SAFETY OF EPIDURAL ANAESTHESIA Eleni Moka, Eriphili Argyra

23 Pages·2012·0.39 MB·English

SAFETY OF EPIDURAL ANAESTHESIA
Eleni Moka, Eriphili Argyra, Ioanna Siafaka

Introduction
Every patient wishes to receive anaesthesia care that is safe, in other words, free from risk, not
involving danger or mishap and guaranteed against failure. Anaesthesiologists usually present a
more realistic view to the patient. The personal view of the hoped – for care will be one in which
the clinical outcome is satisfactory and achieved without complications, since performance has not
deviated from the ideal. By this standard, most deviations are trivial or easily corrected by a perfect
process, whereas patient outcome and a reasonably stress-free life for the clinician are objectives
for all anaesthesiologists [1]. Safety of an anaesthetic technique is characterized by avoidance of
complications, minimal percentages of associated risks and numerous primary and secondary
beneficial endpoints, balanced against the inevitable consequences of method – related dangers.
Consequently, prior to any conclusion regarding safety, reliable evidence must be established for
both sides of the anaesthetic technique benefit-and-risk equation [2, 3].
Epidural anaesthesia – analgesia (EAA) has a long and distinguished history. For many clinicians, it
remains an attractive option and a leading anaesthetic – analgesic modality applied in the
perioperative environment [4, 5]. In regard to the benefit side, there is widespread conviction
among anaesthesiologists that EAA offers significant advantages in certain settings, especially those
involving abdominal and thoracic operative procedures [4, 6]. EAA is highly effective for controlling
acute pain after surgery or trauma to the chest, abdomen, pelvis or lower limbs, with its salutary
effects providing an added therapeutic benefit postoperatively. It has been instituted in various
subpopulations, including cardiothoracic, vascular, paediatric and obstetric patients, with optimistic
and promising results. Usually, the combination of excellent pain relief, associated with minimal
side – effects, results in high patient satisfaction, when compared with other methods of analgesia
[4, 6 – 8].
However, the controversy around EAA still continues, particularly with regard to its true impact on
postoperative morbidity and mortality, as well as its safety, mostly due to the fear of rare, but
potentially life – threatening or catastrophic complications [3, 6, 9]. Controversies represent the
debate and dialogue that ensue when clinicians examine the issue of best practice. Taking into
account the other (“darker”) side of the benefit – and – risk equation, unfortunately, the fear of
complications is still held with almost equal intensity, compared to the enthusiasm which initially
escorts EAA beneficial effects. Since there is always risk attached to all anaesthesia methods,
contemporary EAA practice in certain situations has been controversial. In addition, the main factor
which in the past turned our specialty away from EAA has been the difficulty in making a
reasonable risk/benefit analysis about the technique, resulting in clinicians constantly asking
whether epidurals are effective and whether the technique can finally be safe [2 – 4, 6, 9 – 12].
Nowadays, EAA is widely used again and since it is an invasive method, its benefit/risk ratio
deserves to be appraised, in order to help physicians estimate technique safety, make appropriate
choice among other opportunities (general anaesthesia, peripheral nerve blocks, wound infiltration)
and provide objective information to patients, prior to their written informed consent acquisition [5,
6, 13]. In this context, the “ultimate tool” of evidence – based medicine, the Level 1 evidence of
meta-analysis, has shown over the previous years that EAA decreases postoperative morbidity and
mortality [6, 15], as well as pulmonary complications [16], improves bowel recovery [17] and
reduces incidence of myocardial infarctions [18]. However, evidence to support improved outcome
from surgery remains elusive [19, 20], with the exception of aortic surgery, where mortality and
major complications were significantly reduced [21].
Therefore, it is not surprising that research and more or less scientific and emotive debates in this
area continue, even more so in view of reports of safety, complications and the current medicolegal
environment in many countries [5, 6]. EAA complications have been described by many authors,
although our understanding of the numerous risk factors is limited [2, 10 – 14]. Despite their
relatively infrequent occurrence, the fear of EAA complications exceeds their actual incidence [2, 3,
9, 13, 22]. Incidence knowledge could be an essential component of the clinical decision-making,
objective safety judging and consent processes, although there are few good and detailed data,
which can be quoted to support such discussions, leaving both patient and clinicians in quandary.
Figures (ranging from 1:1.000 to 1:100.000 in general) are quoted, but their doubtful validity
questions the ability to obtain genuinely informed consent from patients offered these procedures
[13]. Although it is impossible to prevent all EAA complications, it may be possible to reduce their
occurrence, by avoiding well-defined risk factors and by using a meticulous epidural technique at all
times. Further studies on clarification of risks, complications’ early recognition and on EAA
technique improvement may reduce adverse effects occurrence and improve EAA safety [13, 14,
22].
In the following paragraphs, the general objective is to focus on the clinical aspects of EAA risks in
terms of their incidence, recognition, prevention and management, in an effort to provide
information that will help clinicians to avoid or minimize complications, incurred during technique
application. In this context, EAA provision can be safe, as well as effective, if described risks and
benefits are carefully considered and the Hippocratic axiom “primum non nocere” is kept in mind.
Complications and Safety of EAA: Aetiology – Evaluation of Evidence
EAA complications represent the unexpected or untoward events following the application of the
technique itself or the exposure to anaesthetic agents. They can be classified as simple adverse
events and common or uncommon/rare complications. Some of them are considered minor side
effects, easy to prevent and to manage, that usually do not raise safety concerns. However, others,
resulting from EAA as unexpected outcomes, are of major concern, due to their potential to cause
neurological damage, leading to long-term or permanent disability. Both major and minor EAA
complications are well – known to clinicians and are also well-annotated. They can occur
immediately, or can be delayed. They can be clinically insignificant, or, in rare situations, life –
threatening. Knowledge and attention to the technique can reduce the likelihood for major
complications. The impact of EAA complications in anaesthesia practice is significant because
anaesthesia care is rarely therapeutic and known to carry risks. Furthermore, there is compelling
evidence that adverse outcomes trigger malpractice claims. As a group, anaesthesiologists seem to
accept the blame of complications that may or may not have been caused by EAA itself.
Furthermore, even when the standard of care has been met, juries tend to assign blame more
readily in EAA complicated cases [2, 9 – 14, 22].
EAA complications [2, 6, 9 – 12, 22] may happen in the whole perioperative period, including the
postoperative one and can be the direct or indirect result of the following:
- the technique itself: usually related to direct or indirect trauma by needle/catheter insertion
and/or catheter presence in the epidural space
- the untoward effects of local anaesthetics/adjuvant drugs instilled
- local and systemic toxicity results of drugs infused
- fatal drug errors
- untoward effects resulting from the anticipated or unanticipated physiologic consequences
of local anaesthetic blockade
- poor/late/no management of physiologic responses and/or adverse effects (arterial
hypotension – fatal cardiovascular collapse)
- intraoperative technical problems/equipment failure
- ignorance regarding complications’ anticipation/recognition of precipitating factors
- delayed complications’ diagnosis
- non – application/delayed application of complications’ preventive measures
- lack of experience/education/audit/monitoring
- human/behavioural/operator/patient factors
EAA complications have been recognized over 100 years ago and the database concerning such
risks is large but confusing. Studies are scarce, and their results difficult to compare. Due to their
rarity, definite studies on EAA risks remain problematic and the quoted ranges of complications for
severe adverse events still vary widely, because the study methodologies vary, much of the data are
retrospective, and the distinction between permanent and temporary disability is not always clear.
Many complications are known through case reports, and these rare events might not be evenly
distributed within the patient population. Retrospective observational studies/surveys and case
reports are important sources for evaluation, providing valuable information on incidence and
possible associations with confounding factors. Observational data are often clinically relevant, can
have a profound influence on clinical practice when a consensus of opinions is formulated, but,
unlike randomized controlled trials, lack a control group, making it difficult to calculate an accurate
incidence of risk. Because the enormous number of patients needed to perform prospective studies
exceeds feasibility, it is important that retrospective studies try to minimize the inherent weakness
of such study designs. Underreporting is common in retrospective studies, causing underestimation
of risk [2, 3, 6, 9, 10, 14, 22].
Complications and Safety of EAA: Incidence
Fortunately, literature data are reassuring, suggesting that EAA blocks carry a low incidence of
major complications, many of which may resolve in up to six months [13]. Anaesthesiologists’
vigilance has enabled early recognition of these diverse and, in some cases, extremely rare
complications, so that corrective action can be taken to prevent permanent harm. Nonetheless,
permanent disabling injuries following EAA are not so rare that we can ignore the issue. We are
obliged to raise the specter of permanent injury to patients in our discussions, despite its
uncommon incidence [2, 3, 6, 9, 14].
In the perioperative setting, overall permanent disability following EAA has been reported, in large
populations, with some of the best information available coming from Europe, US and Australia [5,
6, 13, 23 – 28]. Prior to examination of studies results, one should not forget the Poisson’s Law
regarding distribution of rare events. Actually, it is a statistical model for situations where the
probability of an event to occur is very law, but the opportunity for such an occurrence is very high.
According to this law, when an event has occurred in a sample size of “n” patients, the sample size
associated with a 95% probability to have a new occurrence is “3n”. If no adverse event occurred in
a population of “n” patients, it can only be conducted that the real incidence does not exceed “3 / n”
[5].
Moen V et al published a huge retrospective study involving 450.00 epidurals (including 200.000 in
obstetric cases) administered in Sweden in a ten – year period. Major EAA complications reported
were 1.37/10.000 patients (1/7.258 in general, 1/25.000 in obstetric population and 1/3.600 in all
other patient groups). The incidence of epidural haematoma was 1/22.000 (approximately
0.45/10.000), being lower in the obstetric patients (1/200.000) and much higher in patients subject
to knee arthroplasty (1/3.600) [23]. Auroy et al provided some information on this topic, through
two studies from France. Even though their data are prospective and large numbers are involved,
information is conflicting and studies were not randomized. After 30.413 epidurals performed, 19
serious events were noticed, including 6, 5 and 1 cases of neurological injury, radiculopathy and
paraplegia respectively, with the incidence of permanent neurological injury being calculated as
0.32/10.000 patients [24, 25]. According to Aromaa et al, based on the claims related to regional
anaeshesia, 9 serious complications were associated with the 170.000 EAA administered in Finland,
from 1987 – 1993 (0.52/10.000 patients) [26]. In the UK, Christie and McCabe retrospectively
recorded 12 major complications after 8.100 perioperative epidurals (1 in 675) in one hospital. This
approximates to 148/100.000 epidurals. As nine patients made a full recovery, permanent injury
was 3 in 8.100 (37/100 000, 95% CI 7.6 – 10.8) [29]. Cameron and colleagues reported similar
results, from a retrospective, single – hospital series, from Australia, with two vertebral canal
haematomas and six epidural abscesses following 8.210 “acute pain” epidurals. One laminectomy
was required and there were no cases of permanent neurological injury. Percentages of vertebral
canal haematoma (24/100.000, 95% CI 3 – 8.8), abscess (73/100.000, 95% CI 2.7 – 15.9),
laminectomy (12/100.000, 95% CI 1 – 6.8), and permanent neurological harm (0/100.000, 95% CI 0
– 4.5) are again broadly consistent with those from previous studies [30]. In a recent prospective
survey, conducted in UK, with denominator (procedures performed) and nominator (complications)
data validation through national databases, severe complications following 293.050 EAA have been
calculated in events/100.00 epidurals, with [95% CI] . Permanent injury after EAA is reported to be
4.2 [2.9 – 6.1]/100.000, with incidence of 17.4 [7.2 – 27.8]/100.000 and 0.6 [0 – 3.4]/100.000 in the
perioperative and obstetrical setting respectively [13].
Subgroup analysis from various studies has shown that elderly, female and orthopaedic patients
carry a considerably higher risk of untoward side-effects and are more exposed to complications,
compared to the obstetric or general population [13, 23, 27, 31]. Perioperative hypotension,
potential coagulation disorders, blood in the needle, pain on injection and difficult block may be
considered as alarm events [6]. Cardiovascular collapses [27] and wrong route drug injection errors
[32] have still to be considered. Severe neurologic events are more often reported in the
perioperative period and are mainly related to spinal haematoma (0.17 – 2.4/10.000), epidural
abscess (0.23 – 7.3/10.000) and permanent or transient traumatic neurological injury (0.17 –
2/10.000) [13, 23 – 30, 33 – 35]. Based on the lessons learned from the ASA Closed Claims Analysis,
haematoma, chemical injury and abscess represent 2%, 7% and 3% of spinal cord claims injury
respectively, with paraesthesia during needle insertion and/or drug injection and multiple attempts
to perform the block being the most important associated factors. These data also suggest that
nerve injury now surpasses brain damage as the second leading injury associated with anaesthesia
claims [2, 36]. In the following paragraphs, in regard with EAA safety only major complications
leading to potential disability will be analyzed.
Adverse Events due to Insertion of Needle/Catheter in the Epidural Space
Dural Puncture
Dural puncture occurs in 0.32 – 1.23% of epidural placements, potentially resulting in a post – dural
– puncture – headache (PDPH). Rarely, subdural haematoma, leading to neurological deterioration
has been described after dural puncture; its incidence may be less with loss of resistance to saline
than to air [2, 11, 12, 33, 37]. There is also a risk of pneumocephalus if air is used, which can result
in serious complications. The use of saline may again help to reduce the incidence of this and other
complications that have been associated with the use of air, notably spinal cord and nerve root
compression, postepidural paraesthesia and venous air embolism. In addition, accidental pleural
puncture during epidural catheter insertion has been described, as has haemothorax. There is
currently a large body of evidences suggesting that liquid must be used instead of air for loss of
resistance technique [11, 12, 38].
Direct Trauma
Direct trauma to the spinal cord, conus medullaris, spinal nerve roots or peripheral nerves ascribed
to the needle or catheter is extremely rare, but has been reported, followed by sensory loss and
less frequently motor deficits. As needle or catheter is advanced in the epidural space, intrinsic
spinal cord lesions may happen, possibly due to direct trauma during the procedure and
subsequent injection of fluid into the spinal cord, producing localized hydromyelia [2, 12, 39]. Auroy
et al found five cases of radiculopathy following 30.413 epidurals. In each of these patients, pain or
paraesthesia was noted on needle insertion or drug administration, with the radiculopathy being in
the same distribution as the associated paraesthesias, suggesting a traumatic mechanism [24, 25].
Pathology of the spine may be a risk factor and the impact of spinal stenosis (often asymptomatic)
has been recently pointed out, warranting further attention [5, 39, 40, 41]. Recent reports
demonstrate that either multimodal causes [42], or a preexisting pathology of the spine [43] may
be responsible for such complications, arising in temporal but not necessarily causal relationship to
EAA [6].
To avoid nerve trauma, careful technique and accurate anatomic knowledge are advised. Literature
reports highlight the problem and are fuelling the ongoing discussion and debate as to whether
patients should remain awake during EAA, to respond to painful stimuli, thus serving as possible
indicators of accidental cord trauma or unrecognized nerve injury. Epidural catheterization is most
frequently performed in the awake patient to avoid this risk of neurological damage and needle
advancement should be stopped if the patient complains of pain. In most adults, spinal cord
terminates at L1 vertebral body; however, in some it may terminate above or below this landmark.
The ability of the clinician to correctly identify lumbar spinous interspaces has been questioned,
using MRI. Only 29% of the interspaces were correctly identified, with 51% of clinicians being at a
higher vertebral level than anticipated and with the spinal cord terminating below L1 in 19% of
subjects. Oblique lateral entry into the ligamentum flavum may direct the needle into the dural cuff
region, resulting in potential nerve trauma and unisegmental paresthesia. This should alert the
clinician against persisting with further needle insertion or catheter threading [2, 4-6, 10 – 12, 22,
44, 45].
Traumatic injury when performing EAA also raises the question about technical skills. A learning
curve exists and manual skills improve with increasing experience. It is considered that residents
show significant improvement over baseline after 25 EAA, whereas at least 60 procedures have to
be performed before obtaining a 90% success rate [46]. Some new methods of training, such as
video technology and/or simulator, can be added to the available educational tools and would be
valuable for improving safety [47, 48]. Ultrasound-guided technique may help to teach and also to
perform EAA, especially when difficulties are awaited in specific population categories (obese
patients, parturients, scoliosis, hyperlordodis etc) [49, 50].
Additionally, many attempts have been made to improve techniques for epidural space localization.
The “membrane – in – syringe” technique, a modification of the loss of resistance technique,
combining loss of resistance to air and saline, allows reliable identification of epidural space,
keeping injection of saline into the space to a minimum [6, 51]. Another experimental innovation is
a device combining a visible and acoustic signal for epidural space identification [52]. Although
these techniques are in an early experimental stage, the simple, objective and reliable technique
for confirmation and accurate placement of an epidural catheter by low current electrical
stimulation has become widely discussed [53-55]. A meta-analysis of available studies, investigating
the ultrasound application as a diagnostic tool for epidural space visualization and its effects on EAA
quality and performance, demonstrated a clear advantage over the use of this imaging technique.
Regarding cost and practicability of these techniques, it has to be shown whether they will find
application in everyday practice [4, 6, 56].
However, one of the important and unanswered questions regarding the ultrasound use to guide
EAA is whether this technique will actually result in a lower incidence or severity of neurologic
complications, versus classical methods of epidural space identification. As with any newer
technique, there will be a learning curve when introducing ultrasound into clinician’s practice and
as such anaesthesiologists will need to be familiar with the anatomical landmarks and cognizant of
the potential artifacts and pitfall errors associated with ultrasound-guided regional anesthesia [4].
Transient Neuropathy
Transient neuropathy after EAA with eventual full recovery occurs more commonly, but is still
relatively infrequent; a recent large, prospective, multicentre series involving 30.413 epidurals
reported five cases of radiculopathy (0.016%), over 50% recovering completely within 3 months.
Results are similar to ones previously published in large studies on transient neuropathy: 4 out of
17.439 patients (0.023%) and 0.013% from a retrospective study of 1.304.214 epidurals. Smaller
studies report an incidence of 0.24 – 0.56%. After certain operations, such as tibial fracture fixation,
EAA has been implicated in a higher incidence of neurological complications. However, a
retrospective study demonstrated no significant association between peroneal nerve palsy
development after total knee replacement, with the use of postoperative EAA [2, 10, 12, 22, 23].
The management of transient or permanent postoperative neurologic sequelae requires the
cooperation of the anesthetist, surgeon, and neurologist. Additionally, the advice of the radiologist
and neurosurgeon may also be sought. Although it is easy to blame epidural presence for an
adverse neurologic outcome, it should be borne in mind that other factors can lead to
demonstrable nerve injury. These include undiagnosed preexisting neurologic disorders; ligation of
nutrient spinal cord vessels during abdominal surgery; injury to the femoral nerve during pelvic
surgery, or to the lateral cutaneous nerve of thigh during retraction close to the inguinal ligament;
or, pressure on the fibular head leading to neuroapraxia of the lateral popliteal nerve. If an adverse
outcome occurs, an attempt to localize the lesion by history and examination should be made.
Bilateral symptoms associated with pain should alert one to the possibility of neuraxial pathology.
Injury at the nerve roots affects both posterior and anterior rami. Preservation of sensation over
the paraspinous muscles suggests a more distal injury. Investigations should include blood cultures
and coagulation studies. Immediate MRI is the gold standard for out ruling central lesions.
Electromyography can be used to determine the site of injury and the degree of axonal loss,
although it can take up to 3 weeks after injury for changes to appear [2, 3, 4-6, 9-11].
Adverse Events due to Insertion/Presence of an Indwelling Catheter in the Epidural Space
Epidural Haematoma
Epidural vessels puncture during catheter placement occurs during 3 – 12% of attempts [12].
Bleeding from an epidural vein may occur during needle/catheter insertion, but is usually self-
limiting. However, the subsequent development of a spinal haematoma, defined as symptomatic
bleeding within the spinal neuraxis, which causes neurological damage, is a rare and potentially
catastrophic complication following EAA [10, 22]. Epidural haematoma often occurs spontaneously,
without any relationship with neuraxial anaesthesia. If not detected and treated early, it results in
irreversible paraplegia [10 – 13, 22].
The true incidence of clinically apparent epidural haematoma is unknown, as any study attempting
to quantify it would have to involve an enormous number of patients. The calculated incidence is
approximated to be about 1/150.000 cases of EAA [57]. Because this estimate represents the upper
limit of the 95% confidence interval, the actual frequency should be much less [58, 59]. However,
the series involved in these calculations were conducted before the implementation of routine
perioperative thromboprophylaxis and the risk may increase 15 – fold by concomitant use of
anticoagulant therapy, when appropriate precautions are not taken [60]. In this context, risk rate
may be underestimated, since complications frequency is mainly based on cases reported in the
literature. Recent reports have raised this risk to 1/100.000 after epidural labor analgesia,
1/150.000 in patients who were already receiving heparin or acetylsalicylic acid and 1/70.000 in
patients who had experienced a traumatic spinal tap [57 – 60].
In a study including 1.710.000 patients, the overall incidence of epidural haematomas was around
1/50.000 and increased to 1/3.600 when analysis was restricted to EAA with a catheter for total
knee replacement in women older than 70 years [23]. Other reports calculated the risk to be as
high as 1/3.000 in specific patient subpopulations [3, 9 – 13, 22 – 30, 61]. In a retrospective review,
three neuraxial haematomas were detected in 8.000 EAA, associated with epidural catheter, 2 days
after its insertion [29]. Schroeder et al estimated that spinal haematoma incidence in patients
undergoing EAA in combination with LMWH, in the United States, was 1/3.100 epidural injections
[60]. It is apparent that risk increases substantially in elderly women, in whom epidural catheters
are inserted, concomitantly with anticoagulant therapy and bloody puncture [59]. This high
prevalence could have a double explanation: the frequent dual therapy with antiplatelet agents and
antithrombotic drugs in orthopaedic patients and that, in the past, the majority of Anaesthesia
Society Guidelines that establish a time interval between the administration of the anticoagulant
and the performance of EAA had not been published [62, 63].
Recent case series and epidemiologic surveys suggest that the risk has increased, possibly as a
result of increased use of regional anaesthesia in combination with altered coagulation or of better
reporting of the complication. Overall, the risk of clinically significant bleeding is currently related
not only with concomitant drug administration, but also with procedure – related and additional
personal risk factors. It increases with age, associated abnormalities of the spinal cord or vertebral
column, presence of an underlying coagulopathy, difficulty during needle placement, and an
indwelling neuraxial catheter during sustained anticoagulation (particularly with standard heparin
or low-molecular weight heparin). The need for prompt diagnosis and optimized intervention is also
consistently reported [59, 62, 63].
New anticoagulant and antiplatelet drugs have been introduced recently, giving rise to new
challenges in the management of the anticoagulated patient undergoing EAA. EAA performance
could be considered safe in patients receiving drugs that alter haemostasis, provided there is
appropriate management based on safety intervals, suited to the anaesthetic-analgesic technique
to be carried out and to the characteristics of the anticoagulant. International recommendations for
thromboembolic prophylaxis and EAA application may help the physician to manage safely with
antithrombotic agents when EAA is foreseen. Appropriate guidelines have been prepared by a
number of national societies of anaesthesiologists, but they do not have universal acceptance [3, 9,
10, 12, 22, 59, 61]. The first national recommendations on neuraxial anaesthesia and
antithrombotic drugs were published by the German Society for Anaesthesiology and Intensive
Care in 1997 [64], followed by the American Society of Regional Anesthesia and Pain Medicine
(ASRA) in 1998 [65] and Belgian Anaesthesiologists in 2000 [66]. Since then, new anticoagulant
agents have been introduced and more information, regarding EAA risks with concurrent
anticoagulation, is available [59].
In response to such patient safety issues, the ASRA convened its Third Consensus Conference on
Regional Anesthesia and Anticoagulation. Practice guidelines and recommendations, published in
2010, summarize evidence-based reviews. However, the rarity of spinal hematoma defies a
prospective randomized study, and there is no current laboratory model. As a result, the ASRA
consensus statements represent the collective experience of recognized experts in the field of
neuraxial anesthesia and anticoagulation. These are based on case reports, clinical series,
pharmacology, hematology, and risk factors for surgical bleeding. An understanding of the
complexity of this issue is essential to patient management [62].
Additionally, in 2010, the European Society of Anaesthesiology (ESA) working party on Neuraxial
Anaesthesia and Anticoagulants, composed of academic physicians experienced in this topic,
published guidelines, to assist European anaesthesiologists in their daily clinical practice. The
introduction of new anticoagulants together with recent reports of stent thrombosis in patients
with perioperative cessation of antiplatelet drugs have considerably broadened the issue and made
revision necessary. To overcome deficiencies in content and applicability, the ESA has taken the
initiative to provide current and comprehensive guidelines for the continent as a whole, based on
extensive literature review [63].
Guidelines were designed to optimize both safety and efficacy of prophylaxis in the presence of
EAA. Recommendations and suggestions are drug-specific and usually based on the pharmacologic
profile (pharmacokinetics and pharmacodynamics) of each drug, mainly the time required to reach
maximal concentration, the time to reach maximal antithrombotic activity, the half-life and the
dose regimen. Two important factors also taken into account are how long to delay before
removing catheters and when to restart anticoagulation. The recommendations are usually relevant
to doses used for thrombosis prophylaxis, rather than therapeutic anticoagulation [59, 61 – 63].
All anticoagulants administered can be classified according to their specific target [59, 67 – 69] in
the coagulation pathway:
- Inhibitors of the initiation of coagulation: factor VIIa/tissue factor pathway inhibitors
- Inhibitors of propagation of coagulation: mainly factor Xa (FXa) inhibitors, either direct or
indirect
- Inhibitors of fibrin generation: direct and indirect thrombin inhibitors
- “Global” inhibitors of the coagulation pathway: antivitamin K drugs and unfractionated
heparin
Interestingly, recommendations for prevention of haemorrhagic complications associated with EAA
in patients given LMWH differ from country – to – country, and across continents. Indeed, in most
European countries, the recommendations are that placement or removal of a spinal or epidural
needle/catheter should be delayed at least 12 h after the last anticoagulant dose, but the
recommendation is to delay 20 h and 10h in France and USA respectively. Subsequent
administration of LMWH is not recommended until 4 h after catheter removal in Europe, but is
considered acceptable after only 2 h in the USA. In France, LMWH therapy is not initiated until 6 h
after surgery in patients having EAA [59, 70].
Recommendations related to EAA and antithrombotic agents vary according to patient
characteristics (age, weight, creatinine clearance and concomitant medications), difficulties
associated with needle puncture, and pharmacokinetics of the anticoagulants. Recommendations
and suggested strategies [59, 61 – 63, 68] can be summarized as follows:
 Consider the risk/benefit ratio of EAA for each patient. In general, outcomes sometimes
appear comparable between general and neuraxial anaesthesia.
 EAA in patients receiving full anticoagulation continues to be contraindicated.
 Concomitant administration of medications affecting haemostasis, such as antiplatelet
agents, NSAIDs, or dextran represents an additional risk of perioperative haemorrhagic
complications, including spinal haematoma.
 Appropriate neurological monitoring is essential during postoperative recovery period and
following catheter removal. The final decision to perform EAA in patients receiving drugs
that affect haemostasis has to be taken after careful assessment of individual risks and
benefits.
 Preoperative initiation of anticoagulation is not required for efficacy. When begun within
2 h of surgery, anticoagulation may increase major bleeding. For example, fondaparinux is
effective and not associated with an increased risk of major bleeding if started 8 ± 2h after
surgery; but it increases major bleeding without improving efficacy when the drug is started
within 6h after surgery.