Gas and Hot Air

Journal Reviewed:  Anesthesiology
Issues: November 1999, Volume 91, Number 5
Abstracted by: Dr B Piketh  MB BCh (Registrar, University of the Witwatersrand)

Summary of abstracts

The first article can only be considered very preliminary.  It is the first time Perflubron is being used in a prospective, randomised clinical trial.   Although the conclusion that administering Perflubron can delay the onset of a blood transfusion is not really supported in the article, the research was hampered by its own design.  It is still important as it lays the ground work for the next trial, which can use the safety factor from this trial, to have more liberal withholding of blood.

The second article  promises to change the way we absorb carbon dioxide in breathing systems forever.

The third article has taken an "off the wall" suggestion and turned it into three excellent clinical studies, which may well change our view of intra-operative oxygen supplementation.

The article on using PEEP in normal and obese patients can not be read in isolation.  The authors have not used PEEP to its full advantage, and their finding that it changes chest wall compliance seems to suggest a measurement error.

You may also wish to briefly browse our editorial comment

1.  Perflubron emulsion delays blood transfusions in orthopedic surgery

This article is accompanied by an Editorial view which lays down an interesting background

Tremper, KK.  Perfluorochemical "blood substitutes": Indications for an oxygen-carrying colloid.  Anesthesiology 1999; 91: 1185-1187

Haemoglobin solutions and perfluorochemical (PFC) emulsions have been in clinical testing for almost 20 years. It was Clark in 1965 that did the initial work on PFC. These compounds consist of 8 - 10 carbons in which the hydrogen atoms have been replaced by fluorine resulting in a chemical liquid that is inert, very heavy, immiscible in water and liquids and at room temperature (20 o C) has approximately 20 times the solubility for oxygen as does water (there is a linear relationship between the oxygen partial pressure and the oxygen content vs. The sigmoid shape of the oxygen-dissociation curve of haemoglobin (Hb)).

Clark's initial experiments was on rats that he immersed in the PFC liquid (equilibrated with 100% 02) for 20 min and retrieved the animals alive.

Given intravenously though PFC emulsion causes liquid embolism and circulatory arrest. In 1967 Geyer produced microemulsions of PFC in normal saline. The emulsion droplets have 1/70 the size of RBC, and may therefore be able to perfuse and oxygenate tissue more effectively. When given intravenously though, the droplets are seen as foreign bodies and are quickly cleared from the circulation by the reticulo-endoethelial system (plasma half life approximately 12 Hrs), accumulating in the liver and spleen.  The laden macrophages are transported slowly to the lungs where they are breathed out unchanged in expiratory gases (body half life measured in months).

The study itself is a prospective, multinational, multicentered, randomized, controlled, single blind, parallel investigation on 147 patients, which aimed to evaluate the safety and efficacy of perflubron emulsion, (a 2nd generation PFC) in the setting of acute normovolaemic haemodilution, to reverse physiological transfusion triggers in patients undergoing orthopaedic surgery with significant blood loss.

After acute normovolaemic haemodilution (ANH) and when the patients reached any one of the prospectively defined "physiological transfusion triggers", they where randomised into one of four groups.

Transfusion triggers

  1. Tachycardia 
    • heart rate > 110/min or >125% of post ANH
  2. Hypotension
    • Mean arterial pressure < 60 mm Hg or < 75% of post ANH
  3. High cardiac output
    • > 150% Post ANH
  4. Low P V O2
    • < 38 mm Hg
  5. Low haemoglobin
    • <6g/dl
  6. End of surgery
    • Blood transfused to a haemoglobin of > 8g/dl

Randomisation groups

  1. 450 ml of autologous blood - The last unit from the normovolaemic haemodilution
  2. 450 ml of colloid.
  3. 0.9g/kg of perflubron emulsion mixed with a colloid to make up a volume of 450ml.
  4. 1.8g/kg of perflubron emulsion mixed with a colloid to make up a volume of 450ml.

They then looked at the ability of the instituted treatment to reverse these ‘transfusion triggers’ and for how long the measured parameter remained changed.  They concluded that there was a higher frequency of "reversal of transfusion trigger" and a longer duration of reversal of the trigger in the group transfused with 1.8g/kg of perflubron as compared to autologous blood transfusion group.

The researchers recorded no clinically significant increase in any adverse events/effects within the first 28 postoperative days.

Article 1: Perflubron emulsion delays blood transfusions in orthopedic surgery
Journal:  Anesthesiology 1999; 91: 1195-1208
Article type:  Clinical study
Authors:  Spahn, D.  van Brempt, R.   Theilmeier, G.  et al

2.  Amsorb .  A new carbon dioxide absorbent for use in anesthetic breathing systems

Although soda lime is still widely used , even after 80 yrs, a number of concerns still exist with respect to its use including :

  1. The production of compound A [a vinyl ether ] a breakdown product of sevoflurane that may be nephrotoxic.
  2. Carbon Monoxide production from especially desflurane, enflurane and isoflurane.

The cause of the degradation of the volatile anaesthetics is thought to be the hot monovalent hydroxide bases (sodium and potassium hydroxide) found in currently available soda lime compounds, something that AMSORB doesn’t contain.

The composition of AMSORB is Calcium hydroxide with a compatible humectant calcium chloride and two setting agents calcium sulphate and polyvinylpyrrolidine to improve hardness and porosity.

The new agent fulfills the United States pharmacopoeia xxii national formulary requirements that includes specifications for

  • size of granules
  • moisture absorption
  • hardness
  • CO 2 absorbant capacity
  • loss of wt or drying
  • packaging and storage and
  • labeling

The study compared AMSORB to two other soda lime preparations, Intersorb and Draegersorb with very encouraging results, including

  1. No compound A formation on exposure to very low flow sevoflurane in vitro
  2. No carbon monoxide production when the desiccated Amsorb is exposed to desflurane, enflurane or isoflurane
  3. Reasonable carbon dioxide absorbing capabilities [102 L/kg vs. 120L/kg]
  4. Lower operating temperatures 36.1(+/- 2.3) vs 38.0 (Soda Lime) and 39.1 (Bara Lime)
Article 2: Amsorb .  A new carbon dioxide absorbent for use in anesthetic breathing systems
Journal:  Anesthesiology 1999; 91: 1342-1348
Article type: Product evaluation
Authors: Murray, J.  Renfrew C.  Bedi A.   et al

3.  Supplemental oxygen reduces the incidence of postoperative nausea and vomiting

Post operative nausea and vomiting (ponv) still has an incidence of 20 - 70% despite new advances in anaesthetic drugs, techniques and antiemetics.

The study randomised patients (231) into two groups. One to receive  a 30%/70%-oxygen/nitrogen and the other to receive 80%/20%-oxygen/nitrogen gas mixture intraoperatively and for two hours postoperatively.

All other factors that are known to influence ponv were comparable in both groups and the anaesthetic technique, with regards to induction, maintenance and analgesia were standard for both groups.

A statistically significant result was achieved with an incidence of ponv of 30% in the  30% oxygen group, and only 17% in the 80% oxygen group (p = 0.027). Patients in the 80% oxygen group needed far less rescue antiemetic therapy.

In the discussion the authors of the study speculate as to the mechanism of the outcome concluding that probably decreased gut hypoxia with subsequent decreased 5-hydroxytryptamine release caused the difference.

They also mention other factors such as surgical stress decreasing GIT blood flow, raised intraabdominal pressure - reducing mucosal blood flow, splanchnic vasoconstriction due to altered thermoregulation, retractor positioning and intestinal mobilisations as other causes of intestinal hypoxia, which presumably would have been decreased in the group receiving 80% oxygen. From a physiological point of view this would make sense, but is purely speculative and probably needs proper documentation.

The authors conclude that although the precise mechanism for the decreased incidence of ponv in the group who received 80% oxygen is unknown, because oxygen is inexpensive and essentially risk free, that they would advocate the use of supplemental oxygen as an effective method to reduce ponv.

Article 3: Supplemental oxygen reduces the incidence of postoperative nausea and vomiting
Journal: Anesthesiology 1999; 91: 1246-1256
Article type: Clinical Study
Authors: Greif, R.  Laciny, S.  Rapf, B.   Hickle, R.  Sessler, D.

4.  Positiveend-expiratory pressure improves respiratory function in obese but not in normal subjects during anesthesia and paralysis.

The study aimed to look at the differences that the introduction of PEEP would make in obese (BMI > 40) and normal subjects (BMI < 25) with respect to the improvement of respiratory mechanics and gas exchange during anaesthesia.

The study measured, lung volumes (helium technique), elastances of the respiratory system, lungs and chest wall; pressure volume curves and the intraabdominal pressure.

It is felt that morbidly obese patients develop significant respiratory dysfunction, due to a large amount of atelectasis that causes morbidity and mortality. Interestingly enough no clear cut guidelines exist about the use of PEEP in normal subjects and it is felt that the indiscriminate application of PEEP, is not currently recommended during anesthesia and paralysis in normal subjects.

The study looked at 9 obese and 9 normal subjects all of whom were non-smokers and had no clinical evidence of cardio pulmonary disease.

Patients all received an identical "standard" anaesthetic.  All patients were mechanically ventilated with tidal volumes of 8-12ml/kg of ideal body weight with the  respiration rate adjusted to maintain PaCO 2 35-45 mmmHg.

Measurements were performed in supine position; without any dressing.

  1. Intraabdominal pressure --- measured using transurethral bladder catheter.
  2. End expiratory lung volume -measured using a closed circuit helium dilution method at 0cm and 10cm H 2 O of PEEP.
  3. Respiratory Mechanics - airway pressure, oesophageal pressure as well as elastance of total respiratory system, lung and chest wall were recorded, and pressure volume curves were derived.
  4. Gas exchange parameters were also analysed.

The authors then concluded from all the data that:

  1. Remarkable differences exist at PEEP 0cm H 2 O in lung volumes, respiratory mechanics and gas exchange in the early postoperative period during anaesthesia and paralysis, comparing normal and morbidly obese patients.
  2. The increase in PEEP to 10cm H 2 O led to significant improvement of respiratory mechanics and gas exchange in the obese patients but not in normal subjects.

The reason for this could be, that end expiratory lung volumes dramatically decreased during anaesthesia and paralysis in the obese patients. The author, however, felt that intraabdominal pressure alone was insufficient to cause the changes seen, and that other factors must also contributed to the deterioration seen.  These included blood shift from the abdomen to the thorax, rib cage distortion, cephalad shift of the diaphragm and surfactant alteration. The change in gas exchange in normal subjects was thought to be due to a combination of small airway collapse, V-Q mismatches and compression atelectasis with alveolar collapse and true shunt.   It is also likely that these mechanics are all operating at higher degrees in obese patients.

Increased PEEP only made a difference in the obese patient this was probably related to the alveolar recruitment of higher PEEP and it was felt that the negative effect of PEEP (ie. Decreased cardiac output) caused the failure to improve oxygenation in normal subjects.

I thought this was a good article worthwhile reading. The authors were aware of the limitations of the results they obtained and were reserved about drawing any conclusions without reasonable evidence.

Article 4: Positive end-expiratory pressure improves respiratory function in obese but not in normal subjects during anesthesia and paralysis.
Journal:  Anesthesiology 1999; 91: 1221-12231
Article type:  Clinical Study
Authors: Pelosi, P.  Ravagnan, I.  Giurati, G.   et al.


Editorial pointers

The authors of article 1 are offering us some hope in our ongoing quest to decrease the amount of blood transfused to a patient.  Transfusing blood has always been an emotionally charged intervention, with most of us trying to minimise the patient's exposure to blood that has been stored.  Very few anaesthesiologists appreciate having to give blood, only to see it emerge moments later from the patient and land on the floor.  We are all looking for safe and effective means to delay the actual start of the transfusion, in the hope that the total amount of blood transfused will be less and the physiological consequences of transfusing blood will be minimised.   The conclusion of this article that "Perflubron emulsion (1.8g/kg) combined with 100% oxygen ventilation is more effective than autologous blood or colloid infusion in reversing physiologic transfusion triggers" needs to be looked at closely and not taken at face value.
  1. Perflubron 1.8g/kg combined with 100% oxygen ventilation is more effective than autologous blood at reversing transfusion triggers
    • The only transfusion trigger that showed any clinically significant difference was the mixed PvO 2 .
    • The PFC 1.8 group had hyperoxic ventilation whereas the blood transfusion group were only exposed to 40% oxygen.  Hyperoxic ventilation has already been shown to rapidly improve tissue oxygenation and permit extended hemodilution to a haemoglobin of of 5.8 g per dl until the hyperoxic ventilation induced effects are lost.

Habler OP, Kleen MS, Hutter JW. et al. 
Effects of hyperoxic ventilation on hemodilution-induced changes in anesthetized dogs. 
Transfusion 1998 Feb;38(2):135-44

  • Comparison of the PFC 1.8 group to the colloid group, both of which received hyperoxic ventilation, was not commented on in the paper.  Looking at the only data available there did not appear to be any clinical difference between these two groups in terms of reversal of transfusion triggers
Transfusion Trigger Reversed Colloid PFC 0.9 group PFC 1.8 group Blood
Tachycardia 3/6 2/4 4/5 1/4
Hypotension 10/15 10/12 10/10 17/19
Increased cardiac output 2/3 0/2 1/2 1/2
Lowered PvO 2 22/23 22/24 26/26 12/22

Extract of Table 4.  Transfusion trigger reversal

  1. Perflubron 1.8g/kg significantly delays the onset of a second transfusion trigger
  Colloid PFC 0.9 group PFC 1.8 group Blood
Overall reversal of first transfusion trigger (%) 76 82 97 60
Duration of reversal of first transfuion trigger (min) 30(27-60) 59(40-90) 80(60-100) 55(30-70)

Extract of Table 4.  Transfusion trigger reversal

There appears to be about 40 minutes "bought" by the infusion of Perflubron 1.8g/kg compared to continued haemodilution with colloid infusion.   What needs to be explained is why there was such a poor response to autologous blood transfusion.

The article on Amsorb is a wonderful example of lateral thinking.  Rather than trying to redesign all our volatile anaesthetics, or trying to minimise the implications of toxic breakdown products, why not invent a new way of absorbing carbon dioxide.  Although many different ways of absorbing carbon dioxide do exist, the vast majority are not clinically useful in anaesthesia.  A good review of the topic can be found

Holloway, A.  Possible alternatives to Soda Lime. 
Anaesthesia and Intensive Care 1994; 22: 359-362

I certainly hope the manufacturers read and listen to the advice of the accompanying Editorial View

Kharash, E.  Putting the brakes on anesthetic breakdown.
Anesthesiology 1999; 91: 1192-1194

"From a patient-safety perspective, widespread adoption of a nondestructive carbon dioxide absorbent should be axiomatic.  Assuming a reasonable and only marginally increased cost over currently used absorbents, economic arguments against (Amsorb) should be moot" and its availability will actually change our clinical practice.

Article 3 is a lovely example of grabbing and using other peoples ideas, especially when they are unaware of the value in what they have said.   Prof D Sessler has taken the seemingly flippant remark from a letter to the editor...

Overdyk, F.  Roy, R.  If nitrous oxide induces emesis, maybe 100% oxygen is an anti-emetic
(letter) Anaesthesia and Analgesia 1997; 84: 228-236

And used this to spawn two other very interesting articles in conjuction with article 3.

Greif R, Akca O, Horn EP, Kurz A, Sessler D. 
Supplemental perioperative oxygen to reduce the incidence of surgical-wound infection.  
N Engl J Med 2000 Jan 20;342(3):161-7

The perioperative administration of supplemental oxygen is a practical method of reducing the incidence of surgical-wound infections.

Akca O, Podolsky A, Eisenhuber E, Panzer O, Hetz H, Lampl K, Lackner FX, Wittmann K, Grabenwoeger F, Kurz A, Schultz AM, Negishi C, Sessler D.  
Comparable postoperative pulmonary atelectasis in patients given 30% or 80% oxygen during and 2 hours after colon resection. 
Anesthesiology 1999 Oct;91(4):991-8

The authors conclude that administration of 80% oxygen in the perioperative period does not worsen lung function. Therefore, patients who may benefit from generous oxygen partial pressures should not be denied supplemental perioperative oxygen for fear of causing atelectasis



Old scopes logo