Written by Dr C. Williams
The incidence and characteristics of 3-month mortality after intraoperative cardiac arrest in adults. Hur M, Lee H-C, Lee KH et al. Acta Anaesthesiologica Scandinavica 2017 doi:10.1111/aas.12955
Intraoperative cardiac arrest (IOCA) is thankfully a relatively uncommon event. In 2002 a study in the US reported rates of IOCA as high as 19.7 per 10,000 anaesthetic events. However more recent studies have given rates of 27-43 per 100,000. In contrast to most other in-hospital or out-of-hospital cardiac arrests IOCA generally occur in monitored patients and resuscitation is started immediately. Nevertheless IOCA is generally associated with poor clinical outcomes and a high in-hospital mortality rate (approximately 66%).
The majority of studies looking at IOCA have focused on the characteristics of the arrest or the optimal management. Very few studies have looked at clinical outcomes after IOCA. This study was conducted in Japan and looked at the records of almost 240,000 adult patients who underwent a surgical procedure under general anaesthetic from January 2005 to December 2014. 101 patients who suffered IOCA were identified but after exclusions 50 records were looked at. Excluded were brain dead organ donors and patients on cardiopulmonary bypass or ECMO. The primary outcome was 3-month mortality after IOCA.
The rate of IOCA in this study was 21 per 100,000 patients. Nineteen patients died in the operating room and a further 12 died within 3 months (mortality 62%). Three survivors had unfavourable neurological outcomes at 3 months. IOCA occurred most commonly during the maintenance phase of anaesthesia, followed by pre-induction. The most common initial cardiac rhythm seen was VF/VT. The most common cause of IOCA was pre-operative patient complications (42%), followed by surgical complications (36%) then anaesthetics related events (14%). In immediate non-survivors of IOCA, the incidences of emergency surgery, out of hours operating, pre-operative patient complications, need for intra or post arrest transfusion, were all higher. These patients also had higher total doses of adrenaline and a longer duration of cardiac compressions.
Although the overall incidence of IOCA has decreased, the outcomes remain largely unchanged compared to previous studies. It appears that IOCA in patients being operated on out-of hours results in a higher immediate mortality although not a higher unfavourable outcome in survivors. The exact cause for this is difficult to ascertain, mainly due to the fact that the number of IOCA are thankfully very small meaning that the numbers of patients included in the analysis for this study was only 50. Various other factors appear to be implicated as risk factors for immediate mortality after IOCA but no definite conclusions can be made. The ultimate conclusion is that a large-scale study into IOCA and risk factors associated with 3-month mortality is needed.
Postoperative complications in individuals aged 70 and over undergoing elective surgery for colorectal cancer. Colorectal disease 2017 doi:10.1111/codi.13821
Almost 60% of colorectal cancer cases are diagnosed in patients aged over 65-years. For non-metastatic disease surgery is the best management. In more advanced stages, surgery may be indicated for symptom relief such as obstruction, perforation or bleeding. It is known that elderly patients undergoing major surgery are higher risk. This study identified 190 patients between 2009-2015 undergoing colorectal resection aged over 70 years of age. Medical and surgical postoperative complications were reviewed and outcomes analysed comparing length of stay, critical care admission, 30-day readmission rates, 30-day and 1-year mortality.
97.9% of the patients were classified as ASA III or above with 60.5% assessed as fitting the criteria for frailty on geriatric preassessment. Medical postoperative complications occurred in 40.5% of patients whereas surgical postoperative complications occurred in 17.9%. The most common medical complication was infection with urinary tract infection, lower respiratory tract infections and catheter related sepsis being top of the list. The next most common category of medical complications was transient confusion or altered mental function. The most common surgical postoperative complications were surgical site infections and the need for reoperation.
The complications were classified using the Clavien-Dindo (CD) Classification system into CD grade 2 and CD grade 3 (more severe). CD grade 2 complications occurred in almost 40% and grade 3 in just over 10%. Unsurprisingly the more severe the complication the longer the postoperative length of stay – CD grade 2 resulted in a 114% increased postoperative length of stay, CD grade 3 by 162%. CD grade 2 complications did not significantly alter readmission or mortality (at 30 days or 1 year). However, grade 3 complications were associated with an increase in 1-year mortality rates. A complication of CD grade 2 or above increased the need for critical care admission.
Comparison of the results from this study to others looking at postoperative complications are difficult because many studies do not specify the definitions used for complications or derive data from ICD-9 coding. However, it is clear that complications can have a significant impact on the perioperative surgical course. Medical complications are more common therefore surgeons should have medical expertise concerning the most common medical postoperative complications. For vulnerable, multimorbid older patients collaboration with other medical specialties and a multi-disciplinary approach is of paramount importance to provide the best overall care for this group of patients.
Impact of reversal strategies on the incidence of postoperative residual paralysis after rocuronium relaxation without neuromuscular monitoring: a partially randomised placebo controlled trial. Nemes R, Fülesdi B, Pongrácz A et al. European Journal of Anaesthesiology 2017;34(9):609-616 doi:10.1097/EJA.0000000000000585
There has been many recent papers looking at postoperative residual paralysis and the impact it has on postoperative recovery. The evidence indicated that there is a significant incidence of residual paralysis particularly when neuromuscular function is not monitored. Many anaesthetists argue that they can ensure adequate recovery of neuromuscular function without the need for a nerve stimulator and without having to give neostigmine. There has also been some concern about the use of neostigmine due to its undesirable muscarinic side-effects and its limited ability to reverse even a moderate neuromuscular block. Sugammadex is now well established as an alternative to neostigmine for the reversal of a neuromuscular block by encapsulation of the steroidal neuromuscular blockers. The cost of sugammadex compared to neostigmine is significant meaning that in several institutions its use is closely controlled. Studies published studying the reversal of rocuronium with sugammadex did not find any postoperative residual neuromuscular block in the first 60 minutes compared to a significant proportion of patients given neostigmine.
This was a partially randomised, placebo controlled, double-blind study aiming to investigate the incidence of postoperative residual neuromuscular block using acceleromyography after spontaneous recovery of rocuronium-induced block, compared to administration of sugammadex, neostigmine or a placebo. Patients undergoing surgery were given a general anaesthetic with propfol, fentanyl, sevoflurance and rocuronium. Neuromuscular block was measured by acceleromyography. If the anaesthetist deemed that pharmacological reversal was needed the patient was then randomly allocated to receive sugammadex (2mg/kg), neostigmine (0.05mg/kg) or a placebo. In recovery an independent anaesthetist blinded to the treatment given then reassessed the neuromuscular function using acceleromyography. The main outcome measure was a train-of-four measurement of less than 0.9 on arrival in recovery.
The results indicated that pharmacological reversal was more effective than spontaneous recovery of a neuromuscular block. No agent was 100% effective but there were less incidences of a residual block in patients given sugammadex than those given neostigmine. What was highlighted by this study was that the residual block in patients allowed to spontaneously recovery and those deemed to need pharmacological reversal but randomised to receive a placebo was no different. This would indicate that clinically estimating who should need reversal or not may not be the best method. Studies have indicated that the incidence of a clinically relevant residual block by anaesthetists is less than 1%. This study demonstrated a significant residual postoperative neuromuscular weakness requiring rescue treatment in 8.8% of patients. Although it could be argued that the higher rates may have been detected due to a higher level of monitoring and awareness during the study, this highlights the risk of unmonitored muscle relaxant use.
There are limitations to this study the main ones being very small groups of patients and only rocuronium was studied. Yet overall, this study serves as a reminder that neuromuscular blocks should be monitored and reversed adequately. Other studies have shown that inadequate reversal is implicated in the development of respiratory postoperative complications. The incidence of a residual block after sugammadex was significantly lower than neostigmine. The sharp reality in todays NHS is that the cost of sugammadex is a limiting factor and as such it should maybe be reserved for patients at particular risk of undesirable effects of a residual neuromuscular block.
Angiotensin II for the treatment of vasodilatory shock. Khanna A, English S, Wang X et al for the ATHOS-3 Investigators. New England Journal of Medicine 2017;377:419-430 doi:10.1056/NEJMoa1704154
Vasodilatory shock, the most common type of shock, is characterised by peripheral vasodilation and reduced blood pressure despite preserved cardiac output. Patients with severe vasodilation who have hypotension despite the use of high doses of vasopressors have a poor prognosis with a 30-day all cause mortality of over 50%. This was a randomised, double-blind, placebo controlled phase 3 trial to determine whether the addition of angiotensin II to background vasopressors would improve blood pressure in patients with catecholamine-resistant vasodilatory shock.
Vasodilatory shock requires prompt treatment to ensure organ perfusion through the reestablishment of adequate blood pressure while the underlying cause of the shock is identified and treatment. Vasopressors are used when intravenous fluid resuscitation alone fails to restore blood pressure.
Treatment available for patients with catecholamine-resistant vasodilatory shock are limited and the available treatments often have associated side-effects. Options include glucocorticoids, vasopressin, methylene blue and high-volume hemofiltration. New therapies have so far proved to be disappointing. The nitric oxide inhibitor 546C88 increased blood pressure in patients with septic shock but was associated with more frequent cardiovascular side-effects and increased mortality at 28-days.
In the ATHOS-3 trial (Angiotensin II for the Treatment of High-Output Shock) the primary endpoint was the response of mean arterial blood pressure (MAP) at hour 3 with a response either being a MAP >75mmHg or an increase above baseline of at least 10mmHg without an increase in vasopressor dose. Patients were included if they had vasodilatory shock unresponsive to fluid resuscitation (defined as at least 25ml/kg of body weight over the previous 24 hours) and high dose vasopressors (defined as 0.2mcg/kg/min of noradrenaline or equivalent). Patients were randomly assigned to either receive synthetic human angiotensin II or saline placebo. Significantly more patients given angiotensin II met the primary end-point criteria. In fact, for those given angiotensin II the MAP increased rapidly and the dose of both angiotensin II and vasopressors were able to be reduced. The rate of adverse events (specifically tachyarrhythmias, distal ischaemia, ventricular tachycardia and atrial fibrillation) were similar in the both the angiotensin II and the placebo groups.
The study had a relatively small sample size so the possibility of clinically important side-effects related to angiotensin II cannot be excluded. Also follow-up was only for 28-days so either beneficial or harmful long-term effects of angiotensin II therapy cannot be excluded. Larger trials of longer duration are needed to answer these questions and to directly compare angiotensin II with other vasopressors.
ICU admission after surgery: who benefits? Ghaffar S, Pearse R, Gillies M. Current opinion in critical care 2017;23(5):424-429 doi:10.1097/MCC.0000000000000448
The number of operations performed each year continues to grow and combined with an aging population, more and more procedures are performed on patients who previously may not have been considered for surgery. The last decade has also witnessed an increased recognition of the importance of and improvements in perioperative care. Overall mortality rates after surgery are low at 0.5% but complication rates are higher with the global figure suggested to be approximately 16% with complications being associated with an increase in mortality.
The incidence of adverse events will obviously vary by surgery, institution, region and nation. Differences in the process of care are often given as a cause in particular the access to critical care beds. Admission to ICU has been a standard after certain types of surgery for many years, however, ICU resources particularly in the UK are limited and expensive. Having the ability to identify those patients that would benefit the most from critical care admission is a major issue for those delivering perioperative care. The ability and need to identify high-risk patients, develop pathways to signpost to the most appropriate postoperative care placement and methods to allow as early detection as possible of the deteriorating patient are all vital to improve outcomes and best utilise finite resources.
Evidence from large epidemiological studies suggest that for many patients undergoing major elective surgery routine ICU admission may not provide any additional benefit and may contribute to an increase length of hospitalisation and cost. The evidence also suggests that for patients undergoing major emergency surgery where there is insufficient time to optimise comorbidities or where there is any other major physiological derangement are best managed in a critical care setting.
A study published in 2013 found that ICU bed provision and rate of admission did not correlate with overall mortality. In fact, countries that had the lowest mortality rates after surgery also had the lowest rates of ICU admission. An explanation for this could be the use of alternatives to ICU admission in some of the best performing European countries. Patients recovering from major elective noncardiac surgery require prompt and effective treatment of pain, hypothermia, cardiorespiratory compromise, fluid imbalance combined with early mobilisation and enteral nutrition. The main way that ICU provides a benefit in these scenarios is the access to high quality nursing care. This can be delivered in less intensive environments such as post anaesthetic care units or specialist wards. In this way patients will still receive a higher level of nursing care along with early identification of deterioration and rapid access to relevant services such as critical care outreach.
Another consideration is that of the use of ‘fast-track’ pathways in cardiac surgery for low to moderate risk patients. They have been established for many years and appear to be safe and shorten the ICU stay without increasing complication rates. Maybe this is an area that needs further exploration for noncardiac major elective surgery.
Clearly the benefits of ICU admission for many types of major surgery remain uncertain. Identifying the patients most at risk of death and complications following surgery remains the major challenge for perioperative care. As pointed out in this review, future research should focus on how postoperative care can best be structured to provide patient optimum care within the available resources.
Risk factors for peripheral nerve injuries following neuraxial labour analgesia: a nested case-control study. Haller G, Pichon I, Gay F-O, Savoldelli G. Acta Anaesthesiologica Scandinavica 2017 doi:10.1111/aas.12951
Neuraxial anaesthesia is the most common method for either labour analgesia or anaesthesia for delivery. Post-partum lower extremity motor and sensory dysfunctions occur in between 0.1-9.2% of deliveries and are a well known complication in obstetrics. Usually these symptoms are transient and resolve spontaneously within a year. For a small minority the damage is permanent. While the damage is easily identified the causation may be more difficult. Anecdotally and personal experience is that any type of neurological problem tends to be blamed on the neuraxial anaesthesia. Although most published data indicates that these complications are related to compression of the nerve roots, plexus and peripheral nerves due to obstetric factors (femoral nerve, lateral femoral cutaneous nerve, obturator nerve, pudendal nerve and common peroneal nerve). There are other studies that associate non-obstetric-related factors with peripheral nerve injuries mainly radiculopathies or catheter injuries.
The most feared but fortunately rare complication of neruaxial anaesthesia is an injury to the central nervous system. This paper quotes the rate as 1:145,000 to 1:240,000. This is comparable to the 2009 Royal College of Anaesthetists NAP3 report that quotes the cases of permanent harm relating to central neuraxial block in obstetrics as 1:80,000 to 1:320,000.
This study collected data from almost 20,000 obstetric patients having neuraxial procedures in the Maternity Department of Geneva University Hospital, a tertiary referral centre where over 80% of deliveries are performed under neuraxial anaesthesia. 19 patients (0.96%) were found to have peripheral nerve injuries. According to neurologists 15 of these cases were likely related to compression or tractions by the baby’s head or obstetric manoeuvres/instrumentation. In four cases a nerve root injury due to the Tuohy needle were suspected. Other risk factors for nerve injury included a gestational age >41weeks, late initiation of neuraxial anaesthesia, repeated anaesthetic procedures, assisted delivery with forceps and a newborn birth weight of >3.5kg.
This paper highlights that although patient and anaesthetic factors contribute to peripheral nerve injuries, obstetric related factors are the most prominent risk. Thankfully overall motor and sensory dysfcuntions following delivery and neuraxial anaesthesia are rare and typically self-limiting. Despite this fact the risk of nerve injury by a Tuohy needle or catheter insertion should not be minimised particularly as symptoms can persist for a long length of time and have a significant impact on the patient’s quality of life.
Early definitive treatment rate as a quality indicator of care in acute gallstone pancreatitis. Green R, Charman S, Palser T. British Journal of Surgery 2017 doi:10.1002/bjs.10578
The incidence of acute pancreatitis is rising with it now being one of the most common intra-abdominal emergency conditions. The most common aetiology is gallstones with them being implicated in 35 – 40% of pancreatitis cases. The available evidence and current guidelines recommend that patients with mild gallstone pancreatitis should have definitive treatment of their gallstones during the same hospital admission or within two weeks of discharge. Definitive treatment is namely a cholecystectomy or if unfit for surgery an endoscopic sphincterotomy. Adherence to these recommendations improves patient outcomes and reduces hospital length of stay and the risk of further episodes of pancreatitis. Compliance with guidelines is variable. This study aimed to examine variation in patients receiving early definitive treatment for gallstones following an episode of acute gallstone pancreatitis and determine its validity as an indicator of quality of care.
Using information from the Hospital Episode Statistics database just over 19,500 patients were identified. Of the patients only 6733 received early definitive treatment within two weeks of discharge. In total 28.5% received treatment during their admission and a further 6% had treatment within two weeks of discharge. In the 1-year follow up 23.9% of patients had one or more emergency admissions for gallstone pancreatitis related complications. 33.8% were readmitted within 2 weeks with the remaining 66.2% of readmissions occurring after the point at which the patient should have had early definitive treatment. Early treatment was associated with a 39% reduction in readmissions.
This study seems to support the idea that early definitive treatment for gallstone pancreatitis is associated with improved patient outcome. As well as improving patient morbidity and mortality the reduction in readmission rates with early treatment is associated with potential cost savings. This issue was touched on in the August journal watch.* This study provides further evidence to suggest that maybe surgeons and anaesthetists alike should take ‘hot’ cholecystectomies seriously.
* Improving care for patients with pancreatitis. Siriwardena AK, O’Reilly DA. British Journal of Surgery 2017 doi:10.1002/bjs.10585
How to better identify patients at high risk of postoperative complications? Talmor D, Kelly B. Current opinion in critical care 2017;23(5):417-423 doi:10.1097/MCC.0000000000000445
As already discussed more patients are presenting for surgery at a later age with more co-morbidities. The challenge is how to identify the patients most at risk of postoperative complications. The aim of preoperative risk assessment is to help identify patients at increased risk of complications and allow appropriate postoperative management to be planned, ultimately to lead to improved patient outcomes.
There are a multitude of prediction scoring systems that can be used – covering them all in this short synopsis is impossible. However, most prediction scores predict postoperative mortality with, at best, moderate accuracy. Score incorporating surgery-specific and intraoperative events may improve the accuracy of traditional scores. The more ‘traditional’ risk factors of increased ASA score, emergency surgery, intraoperative blood loss and haemodynamic instability are consistently associated with increased mortality using most scoring systems.
Preoperative clinical risk indices and risk calculators estimate surgical risk with moderate accuracy. Surgery-specific risk factors are helpful in identifying patients at increased risk of 30-day mortality. Particular attention should be paid to intraoperative haemodynamic instability, blood loss, the extent of the surgical incision and the volume of resection in prioritising patient admission to higher levels of postoperative care.
Ultimately the prediction of postoperative complications remains difficult. Independent risk factors for worse outcomes include increasing age, frailty, chronic renal failure and poor cardio-respiratory reserve despite having at best only moderate discriminatory value. Many anaesthetists still appear to feel that risk scoring is a waste of time. It is an area that is developing but at the moment the risk scoring systems we have are the best that are currently available. There is no doubt that they only give predictions and no matter how good a scoring system is, it will never and can never be 100% accurate. But in a healthcare system where resources and critical care beds not infinite they can be useful in trying to prioritise which patients need higher levels of care postoperatively.