolamide
• Normal saline intoxication (or any other sort of exogenous chloride excess)
• Diarrhoea
• Aldosterone antagonists
• Renal tubular acidosis Type 1 (distal)
• Ureteric diversion
• Small bowel fistula
• Hyperalimentation (TPN)

In the question text, the college harps on about the raised potassium. Truly, there is only one sort of normal anion gap acidosis which is stereotypically associated with hyperkalemia:  Type 4  renal tubular acidosis, or the use of spironolactone. Of all the other causes, none have any firm commitment to hyperkalemia.

College Answer

Q7.1a) Normal anion gap metabolic acidosis with appropriate respiratory compensation.

Q7.1b) Yes The anion gap is normal, indicating resolution of ketoacidosis. The persistent acidosis reflects saline fluid replacement coupled with the chloride retention during the period of ketonuria.

Discussion

Let us dissect these results systematically.
 

1. The A-a gradient is normal.
   PAO2 = (0.21 × 713) - (30 × 1.25) = 112.2
   Thus, A-a = ( 112.2 - 100) = 12.2mmHg.
2. There is acidaemia
3. The PaCO2 is compensatory
4. The SBE is -9.9, suggesting a metabolic acidosis
5. The respiratory compensation is adequate - the expected PaCO2(15 × 1.5) + 8 = 30.5mmHg
6. The anion gap is (135) - (114 + 15) = 6, or 9.5 when calculated with potassium. 

Has the ketoacidosis resolved? Well, there is no more anion gap, so yes- one would be forced to conclude that it has. A normal anion gap acidosis now prevails, likely due to either vigorous resuscitation with normal saline, or to the normal hyperchloraemia in the recovery phase of ketoacidosis.

College Answer

Q7.2a) Metabolic alkalosis and respiratory acidosis.
Q7.2b) Cease narcotics; Naloxone (cautious); Reduce FiO2 and titrate to SpO2 90-95%; Reverse metabolic alkalosis (acetazolamide, KCl if hypokalaemia).

Discussion

Let us dissect these results systematically.
 

1. The A-a gradient is almost normal.
   PAO2 = (0.4 × 713) - (65 × 1.25) = 203.95
   Thus, A-a = ( 203.95 - 170) = 33.95mmHg.
2. There is no acidaemia
3. The PaCO2 is raised, suggesting a respiratory acidosis. The history suggests that this is probably the primary disorder.
4. The SBE is 20.9, suggesting a metabolic alkalosis
5. The metabolic compensation is excessive. The PaCO₂ increase is (65 - 40)/10 = 2.5; if this were an acute respiratory acidosis we would expect a HCO₃^- around 26.5 mmol/L (bicarbonate + 2.5 × 1), and if it were chronic we would expect 34 mmol/L (bicarbonate + 2.5 × 4). There must also be a metabolic alkalosis. 
   If the metabolic alkalosis were thought to be the primary disorder, the respiratory compensation would appear to be excessive: the expected PaCO₂ (0.7 × 46.6) + 20 = 52.6mmHg. Thus, any way you look at it, there is both a metabolic alkalosis and a respiratory acidosis.

So. This 80-year-old has enjoyed a nice dose of perioperative opiates, and is now drowsy. The metabolic alkalosis is preventing acidaemia from developing, and the CO₂ climbs ever higher, fogging up the level of consciousness. This is not helped by the generous oxygenation, which has abolished the normal contribution of hypoxic respiratory drive. Imagine: if this patient was on room air, the alveolar gas equation yields a PAO₂ of only 68mmHg; and with his elderly lungs in charge of gas exchange the PaO2 would be in the 50s, driving the respiratory effort.

Thus, there are various ways of dealing with this acid-base disturbance.

• Stop the opiates.
• Stop the oxygen. Go with room air, or whatever is required to maintain an SpO₂ of 88-90% (the college generously recommends 90-95%)
• Reverse the opiates with naloxone
• Stop the diuretics
• Administer sodium chloride, potassium chloride or ammonium chloride.
• Administer acetazolamide

College Answer

Q7.3a) Acute respiratory acidosis following previous compensated respiratory alkalosis of pregnancy. At 38 weeks pregnancy the normal PaCO2 is <30 mm Hg with compensatory HCO3- reduction. The blood gases therefore indicate acute CO2 retention, probably due to pain and narcotics.

Q7.3b) > 250 mm Hg.

Q7.3c) Potential explanations include loss of FRC post abdominal surgery, segmental collapse or consolidation, aspiration, pulmonary oedema.

Discussion

Let us dissect these results systematically.

1. The A-a gradient is high:
   PAO2 = (0.5 × 713) - (42 × 1.25) = 304
   Thus, A-a = ( 304-150) = 104mmHg.
2. There is acidaemia
3. The PaCO2 is contributing to the acidosis
4. The SBE is -4.9, suggesting a metabolic acidosis
5. The respiratory compensation would seem inadequate - the expected PaCO₂(20.5 × 1.5) + 8 = 38.75mmHg; given that at the end of pregnancy the normal CO₂ is around 30mHg, this would represent a respiratory acidosis.

However, one should point out that the empirical rules for blood gas interpretation were developed from data collected in non-pregnant individuals, and do not account for the changes in homeostatic setpoints of pregnancy. Indeed, if we were to consider that the pregnant woman has well-compensated respiratory alkalosis at term (with a chronically decreased PaCO₂, down to 30mmHg) we would expect the HCO₃^- to decrease by 5mmol/L, which would give us an expected HCO₃^- of 19mmol/L.

If this pregnant woman with a HCO₃^- of 19mmol/L were to suddenly develop a respiratory acidosis, the HCO₃^- would increase by 1mol/L for every 10mmHg of PaCO₂ elevation. The expected HCO₃^- in this woman would therefore be slightly over 20mmol/L; ... which it is. Thus, there is probably no metabolic acid-base disturbance here.

The normal arterial oxygen for this woman should be well over 270, especilly if she had a normal PaCO₂.

Why is the A-a gradient so great?

Apart from the guesses offered by the college ("loss of FRC post abdominal surgery, segmental collapse or consolidation, aspiration, pulmonary oedema") one can also suggest PE, amniotic fluid embolism and cardiomyopathy of pregnancy. In addition, merely being supine can cause modest hypoxia in pregnant women at term, though this is a pre-partum phenomenon.

College Answer

Drug                                            Anion
Ethanol                                   - Lactate
Methanol                                 - Formate or formic acid
Ethylene glycol                       - Glycolate / oxalate

Discussion

Ethanol overindulgeance can indeed cause a lactic acidosis, but I would have chosen a different drug:

• Toxicological causes of high anion and high osmolar gap
  • Methanol intoxication (the anion is formic acid)
  • Ethylene glycol intoxication (the anions are glycolic acid and oxalic acid)
  • Diethylene glycol intoxication (the anion is 2-hydroxyethoxyacetic acid, HEAA)
  • Propylene glycol intoxication (the anions are pyruvate, lactate and acetate)
  • Salicylate intoxication (the anions are salicylate and lactate)
  • Any toxin causing massive lactic acidosis, eg. isoniazid
• Endocrine and metabolic causes of high anion and high osmolar gap
  • Lactic acidosis
  • Alcoholic or diabetic ketoacidosis
  • Acute kidney injury

Erstad, Brian L. "Osmolality and osmolarity: narrowing the terminology gap."Pharmacotherapy: The Journal of Human Pharmacology and Drug Therapy23.9 (2003): 1085-1086.

Gennari, F. John. "Current concepts. Serum osmolality. Uses and limitations."The New England journal of medicine 310.2 (1984): 102-105.

Hoffman, Robert S., et al. "Osmol gaps revisited: normal values and limitations."Clinical Toxicology 31.1 (1993): 81-93.

Dorwart, William V., and Leslie Chalmers. "Comparison of methods for calculating serum osmolality from chemical concentrations, and the prognostic value of such calculations." Clinical chemistry 21.2 (1975): 190-194.

Kraut, Jeffrey A., and Shelly Xiaolei Xing. "Approach to the evaluation of a patient with an increased serum osmolal gap and high-anion-gap metabolic acidosis." American Journal of Kidney Diseases 58.3 (2011): 480-484.

College Answer

13.1     Summarise the findings of the blood tests.

•    High anion gap metabolic acidosis (with apparent normal SID). Note AG 33 which is NOT adequately explained just by a lactate of 13 mmol
•    Inadequate or inappropriate respiratory compensation
•    Hypoxaemia (P/F ratio 70)
•    Acute renal failure (note urea:creatinine ratio).
•    Hyperkalaemia

13.2     What are the likely underlying causes of the lactic acidosis?

•    Sepsis with shock
•    Ongoing hypovolaemia
•    Hypoperfusion eg septic cardiomyopathy; abdominal compartment syndrome
•    Possible gut ischemia
•    Perhaps adrenaline (also seen with other catecholamines – unpredictable

13.3     What are your management priorities at this point?

•    Optimise cardiovascular function. Urgent echocardiogram. Volume replacement if possible. Measure continuous cardiac output (PiCCO or PAC). Measure SvO2 or ScvO2. Exclude abdominal compartment syndrome.

•    Optimise ventilation. Exclude pneumothorax. Probably needs more PEEP after some volume. Minimise airway prtessures, limit tidal volume, tolerate hypercarbia (though concerned about pH < 7!!!)

•    Rationalise inotropes. Stop adrenaline, use noradrenaline as required

•    Emergency management of hyperkalaemia with calcium, bicarbonate, insulin, dextrose and then haemodialysis!

•    Urgent CRRT – for both potassium and acidosis use of hemosol buffer

•    Broad spectrum IV antibiotics (rational answer required)

Discussion

Analysis of the biochemistry:

• Life-threatening hyperkalemia
• Acute renal failure

Analysis of the haematology:

• Thrombocytopenia

Analysis of the ABG:

• Severe hypoxia
• High anion gap metabolic acidosis with failure of respiratory compensation (pCO₂ should be about 25mmHg) - thus, a respiratory acidosis is also present.
• The anion gap is (138) - (104 + 11) = 23, or 30.1 when calculated with potassium. It is not clear where the college got their value of 33 from.
• The delta ratio, assuming a normal anion gap is 12 and a normal bicarbonate is 24, would therefore be (23 - 12) / (24 - 11) = 0.84, giving you the impression that there is some non-anion gap acidosis mixed in with the HAGMA. 
   
• The lactate is 13.  Notably, the college forcefully points out that the lactate of 13 is insufficient to explain the entire rise in the anion gap. This could be argued. It must be remembered that the lactate does not need to account for all of the calculated 23 mmol/L of the anion gap.  Depending on the laboratory and the anion gap equation which you decided to use (sans or avec potassium), your acceptable standard anion gap could be 12 or 16, and this scenario's anion gap could be 23 or 30 (...or 33?). Ergo, the anion gap has risen by some value in the low teens (let's say 10 to 14). Therefore, the lactate accounts for almost all of the anion gap increase, with only maybe 1 mmol/L remaining unaccounted for. Given that the college do not follow on with any additional suggestions as to the origin of these mysterious anions, it is unclear where they were going with their "NOT adequately explained" statement, or how it factored into their marking rubric.

Causes of the lactic acidosis in this case:

• Increased production:
  • Hypovolemia and poor tissue perfusion
  • Cardiac injury and cardiogenic shock
  • Sepsis
  • Hypoxia
  • Use of adrenaline
  • Massive rhabdomyolysis
  • Gut ischaemia and abdominal compartment syndrome
  • Toxicity due to mitochondrial toxins
• Decreased clearance
  • Hepatic injury, ischaemic hepatitis or pre-existing hepatic disease
  • Failure of renal clearance (with lactate above 10mmol/L, renal clearance begins to play a role)

Management priorities:

Something like this benefits from a structured approach.

Airway:

• Exclude acute airway obstruction
• Ensure the ETT is well positioned (not in the right main bronchus)
• Ensure ETT is not kinked and not semiobstructed with secretions or blood, and that it is of an appropriate diameter for this patient (CO₂ clearance could be impaired)

Breathing:

• Exclude tension pneumothorax by clinical examination
• Exclude more minor pneumothorax with CXR or bedside ultrasonography
• Increase PEEP to 12
• Increase I:E ratio to 1:1.5 or 1:1 if the EtCO₂ trace does not demonstrate an obstructive pattern (otherwise, hypercapnea could be exacerbated)

Circulation:

• Exclude cardiac tamponade with bedside TTE
• Assess fluid responsiveness by static or dynamic manoeuvres; administer volume
• Assess cardiac output by combination of physical examination, measured variables (PAC or PiCCO) and surrogate measures (ScvO₂, A-V CO₂ difference)
• Wean β-2 agonist medications to minimise pharmacological causes of lactate excess. Move on to pure α-1 agonists in pursuit of a normal MAP.

Disability/neurology is not a matter of priority at present.

Electrolyte derangement however is.

• Administer sodium bicarbonate to replenish buffer systems, correct pH to improve catecholamine efficacy and at the same time shift potassium into the intracellular compartment
• Administer calcium gluconate to stabilise excitable tissues and prevent arrhythmia
• Consider insulin and dextrose
• CVVHDF is indicated:
  • Without renal clearance, dialysis may be the only option to remove the potassium from this organism. One may consider calcium resonium resin if the GI tract is working.
  • In any case, given the history, there is likely massive rhabodomyolysis, and CVVHDF is indicated for this alone.

• The college also recommends broad spectrum antibiotics. Sepsis is not excluded, so a broad spectrum beta lactam like piperacillin/tazobactam would be a sensible choice.

College Answer

a)          Describe the initial acid-base disturbance.

The initial acid-base disturbance is a mixed metabolic and respiratory acidosis with a raised anion gap.

b)         List 3 clinical scenarios which may produce such a pattern of arterial blood gas derangement?

Seizures
Resuscitated cardiac arrest
Near drowning
Near hanging

Discussion

Let us dissect these initial results systematically.
 

1. The A-a gradient cannot be calculated - the FiO₂ is not supplied
2. There is acidaemia
3. The PaCO2 is vaguely compensatory
4. The SBE is not supplied, but the bicarbonate is 9 , suggesting a severe metabolic acidosis
5. The respiratory compensation is inadequate - the expected PaCO2(9 × 1.5) + 8 = 21.5mmHg, and thus there is also a respiratory acidosis
6. The anion gap is raised: (137) - (105 + 9) = 23, or 27.2 when calculated with potassium. The delta ratio, assuming a normal anion gap is 12 and a normal bicarbonate is 24, would therefore be (23 - 12) / (24 - 9) = 0.73, suggesting that there is a mixed high and normal anion gap metabolic acidosis. 

In the subsequent gas, the acidosis is almost completely corrected- and it only took one hour.

What could cause this sort of picture?

Well.

Whatever the extra anion was, it has either been buffered well, or it has been eliminated in some way, or it has been metabolised.

Thus, the following things may have happened:

• The patient received some IV bicarbonate
• The anion was lactate and virtually all of it has been metabolised; that means the cause was reversible, and has been reversed.
  • Circulatory failure (eg. cardiac arrest, tension pneumothoroax, cardiac tamponade)
  • Hypoxia (asphyxiation of some sort, be it drowning, hanging, etc)
  • Seizures

College Answer

a) Give the three most likely diagnoses.

Diagnoses: 3 ischaemic bowel, 2 metformin induced lactic acidosis, thiamine deficiency, pancreatitis

b) List two additional investigations that you would perform based on the above information.

Two of the following investigations: Diagnostic laparoscopy or laparotomy, CT
abdomen, red cell transketolase, lipase

Discussion

This question is identical to Question 21.2 from the second paper of 2009, and Question 26.4 from the second paper of 2013 . The answer to the latter contains a comprehensive discussion.

College Answer

24.1.   List the acid-base abnormalities.

•    High anion gap metabolic acidosis with raised lactate
•    Metabolic alkalosis ( Delta BE < Delta AG)
•    Respiratory alkalosis

24.2.   What are the causes of elevated plasma lactate in sepsis?

•    Circulatory failure due to hypotension and hypoxia
•    Microvascular shunting and mitochondrial failure (cytopathic hypoxia)
•    Use of adrenaline as an inotrope
•    Inhibition of pyruvate dehydrogenase (PDH) by endotoxin.

24.3.   Name three (3) drugs (each from a different class of drugs) which result in plasma hyperlactaemia.

•          Catecholamines
•         Metformin / Phenformin
•          Alcohols
•         Cyanide, nitroprusside
•          Salicylates
•         Lactate containing solutions – HRL, dialysates

24.4.   List two (2) inborn errors of metabolism associated with lactic acidosis.

•  Glucose 6 phosphatase deficiency
•  Fructose 1,6 diphosphatase deficiency
•  Pyruvate carboxylase deficiency
•  Deficiency of enzymes of oxidative phsophorylation

Discussion

The actual ABG itself and the attached patient history are identical to Question 22.2 from the second paper of 2011, and Question 18 from the first paper of 2007.

Generally, this question is frequently repeated, though the exact wording changes.
Question 6.4 from the first paper of 2013 contains a detailed discussion of the answer.

Let us dissect these results systematically.

1. Yes, the A-a gradient is raised: 
   (0.3 x 713) - 28.8 = 185.1, thus the A-a difference is 195.5-107 = 78.1
2. There is no acidaemia.
3. The PaCO2 is compensatory
4. The SBE is -8.6, suggesting a severe metabolic acidosis
5. The respiratory compensation is excessive- the expected PaCO2(15 × 1.5) + 8 = 30.3mmHg, and thus there is also a respiratory alkalosis.
6. The anion gap is raised:
    (147) - (95  + 15) = 37, or 43.7 when calculated with potassium
   The delta ratio, assuming a normal anion gap is 12 and a normal bicarbonate is 24, would therefore be (37 - 12) / (24 - 15) = 2.77
   This delta ratio suggests that there is a high anion gap metabolic acidosis coexisting with a metabolic alkalosis.
    

24.1. List the acid-base abnormalities.

• Respiratory alkalosis
• High anion gap metabolic acidosis
• Metabolic alkalosis

24.2. What are the causes of elevated plasma lactate in sepsis?

Copying directly from Question 22.2:

• Tissue hypoperfusion and hypoxia
• Use of adrenaline (increased glycolytic flux)
• Down regulation of pyruvate dehydrogenase by inflammatory mediators
• Underlying ischaemic tissue
• Microvascular shunting
• Sepsis-associated cardiomyopathy

It is interesting to note that though these two questions are identical, the answers are not.

24.3. Name three (3) drugs (each from a different class of drugs) which result in plasma hyperlactaemia.

• Metformin
• Isoniazid
• Iron
• Toxic alcohols
• Isoniazid
• Cyanide
• Salicylates
• Catecholamines

24.4. List two (2) inborn errors of metabolism associated with lactic acidosis.

These are discussed in (slightly) greated detail in the chapter on the inherited defects of lactate metabolism. The full list is extensive; the college asks for only two.

The sleep-deprived candidate is likely to remember only glucose-6-phosphatase deficiency. "Deficiency of enzymes of oxidative phsophorylation" is a good all-encompassing statement which is general but still accurate.

College Answer

a)  Outline the possible causes of his metabolic acidosis.

High anion gap metabolic acidosis

Lactic acidosis 
•    Severe dehydration from osmotic diuresis from hyperglycaemia.
•    Ischaemic bowel
•     Sepsis
•    Metformin (if known type 2 diabetes)

Keto-acidosis 
•  Diabetic (unlikely to be first presentation at 56 yrs with type I diabetes)
•  Alcoholic keto-acidosis (possible although sugar too high)
•  Starvation keto-acidosis (possible although sugar too high)

Toxins

Salicyates, methanol, ethylene glycol,

Renal Failure

b)  What is the corrected serum sodium?

Corrected Sodium= Measured sodium + glucose/4 (Credit was given to people using a range of different formulae) approximately =149 mmol/L.

c)  Outline your approach to the correction of his metabolic abnormalities.

Fluid replacement:
Generally Normal Saline/Hartmanns solution, even if hypernatraemia present to expand ECF volume rapidly.

After initial restoration of normotension, free water replacement .
Fluid titrated to clinical status- organ perfusion, filling pressures, etc. Elderly patients with likely co-existent cardiac disease need more careful monitoring/less aggressive replacement

•     Insulin
-       Blood glucose can fall rapidly when urine output re-established and dehydration corrected. High risk of hypoglycaemia.
-      Treatment with insulin to decrease the serum osmolality by no greater than
2mosmol/kg/hr and blood glucose by no more than 3-4mmol/hr

•     Electrolytes potassium supplements required. Attention to magnesium and phosphate supplementation also required.

•     Correction of underlying cause (in this case, surgery for bowel obstruction)

d)             List the possible complications of this condition and its treatment.

• Cerebral oedema with rapid correction
• Pre-renal azotaemia & renal failure
•  Shock 
• Hypercoagulable state-  thromboembolic complications- DVT, stroke, AMI
• Fluid overload/congestive  cardiac failure with correction
• Metabolic and electrolyte abnormalities- hypokalaemia,
• hypomagnesaemia,  hypophosphataemia,  hypoglycaemia, hyperchloraemic
(non-anion gap) acidosis (normal saline therapy)
• Infections and sepsis

Discussion

In this question, the candidates are presented with a picture closely resembling a HONK.

The proper calculation of the anion gap is fuddled by the presence of vast quantities of glucose, and thus an inaccurate reported sodium.

Unadjusted, the biochemistry suggests that the anion gap may be significantly raised:

(137 + 4) - (98 + 15) = 28.

However, if we adjust the sodium, we find that it is even worse.

(149 + 4) - (98 + 15) = 40

Thus, the candidate ought to remember that up-adjusting the sodium for hyperosmolar states will only make the anion gap wider.

The delta ratio, calculated from the corrected and uncorrected anion gaps, is either 1.7 or 3.1

(28 - 12) / ( 24 - 15) or (40 - 12) / (24 - 15)

Thus, there is a high anion gap metabolic acidosis which coexists with a metabolic alkalosis.

a) Outline the possible causes of his metabolic acidosis.

The list of differentials provided by the college is comprehensive.

One may divide it into a MUDPILES mnemonic, or organise it by unmeasured anions:

• Lactate:
  • Hypovolemic shock
  • Sepsis
  • Cardiogenic shock
  • Mitochondial toxicity
  • Thiamine deficiency
  • Biguanide poisoning
• Toxic metabolites
  • Methanol
  • Ethylene glycol
• Ketoacids
  • Diabetic ketoacidosis
  • Alcoholic ketoacidosis
  • Starvation ketoacidosis
• Unmeasured non-volatile metabolic acids
  • Renal failure

b) What is the corrected serum sodium?

The college gave the candidates credit for mentioning a number of formulae.

There are indeed several different ways of calculating the corrected sodium.

They all follow the same trend: Corrected Na⁺ = measured Na⁺ + (correction factor)

The correction factor is a matter of debate, and there are several variations:

• Glucose in mmol/L divided by 4
• (Glucose in mg/dL - 100) multiplied by 0.016
• (Glucose in mg/dL - 100) multiplied by 0.024

c) Outline your approach to the correction of his metabolic abnormalities.

One would not need to approach this in an algorithmic fashion. It is a question specifically about the metabolic abnormalities, and so one does not need to go on at lengths about securing the airway.

Thus:

• Correct fluid deficit with fluid resuscitation
• Correct hyperglycaemia with insulin
  • careful monitioring for hypoglycaemia (the college points out that once the renal function begins to recover, urinary losses of glucose will be rapid)
• Correct electrolytes by careful monitoring and replacement
• Correct underlying disorder (bowel obstruction)

d) List the possible complications of this condition and its treatment.

Complications of HONK are discussed in greater detail elsewhere.

In brief, the major ones are listed below:

• Cardiac arrest
• Cardiovascular collapse
• Myocardial infarction
• Stroke
• Cerebral oedema and brain injury
• Venous thrombosis

In addition to these, the college has thrown in infection/sepsis as a complication of bowel obstruction, as well as complication of HONK management (electrolyte bewilderment and fluid overload).

College Answer

•    Raised osmolar gap with raised AG
•     Methanol
•    Ethylene glycol
•     Ethanol
•    Lactic acidosis can lead to a raised OG and AG, however, the osmolar gap does not reach the levels seen here.

Discussion

These biochemistry results are provided without any history.

Let us dissect them systematically.

1. The A-a gradient cannot be calculated
2. The pH is not supplied
3. The PaCO₂ is not reported
4. The SBE is not offered, but the bicarbonate is 15, suggetsing a metabolic acidosis.
5. The respiratory compensation cannot be assessed.
6. The anion gap is raised:
   (138) - (98  + 15) = 25, or 29 when calculated with potassium
   The delta ratio, assuming a normal anion gap is 12 and a normal bicarbonate is 24, would therefore be (25 - 12) / (24 - 15) = 1.44
   This delta ratio suggests that there is a pure high anion gap metabolic acidosis here.
7. The osmolar gap is raised: 340 - (2 × 138 + 6 + 8) = 50

Thus, this high anion gap metabolic acidosis is also associated with a raised osmolar gap.

Most likely differentials:

• Toxic alcohol intoxication, which includes ethanol intoxication. This is probably the only commonly seen cause of such a massively raised osmolar gap associated with a high anion gap metabolic acidosis.

Less likely differentials (osmolar gap is probably a bit too high for these to be realistic differentials):

• Salicylate overdose
• Iron overdose
• Lactic acidosis
• Ketoacidosis
• Pyroglutamic acidosis

College Answer

•    Raised osmolar gap with normal AG
•     Mannitol
•     Glycine
•     Ethanol

Discussion

These biochemistry results are provided without any history.

Let us dissect them systematically.

1. The A-a gradient cannot be calculated
2. The pH is not supplied
3. The PaCO₂ is not reported
4. The SBE is not offered, but the bicarbonate is 21, suggesting a trend towards metabolic acidosis.
5. The respiratory compensation cannot be assessed.
6. The anion gap is normal:
    (125) - (98  + 21) = 6, or 10 when calculated with potassium
7. The osmolar gap is raised: 310 - (2 × 125 + 6 + 8) = 46

So, this is a hyperosmolar state without metabolic acidosis. It stands to reason that the extra osmoles are probably not participating in any sort of metabolism.

This sort of picture is consistent with the following scenarios:

• Mannitol therapy
• Glycine (post-TURP syndrome)
• Alcohol intoxication (early in the night)

Glucose and urea may contribute in other scenarios, but in this set of biochemistry results they are represented by normal values.

The possible contributors to the hyperosmolarity are discussed briefly in the chapter on hyperosmolar hypernatremia.

College Answer

a)  Comment on the acid-base status.

Mixed respiratory and metabolic alkalosis

b)  List 2 likely causes of the acid-base derangement in this patient.

Respiratory alkalosis from the hyperventilation due to the pneumonia

Metabolic alkalosis from vomiting or diuretic use

Discussion

 This question is identical to Question 6.1  from the first paper of 2013.

College Answer

a)  Could this blood gas be consistent with the definition of acute respiratory distress syndrome (ARDS)? Give your reasoning.

No. The P/F ratio is 220. By definition, the problem would be acute lung injury rather that ARDS at this stage.

b)  What dose of sodium bicarbonate (in mmol) would be required to reverse the metabolic acidosis? Show your calculation method.

Dose sodium bicarbonate = Wt (kg) x 0.3 x -SBE = 70 x 0.3 x 10 = 210 mmol

Discussion

Unlike the majority of these questions about ABGs, this one does not require indepth interpretation of the acid-base disturbance.

The first question refers to the now-extinct definition of ARDS, which is distinct from the new definition of ARDS. The issues surrounding ARDS classification are discussed in full elsewhere, and I will not rant about this extensively, except to mention the following ranges of the PaO₂/FiO₂ ratio:

• 300-200 = "mild" ARDS, ~ 27% mortality
• 200-100 = "moderate" ARDS ~32% mortality
• under 100 = "Severe" ARDS, ~ 45% mortality

This patient has an P/F ratio which is 110/0.5 = 220; thus the patient falls into a "mild" ARDS category, or "Acute Lung Injury" according to the old definition. The new ARDS definition has replaced the old as of 2012, about 1 year after the candidates were exposed to this question paper.

Now, as for the bicarbonate:

There are actually several methods of estimating the amount of bicarbonate required to titrate somebody's body fluids back to normal pH. They are all equally inaccurate. Assumptions are made about the volume of distribution of bicarbonate, which is about 50% of body weight at normal pH and about 100% at a low pH. These things are discussed in greater detail in a distant chapter where I attempted to explain to myself the physiological response to an infusion of sodium bicarbonate.

For example, the following equations give an answer as bicarbonate dose in mmol.

• body weight in kg × 1
• body weight × 0.3 × SBE
• body weight × 0.3 × (desired HCO₃^- - measured HCO₃^-)
• body weight × ( 0.4 + 2.6 / measured HCO₃^-)

College Answer

a)  Describe the acid-base status.
Normal anion gap metabolic acidosis with appropriate respiratory compensation

b)  What is the likely cause of this disturbance?
Resuscitation with large volume saline infusion.

c)  What is the underlying biochemical mechanism?
ECF dilution by fluid with strong ion difference of zero

Discussion

 This question is identical to Question 6.3  from the first paper of 2013.

College Answer

a)  List the acid-base abnormalities.
Lactic acidosis
Anion gap elevation (37 mEq/L) Metabolic alkalosis
Respiratory alkalosis

Discussion

This question is frequently repeated. A list of duplicates can be found in a chapter on the causes of lactic acidosis in sepsis. Question 6.4 from the first paper of 2013 contains a detailed discussion of the answer.

College Answer

a)  List the abnormalities  on the blood gas and give the most likely cause of each abnormality.

•    Metabolic acidosis – lactic acidosis secondary to prolonged seizures
•    Respiratory  acidosis  (or  inadequate  compensation)  –  central  hypoventilation  or inadequate mechanical ventilation
•    Increased A-a gradient  - aspiration pneumonia

Discussion

Let us dissect these results systematically.

1. The A-a gradient is high:
   PAO2 = (0.5 × 713) - (35 × 1.25) = 312.75
   Thus, A-a = ( 312.75 - 105) = 207.75mmHg.
   In fact, the PaO₂/FiO₂ ratio is (105/0.5) = 210, which puts this man's hypoxia in a "mild ARDS" category by the 2012 Berlin definition.
2. There is acidaemia
3. The PaCO2 is compensatory
4. The SBE is not offered, but the bicarbonate is low, suggesting a severe metabolic acidosis
5. The respiratory compensation is inadequate - the expected PaCO₂ = (10.3 × 1.5) + 8 = 23.5 mmHg, and thus there is also a respiratory acidosis
6. The anion gap cannot be calculated.

The college suggests the following explanations for this gas:

• Hypoxia = aspiration
• Hypercapnea = obtundation
• Metabolic acidosis = seizure-associated rise in lactate.

This explanation is certainly within reason. One could even get into a discussion about MELAS. Often in the postictal state there is still appropriate respiratory compensation for the lactic acidosis, with most postictal patients averaging a PaCO₂ around 17mmHg in a 1977 study of post-seizure lactic acidosis. However, postictal hypopnoea is a known cause of respiratory acidosis after a seizure, and may be a factor in SUDEP - Sudden Unexplained Death in Epilepsy.

To consider alternative explanations, this mixed acid-base disturbance can also be accounted for by the complications of toxic alcohol ingestion, a hypoglycaemic episode, opiate overdose, and numerous others.

College Answer

a) Interpret these results

Metabolic alkalosis with partial respiratory compensation and severe hypokalaemia

b) List 2 likely diagnoses

• Primary  Hyperaldosteronism   most  likely  secondary  to  an  aldosterone  producing adenoma (Conn’s syndrome – 50-60%) or adrenal hyperplasia (40-50%)
• Licorice ingestion
• Liddle’s syndrome
• Excessive diuretic use

c) Give 2 drugs used to treat this condition

• Aldosterone antagonist (spirinolactone or eplerenone)
• Amiloride

d) List 3 other potential causes of these biochemical abnormalities

1/ laxative abuse
2/vomiting
3/diarrhoea
3/cushings and ACTH tumors
5/ primary metabolic alkalosis

Discussion

Let us dissect this ABG according to the classical rules.

With a bicarbonate level of 40, the PCO₂ should be (0.7 × 40 + 20) - or about 48, ± 5mmHg.

The gas we get from the college gives us a PCO₂ of 46 mmHg, which is within the range of error.

I could call that a compensated disorder, but - lets be fair - its 2mmHg lower than the predicted value, so the savage ABG purist would be forced to label this as "partial" compensation.

The only other (massive) abnormality is the almost-unsurvivable K⁺ level (1.8mmol/L).

What are the causes of metabolic alkalosis, then? Particularly with such hypokalemia?

Answers b) and d) essentially ask the candidate to spew forth a torrent of differentials. Its a game of "how many causes of metabolic alkalosis can you think of in under 3 minutes". Judging by the pass rate, a fair few of us are rather good at this.

Even though this list of differentials is discussed in another chapter (Causes of metabolic alkalosis), I reproduce it here in order to simplify revision.

Lastly, management is asked about. It requires little imagination to answer "block the aldosterone excess", and a receptor blocker such as spironolactone is a convenient and reasonably safe means of doing so.

College Answer

a)  List four potential contributing causes of the metabolic derangement

•    Shock/Underesuscitation/hypovolaemia (elevated Hb and Lactate)
•    Normal (0.9%) Saline fluid resuscitation
•    Carbon monoxide poisoning
•    Cyanide toxicity from smoke inhalation (elevated anion gap acidosis)
•    Other   missed   injuries   e.g.  abdominal   trauma,   bleeding   etc  leading   to  hypo perfusion/shock
•    Potential concurrent ingestions e.g. methanol, ethylene glycol

b)  How would you classify the acid base derangement and explain your reasoning?

•    Mixed metabolic acidosis

(Note: CO2 is also high for pH but less relevant because patient on IPPV)
Delta ratio indicates a greater fall in [HCO3-] than expected given increase in AG. This can be explained  by a mixed metabolic  acidosis,  i.e. a combined high anion
gap and normal anion gap acidosis.

c) The serum albumin is 18g/L. Outline how would this affect the anion gap.

•    The   plasma   proteins   are   the   major   source   of   unmeasured   anions.   Hypo albuminemia may mask an increased concentration  of gap anions by lowering the value of the anion gap. Adjustment of the anion gap can be made by the application of correction factors (see Figge et al, CCM 1998).

d)  Whilst on your ward round the RMO asks your opinion on the Stewart approach to acid base physiology. List the 3 independent variables that comprise this approach

Strong ion difference
Partial CO2 tension
Total concentration of weak acid (ATOT)

Discussion

Logically, (b) is the question one ought to answer first, in order to answer (a).

Thus: a systematic approach:

1. The A-a gradient cannot be calculated; all that can be said is that this gentleman is not hypoxic
2. There is acidaemia
3. The PaCO₂ is compensatory (34mmHg)
4. The SBE is -16.1, suggesting a severe metabolic acidosis
5. The respiratory compensation is inadequate - the expected PaCO₂( 8 × 1.5) + 8 = 20mmHg, and thus there is also a respiratory acidosis
6. The anion gap is raised:
   (145) - (120 + 8) = 17, or 21.8 when calculated with potassium
   The delta ratio, assuming a normal anion gap is 12 and a normal bicarbonate is 24, would therefore be (17 - 12) / (24 - 8) = 0.31
   This delta ratio suggests that there is a mixed high anion gap and normal anion gap metabolic acidosis here.

 If we adjust the anion gap to compensate for the low albumin (which is discussed later), the "normal" anion gap value ends up being about 6.5, which means the delta ratio is in fact 0.65. 

Now, for question (a): the following are potential causes of a mixed metabolic acidosis in a severe burns patient:

• Shock due to hypovolemia (hemoconcentration is apparent from the high Hb level)
• Shock due to sepsis
• High lactate due to carbon monoxide poisoning
• High lactate due to cyanide poisoning

c) Albumin is a negatively charged protein, and thus it contributes to the anion gap in solution. Indeed for every 4 grams per litre of albumin, the "normal" anion gap value changes by 1. Thus, if we assume that the "normal" albumin value is about 40, this patient has an "albumin deficit" of 22g/L, and the normal value for anion gap should actually be 6.5 (12 - 22/4).

d) The college asks the candidate to list the three independent variables of Stewart's interpretation of acid-base physiology.

• Strong ion difference
• Partial CO2 tension
• Total concentration of weak acid (ATOT)

The Stewart approach can be found heavily marketed at acid-base.org. The original paper, describing the dependent and independent variables, can be found here but is available as an abstract only. Fortunately, the acid-base.org people have published Stewart's original book online, for free.

College Answer

a)
Severe compensated metabolic acidosis with a raised anion gap (~20), normal osmolar gap and delta ratio 0.4 (Some candidates might say that there is both a high AG and normal AG acidosis rather than stating the delta ratio and that is also correct).

b)
Renal failure

Pyroglutamic acidosis (renal and liver dysfunction and possible flucloxacillin and paracetamol exposure) 

Sepsis

Metformin related lactic acidosis
(delta ratio suggests mixed AG and NAG MA or renal failure)

Discussion

How did they arrive at these answers?

Normal or slightly raised osmolar gap     

= 300 - ((1.86 x 139) + 15.4 + 5.1 + 9) = 12.

High anion gap                                            
The anion gap is (139) - (115 + 4) = 20, or 24.4 when calculated with potassium
 

Delta ratio = 0.4    
 The delta ratio, assuming a normal anion gap is 12 and a normal bicarbonate is 24, is
(20 - 12) / (24 - 4) = 0.4                                      
 

Thus, this is a high anion gap metabolic acidosis with hyperchloraemia, with normal osmolar gap, and incomplete respiratory compensation.

The MSSA mentioned in question 3.1 is a hint at flucloxacillin exposure, to make you think of pyroglutamic acidosis. Having never heard of pyroglutamic acidosis, I have found this article to explain it to myself. There is also another article here. A local summary of pyroglutamic acidosis is also available.

Pyroglutamic acidosis occurs due to glutathione depletion in patients who receive flucloxacillin or vigabatrin  together with paracetamol. The key feature is loss of feedback inhibition because of glutathione depletion, which results in overproduction of pyroglutamic acid.

College Answer

a)
Metabolic alkalosis
A-a DO2 295

b)
Resolution of respiratory acidosis with delayed correction of metabolic compensation
Diuretic therapy

Discussion

Widened A-a gradient? yes there is; by the standard equation,

(713 x 0.6) – (48 x 1.25) – (79) =288.8

On FiO2 of 60%, the PaO₂ should be about 378.

There is also a metabolic alkalosis with normal chloride and normal potassium. The hint is that this patient has been ventilated for 3 weeks. ARDS ventilation typically involves "permissive hypercapnea", which leads to a gradual renal compensation, with the retention of bicarbonate. As the respiratory acidosis resolves, an alkalaemia develops because the increase in the rate of bicarbonate excretion is delayed.

College Answer

a) Normal anion gap severe metabolic acidosis with incomplete compensation

b)Renal tubular acidosis Type 1 distal secondary to chronic toluene abuse

Discussion

How did they reach these conclusions?

Let us approach this systematically.

1. The A-a gradient is slightly raised:
   PAO₂ = (0.4 × 713) - (17 × 1.25) = 263.95
   Thus, A-a = ( 263.95 - 140) = 143.95mmHg.
2. There is severe acidaemia
3. The PaCO₂ is low, suggesting an attempt at compensation
4. The SBE is -28, suggesting a severe metabolic acidosis
5. The respiratory compensation is either complete, or there is a mild respiratory acidosis:
   Boston rules: the expected CO₂ is (1.5 × 4) + 8 = 14mmHg (plus-minus 3)
   Copenhagen rules: the expected CO₂ is (40-SBE) = 12
6. The anion gap cannot be calculated in the conventional sense, but if you ignore the potassium it ends up being (127 – 113 – 4) = 10.

There is not enough information to calculate the osmolar gap.
This is a normal anion gap severe metabolic acidosis with incomplete compensation.

The solvent abuse mentioned in the question points to toluene.
Toluene toxicity alone would have caused a high anion gap metabolic acidosis (due to accumulation of hippuric acid). However, the hippurate anion is rapidly excreted via the kidneys, whereas the hydrogen ions remains. The RTA due to toluene is a type 1, which is a hyperchloremic hypokalemic acidosis due to a failure to acidify the urine in the distal tubule. Another characteristic feature is the extremely alkaline urinary pH which contrasts with the extremely acidic blood pH; such a situation tends to only occur in the case of a Type 1 (distal) RTA, whereas with the other RTAs the urinary acidification is at least somewhat preserved.

College Answer

a)
Metabolic alkalosis with respiratory compensation 

b)
Diuretic therapy 
Steroid therapy 
Vomiting from gastric outlet obstruction 
Post hypercapnoeic alkalosis

Discussion

I could add nothing more to these answers. The question plainly states there has been profuse vomiting.

Let us dissect these results systematically.

1. The A-a gradient is high:
   PAO2 = (0.4 × 713) - (47 × 1.25) = 155.05
   Thus, A-a = (155.05 - 58) = 97.05mmHg.
2. There is alkalaemia
3. The PaCO2 is compensatory
4. The SBE is 10.2, suggesting a metabolic alkalosis
5. The respiratory compensation is reasonable - the expected PaCO₂(0.7 × 34.8) + 20 = 44.36mmHg, which is within +/- 5mmHg of the recorded value
6. The anion gap is normal: (137) - (92 + 34) = 11, or 13.5 when calculated with potassium

Thus, this patient has a metabolic alkalosis - likely due to vomiting. A diagnostic approach to metabolic alkalosis is offered elsewhere, and goes through this in some detail. In brief, potenial causes include the following:

• Gastric losses by vomiting or drainage
• Diuretics:
  • loop diuretics
  • thiazides
• diarrhoea
• posthypercapneic state (chronic compensatory renal loss)

• Primary hyperaldosteronism
• Mineralocorticoid oversupplementation
• Licorice (glycyrrhizic acid)
• β-lactam antibiotics
• Liddle syndrome
• Severe hypertension
• Bartter and Gitelman syndromes
• Laxative abuse
• Clay ingestion

• Recovery from starvation
• Hypoalbuminemia

• Hypercalcemia of malignancy
• Milk-alkali syndrome

College Answer

a)
Acute respiratory acidosis

At 38 weeks pregnancy the normal PaCO2 is <30 mmHg with a compensatory reduction in bicarbonate. The blood gases therefore indicate acute CO2 retention probably due to pain and narcotics. 
In the non-pregnant patient these results would indicate an uncompensated normal anion gap metabolic acidosis.

b)
A-a gradient – this is raised at 154 mmHg, suggesting shunt and/or V/Q mismatch. Possible explanations are the loss of FRC after abdominal surgery, segmental collapse/consolidation or aspiration

c)
Hb and WCC – the mild anaemia is physiological in pregnancy and the neutrophil leukocytosis is a normal feature during labour and early post-partum.

Discussion

This question closely resembles Question 18 from the first paper of 2006.

Instead of trying to justify why one identical question ended up in the O&G section and the other among the acid-base questions, I will instead focus on the acid-base and gas exchange abnormalities, for a second forgetting that the patient is pregnant. The obstetric flavour is dealt with in the other version of this discussion section.

For b)

There is a mild respiratory acidosis. The normal CO₂ of late pregnancy is around 30mmHg, which is generally sustained with a bicarbonate of 20. In this scenario the bicarbonate has not changed, and the CO₂ is elevated by 12mmHg.

That makes one think: is the pH influenced by any other disorder? One of the standard equations comes to mind. Every 1mmHg of change in Pa CO₂ leads to a 0.008 change in pH. The use of the standard equation yields an expected pH of 7.304 for this 12mmHg change in CO₂- very close to the measured pH (7.31)

The anion gap is normal if you calculate it without the potassium. It is 15.3 with potassium included, trivially elevated (by 3.3).

Now, as for the A-a gradient:

A-a gradient = (FiO2 x (760-47) - (PCO2 x 1.25) - PaO2

The humidity is always 100% so the left side of the equation is always FiO2 x 713.

So: (0.5 x 713) - (42 x 1.25)

Thus, the alveolar oxygen concentration is 303 mmHg, and the A-a gradient is 153 (303 - 150)

Given that there is a history of abdominal surgery and pregnancy, the causes of this could be
- Post-operative atelectasis 
- Aspiration 
- pulmonary thromboembolism or amniotic fluid embolism

for c), I note that a 15% drop in Hb is normal in pregnancy.

Normal acid-base changes in pregnancy are discussed elsewhere.

College Answer

a)  Describe the abnormalities on the above arterial blood gas profile

• Raised A-a DO2 (78)
• High anion gap (37) metabolic acidosis with markedly raised lactate
• Metabolic alkalosis (delta ratio > 3)
• Respiratory alkalosis (PCO2 lower than expected for HCO3)

b)  List three causes of a raised lactate in sepsis

• Tissue hypoperfusion and hypoxia
• Use of adrenaline (increased glycolytic flux)
• Down regulation of pyruvate dehydrogenase by inflammatory mediators
• Underlying ischaemic tissue

Discussion

This is a straighforward question about ABG interpretation; it presents the candidate with a triple disorder.

This question is frequently repeated. The chapter on the causes of lactic acidosis in sepsis contains a list of past SAQs which are either identical or very similar.

Yes, the A-a gradient is raised: (0.3 x 713) - 18.4 = 195.5, thus the A-a difference is 195.5-107 = 88.5.

How they got 78 I am not sure.

The anion gap is (147) - (95 + 15) = 37, or 43.7 when calculated with potassium.
The delta ratio, assuming a normal anion gap is 12 and a normal bicarbonate is 24, would therefore be (37 - 12) / (24 - 15) = 2.77. This suggests there is also a metabolic alkalosis.

The PaCO2 expected for this sort of bicarbonate level is 30.3 (15 x 1.5 + 8) which suggests that there is also a respiratory alkalosis.

As for the causes of raised lactate in sepsis... The question asks for 3 causes. In the past, the examiners have been firm in saying that if you produce more than the [requested number] of answers, you will only be marked on the first [requested number] answers.

In any case one can usually come up with a few reasons a septic patient might have a raised lactate. The topic of pyruvate dehydrogenase being inhibited by inflammatory mediators is explored in this article.

College Answer

a)  What are the abnormalities?

• Metabolic acidosis with increased anion gap (34 mmol) Increased osmolar gap (approx 85 mmol)
• Hyperosmolar hyponatraemia
• Hyperlactataemia
• Mild hyperglycaemia

b)  Give a possible diagnosis

• Toxic alcohol ingestion (eg methanol, ethylene glycol)
• Alcoholic ketoacidosis
• Formaldehyde ingestion
• DKA possible but osmolar gap in this case higher than expected for DKA

c)  What further tests would you consider to elucidate the cause of the acid base disturbance?

• Specific assays for methanol, ethylene glycol, alcohol
• Urinary calcium oxalate crystals (ethylene glycol)
• Formate level (metabolite of methanol)

Discussion

This is a straighforward question about high osmolar gap high anion gap metabolic acidosis.

The anion gap is (119) - (80 + 5) = 34, or 39.5 when calculated with potassium
The delta ratio, assuming a normal anion gap is 12 and a normal bicarbonate is 24, would therefore be (34 - 12) / (24 - 5) = 1.15, making this a pure HAGMA.

There is also a high osmolar gap of 79 (340- (119x2 + 10+ 13))

Hyperosmolar hyponatremia is not asked about, but is still interesting as an aside.

The causes of a metabolic acidosis with a raised anion gap and a raised osmolar gap are numerous, and well discussed in the literature. Specifically, methanol and the other toxic alcohols are a common cause, and are dealt with in detail elsewhere.

Alcoholic ketoacidosis is mentioned as a differential, although the urinary ketones are absent from the scenario sample. This may be because beta-hydroxybutyrate, the major ketone in alcoholic ketoacidosis, is not detected by urine dipsticks (which look for acetone and acetoacetate).

Smith SW, Manini AF, Szekely T, Hoffman RS.. Bedside detection of urine beta-hydroxybutyrate in diagnosing metabolic acidosis. Acad Emerg Med. 2008 Aug;15(8):751-6.

Kraut JA, Xing SX. Approach to the evaluation of a patient with an increased serum osmolal gap and high-anion-gap metabolic acidosis. Am J Kidney Dis. 2011 Sep;58(3):480-4.

College Answer

i.

• Chronic metabolic alkalosis with partial respiratory compensation

• Background of chronic metabolic alkalosis with partial respiratory compensation could be due to diuretic therapy for heart failure (“contraction alkalosis")

ii.

• Superimposed acute respiratory acidosis and metabolic acidosis (raised lactate)

• Concomitant metabolic alkalosis given that the acidosis is not severe for the degree of PCO2 and lactate + BE is normal for the lactate + AG is also normal.

• The acute superimposed pathology:
  • Cardiogenic shock
  • Sepsis
  • Respiratory depression from opiate or illness
  • PE

Any other reasonable cause

Discussion

So, lets.

In the first gas, the result that stands out is the high pH, with a relatively normal (or slightly elevated) pCO₂ and with a raised bicarbonate. This makes one think of metabolic alkalosis. However, is it properly compensated?

For one the pH is 7.52, and the PCO₂ seems too low by gut feel.

The compensation equation for metabolic alkalosis is 0.7 x [HCO₃^-] + 20; so the PCO₂ in this case should be something like 40-50. (the equation gives us 45mmHg, and it allows +/- 5mmHg error). So, in actual fact the metabolic alkalosis is properly compensated, at least according to the commonly used bedside rules.

So, I suppose we can blame this on frusemide.

Now, for the day 5 gas.

The pH is low; there is acidaemia.

The pCO2 is raised; and it was not raised on day 1, so it is an acute hypercapnea.

If this were a purely respiratory disorder, we would expect the bicarbonate to increase by 1mmol/L for every 10mmHg increase in CO₂. There is an increase of 69mmHg of CO₂, which should give us a bicarbonate value of 31mmol/L or so.

However, the measured bicarbonate is 29.5, which means a metabolic acidosis is also in play.

The anion gap is (141) - (97 + 36) = 8, or 12.2 when calculated with potassium - essentially normal. 

There is a lactate rise, which contributes to this tiny increase in the anion gap. We know there is also a metabolic alkalosis in the background.

Thus, this is a triple disorder, where a metabolic alkalosis maintained by frusemide has been upset by an acute respiratory acidosis and an acute lactic acidosis.

The college answer does not delve too deeply into this.

They ask us then, what could have caused such a rise in lactate and such hypercapnea in this cardiac patient?

"any reasonable cause" is what they want. One can argue that this old man might have developed cardiogenic shock and his hypercapnea is due to his having lost consciousness from low cardiac output; or might have developed an aspiration pneumonia (SaO2 is 89%). Or sepsis of some unknown origin.

Any reasonable cause.

College Answer

i.

• Mixed acidosis: Severe metabolic acidosis with a raised anion gap + additional respiratory acidosis

• Renal failure with hyponatraemia and hyperkalaemia

ii.

• Adrenaline
• Pyruvate dehydrogenase inhibtion by endotoxin
• Liver dysfunction/failure
• Tissue hypoxia

Discussion

1. The FiO2 is not known, but the patient has a reasonable PaO2 and oxygen saturation.
2. There is severe acidaemia
3. The CO₂ is contributing.
4. There is a severe metabolic acidosis. The bicarbonate is 10.
   The anion gap is (123) - (95 + 10) = 18, or 23.4 when calculated with potassium
    
5. There is no compensation for the metabolic acidosis - the expected CO₂ would be 18 (using the SBE) or 23 (using the Boston rules), meaning that there is also a respiratory acidosis.
6. The delta ratio, assuming a normal anion gap is 12 and a normal bicarbonate is 24, would therefore be (18 - 12) / (24 - 10) = 0.42. Thus, this is a mixed high and normal anion gap metabolic acidosis, which is exactly the sort of thing you'd see in renal failure. Given that the lactate does not explain the entire magnitude of the anion gap increase, we can speculate that various non-volatile acids are to blame. 

There are many different reasons for why lactate might rise in sepsis. Apart from the shocked state, there is the catecholamine release, microvascular shunting, mitochondrial dysfunction due to pyruvate dehydrogenase inhibition, and diminished hepatic blood flow. I tried to explain this to myself in another chapter.

College Answer

Uncompensated metabolic acidosis (or metabolic + respiratory acidosis) with raised anion gap not solely due to elevated lactate. 
Raised A-a gradient
Sodium adjusted to normoglycaemia is about 153 
Marked hyperglycaemia (candidates will say this but does not deserve a mark)

Hyperosmolar hyperglycaemic syndrome with component of ketoacidosis and post-ictal lactic acidosis.

Discussion

There is a widened A-a gradient, around 270mmHg. The expected PaO₂ is around 550mmHg.

There is marked acidaemia; this is due to a metabolic acidosis.

The metabolic acidosis is poorly compensated - the expected CO₂ is 24. Thus, there is also a respiratory acidosis.

This is because the patient is ventilated, and has no voluntary compensation mechanisms.

The anion gap is (128) - (82 + 11) = 35, or 38.1 when calculated with potassium. Some, but not all of this is accounted for by the raised lactate.
The delta ratio, assuming a normal anion gap is 12 and a normal bicarbonate is 24, would therefore be (35 - 12) / (24 - 11) = 1.76
This suggests a pure high anion gap metabolic acidosis. The glucose of 79 would suggest that ketones may be contributing to the high anion gap.

There is also some renal impairment.

There is hypernatremia- when adjusted for glucose, the sodium level is ~ 150 mmol/L.

(for every 5.6mmol/L of glucose, sodium drops by 1.6mmol/L)

Thus this is a hyperosmolar state with respiratory acidosis following intubation, metabolic acidosis likely due to ketones, and lactic acidosis likely due to seizures.

Specific summaries dealing with these topics are available:

• Causes of respiratory acidosis and alkalosis
• Causes of metabolic acidosis
• Causes of lactic acidosis
• Hyperosmolar non ketotic hypergycaemic coma (HONK)

College Answer

High anion gap metabolic acidosis secondary to pyroglutamic acidaemia.

Can be detected by requesting an organic acid screen, or by plasma or urine pyroglutamate levels.

Management – cessation of precipitating drugs likely paracetamol and flucloxacillin in this case.

N-Acetyl cysteine infusion has been advocated.

Discussion

Let us dissect these results systematically.

1. The A-a gradient is high:
   PAO2 = (0.3 × 713) - (25 × 1.25) = 182.55
   Thus, A-a = ( 182.55 - 84) = 98.55mmHg.
2. There is acidaemia
3. The PaCO2 is compensatory
4. The SBE is not offered but the bicarbonate is 10, suggesting a severe metabolic acidosis
5. The respiratory compensation is adequate - the expected PaCO₂(10 × 1.5) + 8 = 23mmHg (withn +/- 2mmHg of the reported value)
6. The anion gap is raised:
    (143) - (114 + 10) = 19, or 23 when calculated with potassium
   The delta ratio, assuming a normal anion gap is 12 and a normal bicarbonate is 24, would therefore be (19 - 12) / (24 - 10) = 0.5
   This suggests that there is mixed high anion gap and normal anion gap metabolic acidosis here.
7. Urinary pH and electrolytes are not offered (but would be interesting)

So, what is making this drunk so acidotic? The lactate is not raised; so the differentials for the high anion gap here would include the following:

• Methanol and the toxic alcohols in general
• Ethylene glycol
• Diabetic, alcoholic and starvation ketoacidosis
• Pyroglutamic acidosis
• Salicylate overdose
• Iron overdose

Of course, the MSSA story suggests the college want us to consider pyroglutamic acidosis as a cause.
Investigations for these differentials would include the following:

• 5-oxoproline level in urine
• Urinary oxalate
• Methanol levels
• Ketone levels
• Iron studies
• Salicylate levels

One would want to stop feeding this man the paracetamol and flucloxacillin.

College Answer

a)

Raised A-a gradient of 154 mmHg suggestion shunt and/or V/Q mismatch. Potential explanations are loss of FRC post abdominal surgery, segmental collapse/consolidation or aspiration.

Acute respiratory acidosis – normal PCO2 for 38 weeks gestation is 30 mmHg with compensatory reduction in HCO3. CO2 retention is possibly due to pain, narcotics and/or sedation from anaesthetic agents

Normal anion gap

b)

Anaemia and leukocytosis – mild anaemia is physiological in pregnancy. Neutrophil leukocytosis is a normal feature during labour and early post-partum.

Discussion

This question is very similar to Question 6.2 from the first paper of 2013.

Let us dissect these results systematically.

1. The A-a gradient is high; 
   PAO2 = (0.5 × 713) - (42 × 1.25) = 304
   Thus, A-a = (304-150) = 154mmHg.
2. There is acidaemia
3. The PaCO2 is contributing to the acidosis
4. The SBE is -5, suggesting a metabolic acidosis
5. The respiratory compensation is inadequate - the expected PaCO₂(20.3 × 1.5) + 8 = 38.45mmHg, and thus there is also a mild respiratory acidosis (especially considering that in pregnancy the normal CO₂ value is around 30mmHg)
6. The anion gap is normal:
   (137) - (106 + 20) = 11, or 15.3 when calculated with potassium

Thus, the main disorder here is respiratory acidosis (as the metabolic contribution is minimal - the bicarbonate is close to its normal value in pregancy).

The leukocytosis and anaemia are quite normal peripurpureal findings.

College Answer

a)

• Ischaemic bowel
• Metformin induced lactic acidosis
• Septic shock
• Thiamine deficiency
• Pancreatitis

b) 

• CT abdomen
• Blood cultures / septic screen
• Lipase
• Red cell transketolase 
  Diagnostic laparoscopy / laparotomy

Discussion

This question is identical to Question 15.3 from the second paper of 2009, and Question 26.4 from the second paper of 2013. The answer to the latter contains a comprehensive discussion.

College Answer

a) 

• Non-ketotic hyperosmolar state:
• Marked hyperglycaemia – higher than usually seen in DKA
• Hyperosmolar – approx. 342 mOsm/Kg
• Relatively mild anion gap acidosis accounted for by raised lactate

b)

• Cerebral oedema
• Vascular thrombosis
• Electrolyte derangement
• Intercurrent events such as sepsis, AMI Hypotension and shock if inadequate resuscitation
• Death

Discussion

Let us dissect these results systematically.

1. The A-a gradient is high; 194 or 337.6mmHg depending on what FiO2 you think the Hudson mask is delivering.
   PAO2 = (0.4 × 713) - (36 × 1.25) = 240.2
   Thus, A-a = ( 240.2 - 90) = 150.2mmHg.
2. There is acidaemia
3. The PaCO2 is vaguely compensatory
4. The SBE is not offered but the bicarbonate is 18, suggesting a metabolic acidosis
5. The respiratory compensation is adequate - the expected PaCO₂(18 × 1.5) + 8 = 35mmHg
6. The anion gap is raised:
   (140) - (105 + 18) = 17, or 20.9 when calculated with potassium
   The delta ratio, assuming a normal anion gap is 12 and a normal bicarbonate is 24, would therefore be (17 - 12) / (24 - 18) = 0.83
   This suggests that there is a mixed high anion gap and normal anion gap metabolic acidosis here.
    

The glucose of 40, urea creatinine and the raised lactate are all giveaway clues.

Ths patient has HONK and uremic renal failure. The calculated osmolarity is (140 × 2 + 21.8 + 40) = 341.8 mOsm/L.

Complications of HONK, as listed in the abovelinked page, are as follows:

• Cardiac arrest
• Cardiovascular collapse
• Myocardial infarction
• Stroke
• Cerebral oedema and brain injury
• Venous thrombosis

I do not see how listing "death" as a complication earns any marks in this exam.

Hyperglycemic Comas by P. VERNON VAN HEERDEN from Vincent, Jean-Louis, et al. Textbook of Critical Care: Expert Consult Premium. Elsevier Health Sciences, 2011.

Oh's Intensive Care manual: Chapter 58  (pp. 629) Diabetic  emergencies  by Richard  Keays

Umpierrez, Guillermo E., Mary Beth Murphy, and Abbas E. Kitabchi. "Diabetic ketoacidosis and hyperglycemic hyperosmolar syndrome." Diabetes Spectrum15.1 (2002): 28-36.

ARIEFF, ALLEN I., and HUGH J. CARROLL. "Nonketotic hyperosmolar coma with hyperglycemia: clinical features, pathophysiology, renal function, acid-base balance, plasma-cerebrospinal fluid equilibria and the effects of theraphy in 37 cases." Medicine 51.2 (1972): 73-94.

Gerich, John E., Malcolm M. Martin, and Lillian Recant. "Clinical and metabolic characteristics of hyperosmolar nonketotic coma." Diabetes 20.4 (1971): 228-238.

Kitabchi, Abbas E., et al. "Hyperglycemic crises in adult patients with diabetes." Diabetes care 32.7 (2009): 1335-1343.

Kitabchi, Abbas E., et al. "Hyperglycemic crises in adult patients with diabetes a consensus statement from the American Diabetes Association." Diabetes care 29.12 (2006): 2739-2748.

Ellis, E. N. "Concepts of fluid therapy in diabetic ketoacidosis and hyperosmolar hyperglycemic nonketotic coma." Pediatric clinics of North America 37.2 (1990): 313-321.

Pinies, J. A., et al. "Course and prognosis of 132 patients with diabetic non ketotic hyperosmolar state." Diabete & metabolisme 20.1 (1993): 43-48.

College Answer

a)

High anion gap metabolic acidosis

AG 23 not accounted for by just rise in lactate

Marked lactic acidosis

Respiratory acidosis

Hypoxaemia with A-a DO2 145

Anaemia

b)

Septic shock (intra-abdo, lung, other)

Hypovolaemic shock with intra-abdominal bleeding

Ischaemic gut

Cardiogenic shock (myocardial ischaemia or septic ardiomyopathy)

PE less likely with PO2 84 on FiO2 0.4 but cannot be excluded

c)

Resuscitation with simultaneous focussed assessment (history, examination, investigations) to ascertain diagnosis and institution of treatment (supportive and definitive)

Resuscitation – ensure adequate airway and ongoing adequate oxygenation and ventilation. Intubation and IPPV if needed. IV access and fluid resus plus/minus vasopressors.

Focussed assessment – differential diagnosis as above. Look for signs of bleeding, sepsis, intra-abdominal catastrophe, assess myocardial function

Investigations – FBC, U&E, coags, Troponin, septic screen, ECG, CXR, CT abdo (if stable) ± CTPA, bedside echo

Broad-spectrum antibiotics if sepsis suspected

Surgical review, consider proctoscopy, with urgent return to theatre if indicated (anastamotic leak, ischaemic gut)

Other urgent specific treatment as indicated eg stop bleeding, treat myocardial ischaemia Monitoring and transfer to ICU/HDU

Discussion

Let us dissect these results systematically.

1. The A-a gradient is high:
   PAO2 = (0.4 × 713) - (45 × 1.25) = 157.55
   Thus, A-a = ( 157.55 - 84) = 73.55mmHg.
2. There is acidaemia
3. The PaCO2 is contributing to the acidosis
4. The SBE is -19, suggesting a severe metabolic acidosis
5. The respiratory compensation is inadequate - the expected PaCO₂(11 × 1.5) + 8 = 24.5mmHg, and thus there is also a respiratory acidosis.
6. The anion gap is raised:
   (138) - (104 + 11) = 23, or 28.4 when calculated with potassium
   The delta ratio, assuming a normal anion gap is 12 and a normal bicarbonate is 24, would therefore be (23 - 12) / (24 - 11) = 0.84
   The delta ratio suggests that there is a mixed high anion gap and normal anion gap metabolic acidosis here.
    

The lactate is 13, which can account for most -but not all- of the increase in the anion gap.

So: why is it so high, the college asks?

Differentials relevant to this case might include the following:

• Ischaemic gut
• Sepsis with intra-abdominal collection
• Cardiogenic shock
• Haemorrhagic shock
• "Obstructive" shock - PE or cardiac tamponade

Approaching the management systematically, one might respond to the last question in the following fashion:

• Attention to the ABCS, with management of life-threatening problems simultanous with a rapid focused examination and a brief history
• Airway
  • Assess the need to control the airway
  • This patient will probably require intubation
• Breathing/ventilation
  • Aim for a lower PaCO₂ to help correct the acidosis
  • Apply PEEP to improve oxygenation
  • A CXR would be helpful
• Circulatory support
  • Fluid resuscitation should commence
  • Vasopressors will likely be required
  • Central venous access needs to be established
  • Invasive hemodynamic monitoring will be required
  • a TTE would be helpful
• Specific investigations
  • A CT of the abdomen is required
  • A septic screen should be sent
• Specific management
  • A surgical opinion is required
  • Broad-spectrum antibiotics cannot be delayed

Oh's Intensive Care manual: Chapter   45   (pp. 520) Abdominal  surgical  catastrophes by Stephen  J  Streat

Marshall, John C., and Marilyn Innes. "Intensive care unit management of intra-abdominal infection." Critical care medicine 31.8 (2003): 2228-2237.

Gajic, Ognjen, et al. "Acute abdomen in the medical intensive care unit."Critical care medicine 30.6 (2002): 1187-1190.

College Answer

Acute respiratory acidosis with metabolic alkalosis.

Clinical scenario – acute respiratory failure in COAD (Acute on chronic respiratory failure.

Discussion

Let us dissect these results systematically.

1. The A-a gradient cannot be calculated - the information is missing
2. There is acidaemia
3. The PaCO₂ is massively elevated, suggesting a primary respiratory acidosis
4. The SBE is not supplied, but the bicarbonate is 34, suggestive of a metabolic alkalosis
5. The metabolic compensation may not have titrated the pH back to normality, but it appears to be excessive - the HCO₃^- is expected to increase by 1mmol/L for every 10mmHg increase in CO₂ in an acute respiratory acidosis, which would give us (10 × 5 + 24) = 29mmol/L. The fact that the HCO₃^- is 34mmol/L suggests that there is a metabolic alkalosis present.
6. The anion gap is essentially normal: 
   (145) - (101 + 34) = 10, or 14 when calculated with potassium
7. The delta ratio is of no use here.
8. The urinary electrolytes and pH would not be helpful.

Thus, this is a respiratory acidosis with a coexisting  metabolic alkalosis. Anion gap is normal. Scenarios which might explain these findings may be any of the following:

• Hypercapneic respiratory failure in chronic COPD
• Respiratory failure due to acute pulmonary oedema in a patient on chronic loop diuretic therapy
• Hypercapneic respiratory failure due to obtundation in a vomiting opium fiend

Bear, R., et al. "Effect of metabolic alkalosis on respiratory function in patients with chronic obstructive lung disease." Canadian Medical Association Journal117.8 (1977): 900.

Bruno, Cosimo Marcello, and Maria Valenti. "Acid-base disorders in patients with chronic obstructive pulmonary disease: a pathophysiological review."BioMed Research International 2012 (2012).

College Answer

Acute respiratory alkalosis. Clinical scenario – Psychogenic hyperventilation

Discussion

Let us dissect these results systematically.

1. The A-a gradient cannot be calculated - the information is missing
2. There is alkalaemia
3. The PaCO₂ is decreased, suggesting that it is contributing to the alkalosis
4. The SBE is not supplied, but the bicarbonate is 22, suggestive of a metabolic acidosis or metabolic compensation
5. The metabolic compensation may not have titrated the pH back to normality, but it appears to be adequate - the HCO₃^- is expected to fall by 2mmol/L for every 10mmHg of acute decrease in CO₂ in an acute respiratory alkalosis, which would give us 20mmol/L. The fact that the HCO₃^- is 22mmol/L suggests that there is a mild metabolic alkalosis present.
6. The anion gap is normal:
   (135) - (105 + 22) = 8, or 12 when calculated with potassium
   The delta ratio is of no use here.
7. The urinary electrolytes and pH would not be helpful.

Thus, this is a respiratory alkalosis with virtually no other biochemical changes. A scenario which might explain these findings may be any cause of hyperventilation:

• Anxiety
• Fear
• Pain
• Dysregulation of central respiratory control (eg. severe brain injury)
• Ridiculous ventilator settings

Barker, E. S., et al. "The renal response in man to acute experimental respiratory alkalosis and acidosis." Journal of Clinical Investigation 36.4 (1957): 515.

Foster, Guy T., Nostratola D. Vaziri, and C. S. Sassoon. "Respiratory alkalosis." Respiratory care 46.4 (2001): 384-391.

Giebisch, Gerhard, et al. "The extrarenal response to acute acid-base disturbances of respiratory origin." Journal of Clinical Investigation 34.2 (1955): 231.

College Answer

6.1 a)

Mixed respiratory and metabolic alkalosis

6.1 b)

Respiratory alkalosis from the hyperventilation due to the pneumonia.

Metabolic alkalosis from vomiting or diuretic use.

Discussion

Let us dissect these results systematically.

1. The A-a gradient is high: 
   PAO2 = (0.3 × 713) - (31 × 1.25) = 175
   Thus, A-a = ( 175 - 83) = 92mmHg.
2. There is alkalaemia
3. The PaCO2 is contributing to the alkalosis
4. The SBE is 3.3, suggesting a mild metabolic alkalosis
5. There is no respiratory compensation.This disorder ends up a mixed metabolic and respiratory alkalosis whichever way you calculate it.
   The expected PaCO₂ is (0.7 × 25) + 20 = 37.5mmHg, and thus there is also a respiratory alkalosis. 
   If you were to assume that the respiratory disorder is acute and the primary cause of the problem, you would calculate a 2mmol/L decrease in HCO₃^- for every 10mmHg drop in PaCO₂. That would suggest that the expected HCO₃^- for this patient would be somewhere in the vicinity of 22mmol/L.

When offering a "likely diagnosis" to explain this gas, one can resort to mix-and-matching any cause of metabolic alkalosis (eg.vomiting, diuretic use) with any cause of respiratory alkalosis (eg. anxiety, pain, etc). For all we know, she maybe prone to panic attacks and has been binge-eating licorice.

McCurdy, Donna Kern. "Mixed metabolic and respiratory acid-base disturbances: diagnosis and treatment." CHEST Journal 62.2_Supplement (1972): 35S-44S.

Khanna, Apurv, and Neil A. Kurtzman. "Metabolic alkalosis." J NEPHROL 2006; 19 (suppl 9): S86-S96

Barker, E. S., et al. "The renal response in man to acute experimental respiratory alkalosis and acidosis." Journal of Clinical Investigation 36.4 (1957): 515.

College Answer

1. a)

Increased A-a DO2 secondary to decreased FRC post surgery or possible collapse/consolidation or aspiration or pulmonary oedema.

Acute respiratory acidosis on background compensated respiratory alkalosis of pregnancy. CO2 retention secondary to e.g. hypoventilation post anaesthetic.

Discussion

Let us dissect these results systematically.

1. The A-a gradient is high:
   PAO2 = (0.4 × 713) - (42 × 1.25) = 232.7
   Thus, A-a = ( 232.7 - 110) = 122.7mmHg.
2. There is acidaemia
3. The PaCO₂ is contributing to the acidosis(if one recalls that in pregnancy, the PaCO₂ is usually somewhat lower than normal)
4. The SBE is -4.9, suggesting a metabolic acidosis
5. The respiratory compensation is inadequate - the expected PaCO₂(20.5 × 1.5) + 8 = 38.75mmHg, and thus there is also a mild respiratory acidosis. 
   One can also view this in terms of metabolic compensation. The chronic respiratory alkalosis of pregnancy drives the PaCO₂ down to about 32-30mmHg, and thus the bicarbonate decreases by 5mmol/L to 19-20mmol/L. Knowing this normal range, one can surmise that in this patient the bicarbonate is actually within the normal range.
6. The anion gap and delta gap cannot be calculated

Thus: this post-LSCS patient is hypoxic and acidotic, with a predominantly respiratory acidosis. This is not a completely unexpected picture. If she had a spinal, one would be tempted to check her sensory level, to ensure that the respiratory muscles are unaffected. However, the college tells us there was a general anaesthetic. Atelectasis, loss of FRC and opiates are to blame.

Loverro, G., et al. "Indications and outcome for intensive care unit admission during puerperium." Archives of gynecology and obstetrics 265.4 (2001): 195-198.

College Answer

a)

Normal anion gap metabolic acidosis with appropriate respiratory compensation.

b)

Resuscitation with large volume saline infusion.

c)

ECF dilution by fluid with strong ion difference of zero.

Discussion

Let us dissect these results systematically.

1. The A-a gradient is probably high; the FiO₂ is not supplied, but the patient is hypoxic.
2. There is acidaemia
3. The PaCO2 is compensatory
4. The SBE is -9, suggesting a metabolic acidosis
5. The respiratory compensation is adequate - the expected PaCO₂ is (16 × 1.5) + 8 = 32mmHg.
6. The anion gap is normal:
    (147) - (124 + 16) = 7, or 10.6 when calculated with potassium
7. Urinary pH and electrolytes would be interesting, but are not supplied.

Thus, this is a normal anion gap metabolic acidosis, which is well compensated. One possible reason for this might be an enthusiastic overabundance of saline.

Judging from the answer, the college was expecting a  Stewart-based physicochemical approach to acid-base analysis, which is rather refreshing to see. Normal saline intoxication is dealt with elsewhere.