This chapter is relevant to Section J1(iii) of the 2017 CICM Primary Syllabus, which expects the exam candidates to "explain the physiological basis to clinical acid – base disturbance". That sounds like a lot, and has been hugely misinterpreted by the author, giving rise to this giant unwieldy acid-base section. This chapter is specifically concerned with the changes in pH and serum bicarbonate which result from acute fluctuations in dissolved CO2, as a consequence of acute changes in ventilation. It is a more detailed look at the way CO2 interacts with the human body fluid, and the resulting changes which develop in the serum bicarbonate concentration and pH.
Let us consider the favoured model of acute respiratory acidosis, the patient who has stopped breathing.
Conventional wisdom dictates that so long as the oxygen supply continues to mass-transfer its way into the patient, then the patient will continue to produce CO2, and as a result of this metabolic activity the PaCO2 will rise at a rate of around 3mmHg every minute. This technique of "apnoeic anaesthesia" is well known to anaesthetists, and has enjoyed a fluctuating level of interest since the sixties. With a high PEEP and a sufficient attention to detail one may go through the entire hour-long case without any breaths being taken by the patient.
But, let us consider a situation where the airway is not patent, and a constant supply of oxygen is not available. The patient has stopped exhaling CO2. What will happen?
Well, the PaCO2 will rise by about 12mmHg over the first minute, and by about 3.4 mmHg per minute for every minute after that.
How do we know this? Because in 1989, 14 volunteers consented to having their tube clamped during an anaesthetic. The clamps were released after 5 minutes, or if the patients became dangerously hypoxic.
Knowing the change in PaCO2, one can attempt to model the drop in pH during a period of apnoea. The variables pCO2, pH and HCO3- are all linked by the Henderson equation, which is as follows:
Using this equation, one can generate a graph of slowly dropping pH in an apnoeic patient:
Thus, according to the Henderson equation, when there is an acute rise in PaCO2 to 70mmHg the expected bicarbonate level will be about 27, and the pH will drop to about 7.21.
Though trimmed of all fat, Henderson equation can still be difficult to use in the clinical battlefield.
A mathematical shortcut exists, for bedside consumption:
For every 10 mmHg increase in PaCO2, the pH will decrease by 0.08
In other words, pH = 7.40 - ((PaCO2 - 40) × 0.008))
One can see that this shortcut somewhat overestimates the magnitude of the pH change. For a rise of PaCO2 from 40mmHg to 70mmHg, this shortcut would expect a pH of 7.17, whereas the Henderson equation reports a more conservative pH (7.21).
This is not a minor discrepancy, and is a demerit against the practice of using this pH-based compensation rule. Though this "0.008" coefficient is quoted widely, I cannot find its origin anywhere. Who came up with this? Why does it keep appearing in authoritative texts? Even "The ICU book" uses this value in the ABG interpretation chapter.
Close inspection reveals a hard stoichiometric origin for this rule. It appears to be the derived from the use of the Henderson equation without changing the bicarbonate. In this version of the equation, all buffering of H+ by intracellular proteins and phosphate is ignored, and the bicarbonate remained 24mmol/L regardless of the pCO2. A falsely decreased pH value is the result.
So, how do you account for protein and phosphate buffering?
The last equation above uses the "1 for 10" rule, one of the Bedside Rules for Acid-Base Compensation.
For every 10 mmHg increase in PaCO2, the HCO3- will rise by 1 mmol/L
(calculating from a baseline of 40mmHg PaCO2 and 24 mmol/L HCO3-)
In other words, expected HCO3 = 24 + ((PaCO2-40) / 10)
This predicted bicarbonate increase was not calculated using theoretical equations estimating protein or phosphate buffering, but was derived from empirical measurements of the changes in bicarbonate occuring in healthy volunteers. Thus, obviously it is also an estimate, and subject to error. Where are all these healthy volunters in ICU? Does this coefficient apply to our anaemic patients, with their massive oedema, albumin depletion, and numerous other buffer system derangements? What if you shove some THAM up there, how would that work?
Those concerns aside, the rule is fairly easy to implement. If the "expected" bicarbonate is different from the "actual" bicarbonate (derived from the Henderson-Hasselbalch equation), then a metabolic acid-base disturbance is present.
An acute increase in PaCO2 will not change the Standard Base Excess.
What a boring rule.
The Copenhagen school of ABG analysis goes to some considerable lengths to separate the physicochemical effects of increasing PaCO2 on bicarbonate from the role of bicarbonate as a buffer for metabolic disorders. This might seem to competely obliterate its predictive utility in acute respiratory acid-base disturbances. However, that is not the case.
Without having to worry about bicarbonate, we can separate PaCO2 and SBE, allowing them to do their own thing. PaCO2 measures the respiratory disturbance, and SBE the metabolic. If the PaCO2 is raised but the SBE is normal, there must be an acute respiratory acidosis present, and there cannot be any metabolic acid-base disturbance. Metabolic compensation for a chronically raised PaCO2 does occur, in which case the SBE will rise over time by a predictable amount (0.4 mEq of SBE per every 1mmHg change in PaCO2), but acutely there will be no change whatsoever because the bicarbonate buffer system cannot buffer itself, and though the actual bicarbonate concentration might change, the buffering capacity of the extracellular fluid will remain the same.
In short, all of these compensation rules are imperfect. We use them to generate a sensation of control over a system with too many unknown variables, in order to prevent a descent into acid-base anarchy.