Thank you for that introduction. As you heard, my name is Jake Wolf, and I am presenting the topic for today entitled Arterial Blood gases, Obtaining a sample and Avoiding Pre-analytical Errors. When thinking about the term preanalytical errors, A number of studies have explained this quite nicely in describing a brain to brain loop concept.
Where An initial test is ordered. First grade. And then it is evaluated later on.
The second brain concept. And in between, we have the pre-analytical portion, which marks the time from the ordering of the test to inserting the sample into a machine. The analytical period, which involves the machine's analysis of the sample, And then the post analytical period or the physician's interpretation of the results produced by the machine.
Importantly, When you look at studies in human medicine have shown that pre-analytical errors account for approximately 75% of the errors in blood gas analysis. Therefore, it's very important to think about the possible causes of these pre-analytical errors when obtaining a sample. First, let's remind ourselves on why we may perform a blood gas.
We usually receive these results immediately, within a few minutes, and thus they can be an essential tool in in in an emergency room or in a busy ICU. Blood gases give us an acid-base assessment. And if they're an arterial sample, they also allow us to evaluate a patient's oxygenation and ventilation.
And studies in human medicine have shown that blood gases are more likely to identify hypoxemia in patients, especially amongst outpatients, than a physical exam or pulse oximetry. A number of blood gas machines also allow us to assess other clinopathologic variables and have added in analysis of electrolytes, kidney values such as BUN and creatinine, and blood glucose, expanding their utility. Our first focus is going to be on arterial blood gas specimen collection in dogs and cats.
There are a number of sites available for arterial blood sampling. Sites include the dorsal pedal artery, which is probably the most common location in conscious dogs and cats. The aricular artery.
The coccygeal artery, which is more commonly performed in anaesthetized cats, And then less commonly performed areas include the femoral artery and the lingual artery, the latter of which you would expect those animals need to be heavily sedated. Arterial catheterization is another way to obtain arterial blood gas samples, and this is especially helpful if you are planning on obtaining repeated arterial samples, whether it's under anaesthesia or for a more critical patient. The most common artery to catheterize is the dorsal pedal artery on the medial aspect of the distal hind limb.
While the video on this webinar will not play, the slides sent to you after will showing catheterization of this artery. The area should be clipped and then prepped in sterile fashion. And then the artery is palpated, usually with two fingers, the middle finger and the index finger, to get an idea of the pathway that the artery travels, usually between the 3rd and 4th digits.
The The artery is then catheterized by placing a small gauge catheter. From distal to proximal into the artery. The acular artery also may be considered as a location for catheterization, especially in hound dogs such as basset hounds.
The coccygeal artery can be a common location to catheterize, especially in cats, under anaesthesia, although this is not an option for awake animals. Much less commonly, the femoral, radial, and brachial arteries may be catheterized. There are a number of potential complications of arterial catheterization or sampling.
These can include haemorrhage from the artery, and it may be best to avoid arterial sticks in thrombocytopenic or coagulopathic animals. Hematoma formation is probably the most common complication found in our domestic species, and therefore applying a tight bandage for a prolonged period of time is necessary post sampling. Thrombosis and distal ischemia.
Infection Pain and distant thrombi have also been reported. Anaemia with repeated sampling is also common, and studies in humans have shown that sampling, repeated sampling can account for 12 to 21% of a patient's blood loss while in hospital, with arterial blood gases themselves totaling 24 to 57 millilitres per day removed from a patient. And in many of our sickest patients who we may be performing arterial blood gases in.
This is especially important to remember, as they may already be anaemic coming in. Following sampling, it's not uncommon for the artery to spasm after being catheterized or sampled from. And that may make future draws from that artery more challenging, as you will not be able to palpate the artery as easily.
There are a number of carrier options when attaining a blood gas sample. Gastite syringes are probably the most common carrier to use for an arterial blood gas. There is a long-standing debate within medicine about the use of glass versus plastic syringes.
Glass syringes were previously preferentially used as plastic syringes are partially permeable to oxygen, and therefore can alter your partial pressure of oxygen in the arterial sample. However, if samples are analysed in under 15 minutes, these values should be acceptable in a plastic syringe. It's important to note that oxygen permeability may worsen with decreased temperature, so a plastic syringe that's placed on ice will be more permeable to oxygen, and therefore may falsely increase your oxygen levels.
Additionally, specialised syringes with a vent do exist to allow for automatic filling of the syringe with the arterial pulse, which is a good way to note that you have entered an artery rather than a vein. Tight fitting plungers or stoppers are essential for these syringes to prevent. The equilibration with atmospheric oxygen and CO2.
You may also draw arterial samples up in a syringe and then place them in an anticoagulated. Test, blood tube. Or draw them up into, draw them up into a syringe and place them in a capillary tube as seen on the right.
Which carrier option you use may be governed by the type of analyzer you have, and most analyzers require at least 0.3 millilitres of whole anticoagulated blood. As alluded to in the previous slide, anticoagulation is essential for evaluation of arterial blood gases.
Heparin is the anticoagulant of choice, as EDTA will chelate cation and prevent evaluation of your Sodium, potassium, as well as your chloride, and therefore your acid-based status as well. There are 2 options for heronization. Sodium heparin, which is most commonly found in its liquid form, and lithium heparin.
Sodium heparin may bind c ions and therefore will affect your sodium, potassium and chloride levels. It may also cause a decreased PA CO2 and bicarbonate if diluted to more than 10%. 60 units per millilitre of sodium heparin has been shown to significantly decrease the aforementioned values, so your PA CO2, your bicarbonate, and your electrolytes.
As well as your PAO2. However, That quantity of sodium heparin may not cause clinically significant changes in your pH. A study performed in client-owned dogs showed significant changes in CO2, base excess, and lactate, calcium, and potassium levels, all decreased using liquid sodium heparin.
This was especially pronounced with 0.1 millilitres of heparin used per millilitre. There have been conflicting study results on whether pH and PO2 are more significantly affected.
However, sodium heparin solution has a pH of 6.4. PCO2 of 7.5 and a PO2 of 160.
So it would not be surprising that Overusing the amount of sodium heparin. Via Dilution could cause significant changes in those values. Many people, when drawing up an arterial blood gas, will use an evacuated syringe technique where you will draw up sodium heparin.
Several times, line the syringe and then inject back into the bottle of sodium heparin. This technique may be acceptable for all. Evaluations except ionised calcium.
When using sodium heparin, you should dilute the sample no more than 3.8%, which is essentially 40 units for 1 millilitre. However, if you acquire less blood than that on your sample, during your sampling, This will further dilute and may affect your values.
Truly, 1 unit per millilitre of heparin is all that is likely necessary to anticoagulate your sample. Lithium heparin is probably more commonly used in its liopolized form in pre-purchased syringes that have already been anticoagulated, and you must use this form if your analyzer has an ionised calcium, and you are analysing the ionised calcium. It is important to confirm that you have drawn an arterial sample.
When Evaluating your arterial blood gas. This is especially true because in hypoxemic patients, the PAO2 may be significantly low, and therefore, it can be confusing whether this is a venous or arterial sample. The best way to do this is to draw a concurrent Venus sample.
Evaluation of the two samples will show a Lower PO2 for the Venus sample than the PAO2. And the PVCO2 will be higher than the PACO2, usually by about 5 millimetres of mercury. Because the CO2 will be higher in the venous rather than the arterial CO2, the pH of the venous sample will also be lower than the arterial sample.
Let us now briefly touch on acquiring Arterial samples and exotic species. As it is not uncommon to have exotic animals present through the emergency service or through your ICU. I've included only 2.
Groups of exotic species, rabbits and sitvicenes. As Studies evaluating samples in other species, such as guinea pigs, ferrets, rats, and mice have been performed either only on laboratory animals or on Patients who are under general anaesthesia due to the difficulty in acquiring an arterial sample. Therefore, those results may not be as applicable to the ones that we see through our services.
I'll just start with rabbits. Most Important study evaluating rabbit arterial blood gases was performed in conscious dwarflops that presented for castration. This arterial blood gases may be important, as rabbits often hyperventilate or have altered respiratory status due to pain or abdominal diseases.
Blood gases may also be important because rabbits may breath hold when nervous, which can complicate your your findings. And pulse oximetry is often unreliable in rabbits. The central articular artery is the artery of choice for sampling from.
Domestic rabbits. It is recommended that lidocaine gel is applied 60 minutes prior to sampling from the artery to decrease the pain associated with drawing from this this location. Hematoma formation is the most common risk associated with sampling the acular artery.
Importantly, no ear tip necrosis was noted in rabbits that had arterial samples drawn from their arricular artery. When sampling from rabbits, it is important to note that the mean AA gradient for rabbits is higher than for other species, as a mean was found in the mid-twenties. Rabbits also have a higher dead space fraction than other species, and therefore are less able to ventilate properly with higher hyperventilation due to the rebreathing of CO2 that occurs.
It's calculating your alveolar arterial gradient. Using an alternative respiratory quotient is essential due to rabbit's herbivorous diet, and it's recommended using an RQ of 0.755.
A lower arterial partial pressure of CO2 was noted for rabbits in this study than for other species. In sampling from parrots, the deep radial artery is the most common site for birds larger than 200 grammes, and the superficial ulnar artery was the most common site for Birds less than 200 grammes. You may also consider the cranial tarsal or dorsal metatarsal arteries, especially for long-legged birds and water birds.
Studies evaluating Amazon parrots showed that this species has a lower PACO2, ionised calcium, and bicarbonate than we see with our mammals. Most species of birds also have a higher pH, up in the mid 7.4, up to 7.5 as their main pH.
Which is important to keep in mind when evaluating against. Mammal species. This slide nicely demonstrates the most common site for arterial sampling or catheterization in cytosines.
Circled arteries include the deep radial artery and the superficial ulnar artery, with the deep radial artery being the more common site of sampling. Let us next move on to arterial blood gas sample handling. Arterial samples should be evaluated as soon as possible to avoid any changes to metabolic or respiratory status for a patient.
If immediate evaluation is not possible, then samples should be evaluated in less than 15 minutes, and this should preserve most analytes. The sample should be inverted several times before analysis to prevent clot formation. It is also essential to evacuate any air bubbles that may have been acquired during sampling, as this will significantly affect results.
Whether to ice or not is one of the biggest debates in Medicine regarding arterial blood gases. As we discussed previously, if the sample is in a plastic syringe, icing may falsely increase your PAO2. Indeed, icing is probably unnecessary if the sample is analysed within.
30 minutes. As you can see from the chart below, Icing, may be indicated depending on the time course of the sample. As you can see, PAO2 and PACO2 are equally affected within 15 minutes, whether you ice or not.
However, icing can prolong your evaluation of pH, base excess, and lactate. There are statistically significant, but clinically insignificant changes in electrolytes that occur. When, when there are delays in icing and evaluation.
Importantly, lactate increases rapidly when not cooled. It increases on average by about 0.036 millimoles per litre over 30 minutes, or 0.01 millimo per litre per minute.
Indeed, a significant difference was noted within 5 minutes between icing and noticing, and therefore evaluation of uniced lactate concentrate lactates should be processed as quickly as possible. Another common debate in blood gas analysis is whether temperature correction is whether it is necessary for either arterial or venous blood gas analysis. With a decreasing temperature, the solubility of oxygen and CO2 increases.
This results in a higher affinity for oxygen to haemoglobin. This Causes increased binding between the two of these, and therefore decreased diffusion of oxygen to the tissues. Therefore, a decreased temperature will result in a decreased CO2, a decreased oxygen, and an increased pH.
Therefore, temperature correction would likely be ideal. However, we do not know the normal partial pressures for various temperatures, and therefore, this makes temperature correction difficult to interpret, because we do not know the true reference ranges for various temperatures. Additionally, we are using rectal temperatures rather than pulmonary arterial temperatures.
Because of the air blood interface in the lungs, pulmonary arterial temperatures are likely lower than rectal temperatures, and therefore, this may also affect the results, and studies in horses and ox. Supported these findings. There is no consensus on whether temperature correction should occur for Arterial blood gas analysis.
Therefore, the most important thing is to probably be consistent in either temperature correcting or not. When evaluating your sample. Let us now move on to the most important part of the talk, discussing arterial blood gas, pre-analytical errors.
First, we'll discuss errors with the partial pressure of arterial oxygen. If you draw a sample and There are air bubbles within the sample. Even if they occupy only 1 to 2% of the blood volume, this may artificially increase your PAO2.
If your sample is allowed to mix with room air, whether it sits too long, is iced, or is uncapped, this will often cause the PAO2 to increase to greater than 100 millimetres of mercury. Severe leukocytosis may also cause consumption of oxygen and artactually lower your PAO2. There are also a number of errors that we can see with the arterial partial pressure of carbon dioxide.
As we discussed before, modest increases occur after 30 and 60 minutes at room temperature. Because atmospheric. CO2 is less than arterial CO2.
If mixing with room air is allowed, this will falsely decrease your arterial CO2. With an increase in your arterial oxygen, you may also have a concurrent decrease in your CO2. When this occurs, since your CO2 is an acid.
Your pH will increase. The errors with bicarbonate and base excess are often less significant than those of other analytes. As we mentioned before, There may be dilutional changes in bicarbonate, and therefore base excess with Increased levels of sodium heparin.
Increased lactate levels due to altered sampling time may cause a worsening metabolic acidosis, as reflected with decreased bicarbonate or base excess. Bicarbonate levels may also drastically change depending on the CO2 levels, which are dependent on the listed variables in the previous slide. There are also a number of pre-analytical errors that can occur with the electrolytes that we evaluate on the blood gas.
In alkalemia from a decreased CO2 may cause an increased affinity of calcium for albumin. This will result in a decreased ionised calcium. And therefore, if your sample has sat out too long and causes a decrease CO2, we're going to have a concurrent decreased ionised calcium.
As we mentioned before, liquid heparin will also cause calcium binding, and therefore, if you use sodium heparin in your syringe, you cannot eva evaluate your ionised calcium. There are a number of errors that can occur with sodium. Analysis on arterial blood gases.
A study looking at preterm infants in paediatrics noted that sodium levels were significantly lower on arterial blood gases and on a central analyzer, with a mean difference of 4.2 millimole per litre between the two analyzers. This difference was even more pronounced in patients with hypernatraemia.
Hypoalbuinemia may also exacerbate sodium changes. There's also the phenomenon of pseudohyponatremia to consider with sodium analysis. Blood gas analyzers use the percent of water compared to non-aqueous components to construct the sodium concentration in a sample.
The normal percent of water to non-aqueous component is 93 to 7%. If a patient has hyperlipidemia or hyperproteinemia as seen with multiple myeloma, This may alter the water to non-aqueous ratio by increasing the non-aqueous component. This will artfactually decrease the sodium on your analyzer, since they are still using the 93 to 7% comparison.
A number of errors can also occur with potassium evaluation. Homolysis may cause modest changes in A variety of parameters. However, homolysis can cause a significantly increased potassium.
Due to the high potassium concentration intracellularly. There is no reliable indicator of homolysis on most blood gas machines. And therefore, either measuring cell-free haemoglobin on the analyzer.
Or spinning down a micro hematocrit tube could allow you to evaluate for levels of homolysis. This is especially important in patients who may have difficult blood draws, or in patients in which the artery was stuck repeatedly. Causing Altered sampling.
Marked thrombocytosis or leukocytosis may also cause increased potassium levels. There are a number of errors we can see with blood glucose on the arterial blood gas machine as well. A prolonged pre-analytical time may allow for metabolism of glucose by erythrocytes and therefore falsely decrease your blood glucose.
There is great variability between arterial blood gas devices in their reliability for measurement of glucose. And indeed arterial blood gases may have up to 42% variability compared to a central laboratory in some studies. However, Human studies also found that arterial blood gases were likely more accurate than a bedside glucometer.
Arterial blood gas machines also have often have unknown accuracy for patients with severe hypoglycemia, and therefore these results should be double checked. Let us now evaluate a few cases to reinforce these ideas. Our first patient is Arthur.
He is a one year old male castrated French bulldog who presented to your emergency service for an acute history of respiratory distress. On your triage exam, he is orthopic with a respiratory rate of 30 and markedly increased respiratory effort. You perform an arterial blood draw.
However, the machine was calibrating, so it sat for 20 minutes. You get your results on the arterial blood gas, your pH is 7.53.
Your CO2 is 20. And your PO2 is 95 millimetres of mercury. Your base excess is -4.7, your lactate is 2.5, your blood glucose is 50, and your ionised calcium is 0.79.
Pretty clearly, sitting for 20 minutes probably altered a number of these variables. First, as we discussed with the acid-base status, For every minute that a sample sits, the lactate may increase by 0.01 millimoles per litre, and therefore, sitting for 20 minutes probably decreased our increased our lactate from 2.3 to 2.5.
And also would have caused a similar change. And a decrease for our base excess. Sitting for that long they have also allowed for metabolism of our Glucose in the sample, so that should be interpreted with caution.
My main concern for the oxygenation ventilation status for this patient is that the tube was probably not capped, and this would have artfactually increased the PO2. Artifactually decreased the CO2, and because of the decrease in the CO2, increased the pH, making the sample alkalemic. This alkalemia would have also caused the decreased ionised calcium.
Therefore, this sample does not allow us to truly Address the patient's oxygenation, ventilation. Acid-based status or electrolyte status and a new sample would need to be drawn. Our next patient is Loki.
He is a six year old male castrated sphinx cat who presented to us for a one-day history of stranguria and urea and vomiting. On your physical examination, you know a large firm, non-expressible bladder and an increased respiratory rate. Because if you do, you do not know whether the increased respiratory rate is due to pain, anxiety, or an altered respiratory status, you obtain an arterial blood gas.
Unfortunately, Loki is not the most pleasant patient, and you have a difficult blood draw from his femoral artery. On your arterial blood gas, you have a pH of 7.28, a PCO2 of 37 millimetres of mercury, and a PO2 of 88 millimetres of mercury.
Your base excess is -7, your lactate is 3.2, your potassium is 6.7, and your creatinine is 1.9.
Therefore, you have a mild azotemia with a moderate hyperkalemia. And a mild hyperlactatemia. My suspicion is the hyperlactatemia is probably due to the difficulty of the blood draw.
Your creatinine should go up before your potassium in a patient with a post-renal azotemia. And therefore, this More significant increase in potassium than creatinine is suspicious. I would worry that there could be homolysis of the sample due to the difficult blood draw.
And I would recommend spinning a sample down to see if there is evidence of hemolysis when examining the plasma. Our next patient is Ichabo. He's a 9 year old male castrated greyhound, who's actually sitting right next to me as I'm recording this.
He was hit by a car and presents to you laterally recumbent. On your physical examination. And On AA, you suspect a hemoabdomen.
He also has an increased respiratory rate and effort, and you are worried about about pulmonary contusions. You quickly grab a syringe and heronize it via an evacuated syringe technique. And acquire a sample from the dorsal petal artery.
Your arterial blood gas reveals a pH of 7.3, CO2 of 25 millimetres of mercury, and an oxygen of 65 millimetres of mercury. Your sodium is 180, your II's calcium is 0.3, and your bicarbonate is 17.3.
Pretty clearly, the first thing that you may worry about is that your sodium is falsely elevated from using sodium heparin, and that your ionised calcium. Must be discarded because you cannot evaluate an ionised calcium when using sodium heparin. Only liophyli lithium heparin can be used.
As we discussed before, a dilutional effect may also occur when using sodium heparin, and this may cause abnormal CO2 results, which can also affect your bicarbonate results, as well as oxygen results. And it would probably be recommended. To redraw a sample using only one.
5 units of pepper. Let us now consider our last case. We have Athena, a 12 year old female spayed Persian cat, who presented to us for a carpal fracture after a mousetrap injury.
She was sedated for radiographs and 10 minutes later developed an increased respiratory rate and respiratory effort. You decide to perform an arterial blood gas and obtain a sample from her coccygeal artery. The arterial blood gas reveals a pH of 7.45, a CO2 of 35, and a PO2 of 50.
Her lactate on the sample is 1.9. Due to the markedly decreased PO2, you're concerned that this could potentially be a venous sample or a mixed arteriovenous sample, and decide to obtain a second arterial sample.
From a dorsal pedal artery. On this sample, the pH is 7.48.
The CO2 is 30 millimetres of mercury, and the PO2 is 90. The lactate is 1.7.
As you can see, these findings are consistent with the first sample being a venous sample and the second sample being an arterial sample. As the lactate is lower on the arterial sample, The PO2 is significantly higher on the arterial sample. And the PCO2 is higher on the Venus sample.
Because the CO2 is higher on the Venus sample, the pH is also slightly lower on the Venus sample. With that, the lecture portion of our webinar has concluded. I would like to thank you all for joining us this evening, and I would also like to thank my mentor, Doctor Deb Silverstein, for all her help, as well as Ron and Julia and all those at Nova Biomedical for assisting with this presentation.
I will now happily take any questions that you might have. And here are my listed references should you need any further sources to go for any of this material. OK, thank you everyone for again listening to to this webinar.
This is Jake. I'm happy to take any questions you may, you may have. You can just type them into the Q&A box there, and I'll read the question out to the rest of our listeners and do my best to answer it for you guys.
Like
