Article Notes
Although there is some evidence of different efficacy among commonly used vasopressors, translating this to clinically-significant outcome differences is still uncertain.
Singh's 2020 Bayesian network meta-analysis is the most comprehensive study to date investigating this issue. The researchers concluded that norepinephrine, metaraminol, and mephentermine showed the lowest probability of adverse neonatal acid-base effects, and ephedrine showed the greatest.
Previously phenylephrine infusion has been the consensus recommendation.
Nonetheless, other than ephedrine which should not be a first-choice pressor during Caesarean section, there is not enough evidence to strongly recommend one pressor over another. Clinical familiarity and institutional availability are probably the most important factors when choosing a vasopressor.
Daniel Jolley
Why is this important?
Hypotension associated with spinal anaesthesia for Caesarean section is common. Increased interested over the past decade has resulted in some consensus recommending phenylephrine infusions, however there are few studies that directly compare this to other vasopressors.
What did Singh and team do?
By analysing 52 high-to-moderate quality RCTs and over 4,000 patients, Singh performed a Bayesian network meta-analysis to indirectly compare various vasopressors.
It's notable that umbilical artery base excess was used as the primary outcome, although other neonatal and maternal outcomes (nausea, vomiting, bradycardia) were secondarily assessed. Nonetheless, this study prioritised the fetal effects of hypotension management.
"We selected umbilical arterial BE as our primary outcome because it is thought to represent the effect of pronounced fetal hypoxaemia, anaerobic metabolism, and accumulation of non-volatile acids, that is the metabolic component of acidaemia."
Ok, what's a Bayesian network meta-analysis anyway?
A network meta-analysis compares trial interventions indirectly, when researchers are interested in a comparison between two factors (eg. use of metaraminol vs phenylephrine) that have not been directly compared by included RCTs (eg. a study comparing metaraminol vs ephedrine, and a study of phenylepherine vs ephedrine). A Bayesian NMA allows simultaneous comparison of multiple-arm trials, considering prior probability along with the likelihood of outcome rank between interventions.
A Bayesian NMA acknowledges the uncertainty of research conclusions and the probabilistic nature of clinical decision making.
Singh concluded...
Norepinephrine (noradrenaline), metaraminol, and mephentermine showed the lowest likelihood of adverse neonatal acid-base effects, and ephedrine the greatest.
"...norepinephrine, metaraminol, and mephentermine had the lowest probability of adversely affecting the fetal acid-base status as assessed by their effect on umbilical arterial base excess (probability rank order: norepinephrine > mephentermine > metaraminol > phenylephrine > ephedrine)."
When combined, there was a 66% probability that norepinephrine & mephentermine are the best agents for supporting umbilical a. BE.
There was a 66% probability that metaraminol is the best treatment for optimising umbilical artery pH, an 85% combined-probability that metaraminol & norepinephrine are best for umbilical a. pCO2, and 85% that they are the two best agents for avoiding maternal nausea and vomiting.
Be smart
Given the very nature of meta-analyses and the challenge of indirect comparison among agents from heterogenous studies, the conclusions are only suggestive of the benefits of phenylephrine alternatives. A large RCT is still needed! (And despite it's popularity in some countries, there are still only a small number of trials of metaraminol.)
Nevertheless, other than for bradycardia, ephedrine was most likely the worst for all outcomes, reinforcing past conclusions that there are better pressor choices.
Daniel Jolley
Recommendations from the guidelines:
- Hypotension following spinal or combined spinal-epidural anaesthesia at caesarean section causes both maternal and fetal/neonatal adverse effects.
- Hypotension is frequent – vasopressors should be used routinely and preferably prophylactically.
- α‐agonist drugs are the most appropriate agents to treat or prevent hypotension following spinal anaesthesia. Although those with a small amount of β‐agonist activity may have the best profile (noradrenaline (norepinephrine), metaraminol), phenylephrine is currently recommended due to the amount of supporting data. Single‐dilution techniques, and/or prefilled syringes should be considered.
- Left lateral uterine displacement and intravenous (i.v.) colloid pre‐loading or crystalloid coloading, should be used in addition to vasopressors.
- The aim should be to maintain systolic arterial pressure (SAP) at ≥ 90% of an accurate baseline obtained before spinal anaesthesia, and avoid a decrease to < 80% baseline. We recommend a variable rate prophylactic infusion of phenylephrine using a syringe pump. This should be started at 25–50 μg.min−1 immediately after the intrathecal local anaesthetic injection, and titrated to blood pressure and pulse rate. Top‐up boluses may be required.
- Maternal heart rate can be used as a surrogate for cardiac output if the latter is not being monitored; both tachycardia and bradycardia should be avoided.
- When using an α‐agonist as the first‐line vasopressor, small doses of ephedrine are suitable to manage SAP < 90% of baseline combined with a low heart rate. For bradycardia with hypotension, an anticholinergic drug (glycopyrronium (glycopyrrolate) or atropine) may be required. Adrenaline (epinephrine) should be used for circulatory collapse.
- The use of smart pumps and double (two drug) vasopressor infusions can lead to greater cardiovascular stability than that achieved with physician‐controlled infusions.
- Women with pre‐eclampsia develop less hypotension after spinal anaesthesia than healthy women. Abrupt decreases in blood pressure are undesirable because of the potential for decreased uteroplacental blood flow. A prophylactic vasopressor infusion may not be required but, if used, should be started at a lower rate than for healthy women.
- Women with cardiac disease should be assessed on an individual basis; some conditions are best managed with phenylephrine (an arterial constrictor without positive inotropic effect), whereas others respond best to ephedrine (producing positive inotropic and chronotropic effect).
Daniel Jolley
McCombe and Bogod report on their analysis of 55 medicolegal claims relating to obstetric neuraxial anaesthesia and analgesia.
Why is this important?
Not only is neurological injury the second most common reason for obstetric anaesthetic claims (behind inadequate regional anaesthesia resulting in pain during Caesarean section), the average claim cost is greater.
McCombe and Bogod provide a factful exploration of many of the causes of neurological complications.
Which themes emerged from their analysis?
- Consent, particularly around providing inadequate pre-procedure information of the risk of neurological injury1 and the challenges, medical and legal, to achieving informed consent.
- Nerve injury and it's mechanisms: non-anaesthetic causes2, direct trauma, chemical, and compression (abscess, haematoma).
- Complication recognition & management means timely follow-up and assessment, and maintaining a high index of suspicion for abnormalities. Remember the 4 hour rule: blocks should be regressing 4 hours after the last dose.
Important reminders
The level of spinal cord termination varies a lot among individuals, as does the level of Tuffier's line3. Considering the inaccuracy of spinal level identification by anaesthetists, there is a lot of potential to place a needle higher than expected.
Bottom-line: the intrathecal space should be accessed at the lowest possible level, and "the L2/3 interspace should not be an option."
And never allow chlorhexidine to contaminate gloves, the sterile workspace or neuraxial equipment.
-
Noting from NAP3 the risk of nerve injury ranges from, temporary injury 1:1,000, prolonged (>6 months) 1:13,000, to severe (including paralysis) 1:250,000. ↩
-
'Obstetric palsy' (pelvic nerve compression) estimated by Bromage as occurring in 1:3000 deliveries; arterial obstruction & ischaemia 1:15,000; AV malformations 1:20,000. A prospective French study found postpartum neuropathy in 0.3%, 84% were femoral, and 69% resolved at 6 weeks. ↩
-
Although generally accepted as being at the L4/5 interspace, in up to 50% of people the intercristal line might be at or above L2/3! ↩
Daniel Jolley
"Anaesthetists may feel justified in using a high lumbar interspace for spinal anaesthesia when it is condoned by so many texts, but more caution must be exercised. The L2/3 interspace should not be an option, and several textbooks need to be re-written to allow for human variability."
Daniel Jolley