Case
A long time ago, a 2 8/12-year-old girl, 14 kg, was scheduled for full-body scintigraphy because of suspected osteomyelitis in the presence of recurrent febrile episodes. The intention was to induce sedation with propofol 15 mg (1 mg/kg) and ketamine 10 mg (0.7 mg/kg). The anaesthetist injected 1.5 ml of milky solution, which was propofol, and 1 ml from a syringe with a green label, typical for racemic ketamine at the time.
Shortly after the injection, gurgling respiration and upper airway obstruction occurred, followed by apnoea. For the skilled anaesthetist, mask ventilation was easy and uneventful, but soon afterwards a reddish skin rash became visible. Because the whole episode was unclear for the team, the author was informed. Apnoea and skin rash are typical signs of a rapid injection of atracurium, and because at least two similar cases had occurred in his own practice, the diagnosis of a confusion between atracurium and ketamine (both with a green label) was made by phone.
A laryngeal mask airway (LMA) size 2 was inserted and the child was ventilated with oxygen and nitrous oxide; two additional doses of propofol were given for sedation. At the end of the examination the residual neuromuscular block was antagonized with neostigmine 50 µg/kg combined with glycopyrrolate 10 µg/kg. The girl made a completely uneventful recovery and the parents were informed about the event.
Discussion
This case illustrates a classic medication error, confounding two compounds with a labelling of similar colour (‘lookalikes’). Many of the recognized medication errors concern muscle relaxants, because the apnoea makes it immediately obvious that something has gone wrong. In contrast, other errors are often initially overlooked, e.g. profound sedation caused by an incorrect concentration of a hypnotic (S-ketamine 25 mg/ml instead of 5 mg/ml), tachycardia because of ephedrine instead of thiopental 10 mg/ml (both in a 5 ml syringe), increased diuresis because of furosemide instead of metamizole (lookalike vials). During his professional career the author witnessed or himself made many medication errors.
A major step forward was the introduction of colour-coded labelling in many but not all countries (Balzer et al. 2012). Muscle relaxants now have a red label (Fig. 6.1), and confusion between two different relaxants has not had the same impact as the administration of a relaxant instead of a hypnotic agent, as in the presented case.
Figure 6.1 Colour-coded labels have greatly increased safety, but thiopental is still occasionally injected instead of S-ketamine.
The institution has to be made resilient against errors, which will always occur. The most catastrophic medication errors are flushing an intravenous line with potassium chloride instead of normal saline. The use of the wrong concentration of a local anaesthetic, e.g. ropivacaine 0.75% instead of 0.2% for a caudal block, can induce a toxic reaction (Hübler et al. 2010). In oncology patients, the deadly intrathecal injection of vincristine instead of methotrexate is an ever-present threat (Liu et al. 2017). Special precautions have to be taken to avoid these disasters and to make the working surrounding resilient against mistakes by an individual staff member. For example, in the children’s hospital in Lucerne no potassium vials are kept in stock in the operating theatres; if needed, potassium has to be ordered from the ward. Only ropivacaine 0.2% is available in the children’s hospital, and levobupivacaine 0.5% has to be used for more dense blocks. The author is convinced that in paediatric anaesthesia a muscle relaxant has always to be ready in a syringe for immediate injection. This means that it needs to be stored on the anaesthesia trolley in a dedicated place in order to avoid confusion with other medication (Martin et al. 2017). For neuraxial anaesthesia and in oncology, non-Luer connectors are intended to be introduced, in order to avoid the intrathecal injection of a compound exclusively suited for intravenous injection (Lawton et al. 2009).
Special attention is needed when verbal orders are given to staff not necessarily familiar with the medication they are told to inject. The culture of closed-loop communication, to repeat what was heard, has to be both taught and practised (Davis et al. 2017).
Dosing errors are a typical issue in paediatrics. Because of the hundredfold difference in body weight between a preterm baby and an adolescent, no single dose size can be remembered by the practitioner. In adults half to one ampoule of many drugs may be correct, but in paediatrics the dose has to be calculated for every individual child. Electronic prescribing systems have the potential to reduce such errors on the ward and in intensive care settings (Jani et al. 2010), and at the children’s hospital such an electronic system has been running successfully for almost a decade. However, they are probably not suitable for the busy surroundings of the operating room. Dosing errors by a factor of 10 do occur; this has been reported to occur once a week in a large paediatric institution (Doherty & McDonnell 2012). The author has twice observed nalbuphine dosed at 10 mg instead of 1 mg; fortunately, nalbuphine is a compound with a large margin of safety and high doses are well tolerated even by small babies (Schultz-Machata et al. 2014).
Summary and Recommendations
This case illustrates a medication error, confusing a relaxant and a hypnotic compound, which probably occurred because of the similar colour of the labels.
The introduction of colour coding has reduced the severity of confusion, because a drug of the same class is administered. However, continuous care and clear communication remain the cornerstones of safe practice.
The anaesthesia workplace has to be made resilient against mistakes. Confusions with immediately lethal consequences, e.g. potassium instead of normal saline for flushing, are best avoided by physically eliminating certain compounds from the busy theatre workplace.
References
Case
A 7-month-old girl, weighing 8 kg, was scheduled for surgical release of a tethered cord. The paediatric anaesthetist went to the neurosurgical unit. Anaesthesia was induced by mask; after achieving venous access, nasotracheal intubation was performed. The further instrumentation included an additional venous access; an arterial line and a bladder catheter were inserted. During this preparation phase and the placement of the leads for monitoring the motor and sensory evoked potentials, haemodynamic stability was well maintained.
In order to enhance the registration of evoked potentials, the decision was made to switch from the inhalational to a total intravenous technique. The anaesthetist instructed the assistant to start the propofol perfusor at a rate of 80 mg per hour (10 mg/kg/h) and sevoflurane was switched off. In addition, remifentanil was started at the rate of 0.1 µg/kg/min. A few minutes later mean arterial pressure had dropped close to 30 mmHg, and the BIS monitor showed a flat EEG tracing which predominantly showed the electrocardiographic spikes of the QRS complex (Fig. 6.2). At this point it was noticed that the propofol perfusor was running at a rate of 80 ml per hour instead of 8 ml per hour, which would have corresponded to the desired 80 mg per hour. Propofol administration was stopped for more than 20 minutes until the EEG had recovered from isoelectric to regular activity. The further course was uneventful.
Figure 6.2 An EEG tracing showing an isoelectric EEG in a child; this is not obvious at first glance, but when the ECG signal is visible in the EEG, this is usually the case.
Discussion
This is a typical failure in communication leading to a 10-fold error in dosing propofol (Doherty & McDonnell 2012). It is part of a good safety culture to dose medication always in milligrams (mg) or micrograms (µg) and not in millilitres (ml). Even more correct would have been to say ‘80 milligrams per hour, this equals 8 millilitres per hour’. In addition, a closed-loop communication is recommended, which means that the recipient repeats the information as heard, which is then confirmed. Medication errors, especially errors in dosing, are much more common in paediatric practice than in adults because there is no ‘usual dose’ in view of the body weight varying from 500 g to 130 kg (Kaufmann et al. 2017). In adults one or half an ampoule rarely causes major harm if it is not a vial designated to be admixed to an infusion: in children this is definitively not the case.
This neurosurgical intervention was performed outside the paediatric hospital, and therefore the paediatric anaesthetist was working in strange surroundings with personnel not used to paediatric patients. In such a situation, good team performance cannot be taken for granted, and a clear closed-loop communication style, always reconfirming the message, is even more important than it is in the paediatric hospital, where everybody is familiar with paediatric doses and nobody would infuse 100 mg/kg/h of propofol to a child. The same problem occurs when new and unfamiliar drugs are used: the author remembers a very similar event with a rapidly administered overdose of dexmedetomidine, 20 ml instead of 20 µg, which led to profound bradycardia and hypertension in a sick child.
Dosing errors are probably more common with total intravenous anaesthesia than with inhalational anaesthesia, especially when using ordinary infusion pumps for the propofol administration (Pandit et al. 2014). Pumps are switched off for transfer and erroneously not restarted in theatre, which can lead to unwanted intraoperative awareness; or, as in this case, a 10-fold overdose can happen. Very likely the use of target-controlled infusion (TCI) would have avoided this event. Although the Paedfusor system is only recommended in children over 1 year of age, it can be used in infants too, and, in the author’s opinion, it is advantageous because it is still safer than a manually driven pump. TCI increases safety because it eliminates the errors that can typically be made in manual systems (Schnider et al. 2016). It provides a constant plasma level of propofol by administering a bolus followed by a gradually decreasing infusion. The TCI models are not very precise in predicting the exact height of the plasma concentration and they do not tell us at all if this concentration corresponds to the clinical needs of the patient. Nevertheless, a manually driven infusion never performs better than a TCI pump. In the author’s hands the Schnider model (Schnider et al. 1998) is well suited to children older than 5 years (Rigouzzo et al. 2010); below this age the Paedfusor (Marsh et al. 1991) or the Kataria (Kataria et al. 1994) model should be used.
EEG-based monitoring, e.g. BIS or Narcotrend, should be used whenever possible during total intravenous anaesthesia in order to have some kind of pharmacodynamic feedback (Louvet et al. 2016). The primary goal is not avoidance of awareness, but optimizing dosing; in this case a flat EEG first raised the suspicions of the anaesthetist. Both a massive overdose leading to a flat EEG, and a failing administration because of paravenous infusion or a failing pump can be detected with the aid of these devices – even though, in children below 1 year of age, the EEG cannot be reliably used to guide hypnosis during general anaesthesia (Hayashi et al. 2012).
Summary and Recommendations
This case of a 10-fold overdose of propofol during total intravenous anaesthesia illustrates the importance of clear communication.
Target-controlled infusion (TCI) increases safety because typical errors are avoided.
During total intravenous anaesthesia, EEG-based monitoring is helpful to recognize an overdose or a failing drug administration.
As part of safety culture, the dose of a medication should always be quoted in weight units (µg, mg or g) and additionally in volumes (ml).
References
Case
A school class visited the children’s hospital in order to learn a bit about medicine and the hospital environment. The pupils, aged 7–9 years, had a look at some surgical instruments, they were shown how a cast was put on to one of them, and finally they were shown an anaesthesia induction room. An anaesthetist and an anaesthetic nurse demonstrated the anaesthesia machine and the monitoring devices. The most adventurous girl placed herself on the operating table, so that the monitoring could be demonstrated: a blood pressure cuff was put on her arm, and the ECG and pulse oximeter probe were placed. By now she already looked pretty nervous. A flavoured face mask was given to her to be placed over her mouth and nose with the aim of demonstrating the capnographic tracing. She got even more nervous and started to breathe deeply, showed jerky movements, and finally became unresponsive. The anaesthetist tried to calm her, because he was convinced that what he was observing was a hyperventilation episode in a young girl. He placed her in lateral position on a stretcher and explained the episode to the upset class and the anxious teacher. Within 10 minutes the girl made an uneventful recovery without nausea and vomiting. The school class left the children’s hospital, the girl with the stigma of being a little bit special.
Later in the afternoon the sevoflurane vaporizer was found to be set at 4 V% (Fig. 6.3). So the girl had just accidentally received a sevoflurane anaesthetic.
Figure 6.3 The vaporizer is left open, and instead of pure oxygen the child receives 4% sevoflurane.
Discussion
This case of inadvertent drug administration is just a spectacular presentation of a common phenomenon. The vaporizer is still set at high concentrations of the inhalational agents, whereas the fresh gas is turned off. The next user just turns on the fresh gas and places the mask on the patient’s face, e.g. in order to preoxygenate. The author remembers numerous cases occurring at induction of anaesthesia, when the child rejected the mask because of the smell of the inhalational agent. Similarly nitrous oxide is sometimes accidentally given, because some devices have a simple switch for deciding between air and nitrous oxide, and the gas no longer has to be dosed by a rotameter. The author has been called several times because a patient did not wake up and the only reason was the continuing administration of nitrous oxide. The author wonders that the industry has not yet made changes in a way that the fresh gas supply can only be turned off when the vaporizer is closed, and it is surprising that even some modern anaesthesia workstations can be switched off with an open vaporizer.
A similar type of error is leaving open the fresh gas supply, usually oxygen, after the termination of anaesthesia. Flushing a circle system over many hours, typically over a weekend, leads to a completely desiccated absorbent. Desiccated absorbent reacts in a different way with inhalational agents. This leads to the degradation of the inhalational agent in an exothermic reaction with increased formation of carbon monoxide and other degradation products (Wissing et al. 2001). Whereas with desflurane it is mainly carbon monoxide production that is in the focus of interest, with sevoflurane the absorbent can reach temperatures of several hundred degrees Celsius and compounds irritating the airway are formed (Wissing et al. 1997). The author remembers a 3-year-old girl who was the first case on a Monday morning. She was induced by mask with sevoflurane and nitrous oxide and went to sleep, then the nitrous oxide was turned off. One minute later she suddenly began to cough, sat up and pushed the mask away and was fully awake with a still irritated airway. The absorbent canister was extremely hot. It was clear that sevoflurane had been reaching the patient at first, but then with increasing temperature sevoflurane was degraded in the absorbent canister before it could reach the patient. As the phenomenon was already known from the literature, early diagnosis was made, and it could be reconstructed that oxygen was indeed already turned on, when the circuit check was made. In the late 1990s this led to a change of practice such that the fresh gas supply was always detached from the wall connectors over the weekend. Happily this type of complication is no longer possible with modern anaesthesia workstations, because switching off the device automatically stops the oxygen supply.
The accumulation of carbon monoxide with reduced fresh gas flow, even with a correctly hydrated absorbent, has been discussed in the past. There are mainly three sources for the elevated concentration of carbon monoxide haemoglobin: the degradation of sevoflurane, parental (or, rarely in children, the patient’s) smoking, and haemoglobin breakdown.
Accidentally administering anaesthesia to a healthy schoolgirl visiting the hospital is a delicate affair, and careful and timely communication is recommended. It is wise and correct to inform the accompanying teacher as well as the parents about the event, telling them what happened, that we are sorry, and, most important, that there will be no long-term sequelae. However, in the presented case, even the anaesthetist did not immediately check what was going on and the open vaporizer was only detected afterwards. It has been the author’s practice to call parents down to the operating suite whenever something unexpected and unwanted has happened, e.g. a massive extravasation, in order to ensure that it is brought to the parents’ notice immediately and they do not first hear about it afterwards from the nurses. Parents usually do not complain because a complication happened, but because they were not immediately and fully informed.
Summary and Recommendations
Before placing the mask on the patient’s face it must be checked that only the desired gases are administered. There is no excuse for negligence.
Desiccated absorbent, which occurs after an expanded dry gas flow, enhances the breakdown of halogenated agents and has to be replaced.
In case of an inadvertent event, it is of paramount importance that the parents are informed in a timely and personal, face-to-face manner, in order to retain their confidence.
Case
Many years ago, a 2-week-old girl, born at 23 3/7 weeks of gestation, weighing 640 g, was scheduled for emergency laparotomy because of intestinal perforation. Following the institutional practice, surgery was performed in the neonatal intensive care unit (NICU) and the neonatal ventilator was used. The anaesthetic regime was based on a high-dose opioid technique: a total of 120 µg/kg fentanyl and 0.2 mg/kg pancuronium was administered over the 3-hour procedure. During surgery, which included the insertion of a Broviac catheter, small bowel resection and ileostomy, 20 ml/kg of packed red cells and 10 ml/kg of fresh frozen plasma were given in addition to crystalloids. The baby remained remarkably stable, and the only vasopressor needed was dopamine up to 5 µg/kg/min.
Forty-seven hours later duct ligation was indicated. The intention was again to administer pancuronium and fentanyl. At the arrival of the anaesthetist the mean arterial pressure was high, at 55 mmHg; pancuronium and fentanyl were given, the dopamine infusion was turned off, and the baby was positioned for surgery. There was no reaction to skin incision and surgery proceeded as planned. But at the end of the case, when the drapes were removed, the anaesthetist broke out in a sweat when he suddenly noticed that the three-way-stopcock was in a position which impeded the administration of fentanyl (Fig. 6.4a).
Figure 6.4a A three-way stopcock in the wrong position impeded the administration of fentanyl.
Discussion
This case illustrates that failed drug administration is an inherent risk of every total intravenous technique. In this case the three-way stopcock was turned to the wrong position; other pitfalls include infiltrated lines, disconnected tubing (Fig. 6.4b), leaking syringes, failing pumps and, most important, the anaesthetist simply making a mistake. The author has painfully experienced all these situations. Therefore all these potential failures have to be excluded before starting the case and the specific sites have to be checked repeatedly throughout the case. However, in neonatal surgery access to the patient is very limited. A failure in the administration of propofol, e.g. the pump is turned off and not restarted, is a common cause of awareness (Pandit et al. 2014). In addition, some kind of pharmacodynamic feedback is recommended, e.g. an EEG-based monitor (Constant & Sabourdin 2012); but the EEG can only reliably be used for monitoring the depth of hypnosis in children older than 1 year.
Figure 6.4b Interrupted administration of propofol because of defective tubing in another patient.
In this 640 g baby, for the initial operation a high-dose opioid anaesthetic was used. There is no consensus about the optimal anaesthetic technique for this type of patient. They are usually dependent on a neonatal ventilator with no possibility of administering inhalational agents, and the only option is an intravenous technique. Two questions need to be answered: the necessary dose for sufficient analgesia and the need for a hypnotic.
First, for these extremely preterm patients there are no data about the necessary dosage. Some practitioners use moderate doses of morphine, e.g. 250 µg/kg – but maybe sufficient analgesia is only mimicked because the hypotensive side effect impedes the increase in blood pressure at skin incision. Experts recommend a dose of 25 µg/kg fentanyl for duct ligation (Wolf 2012). In a pharmacokinetic study, 30 µg/kg fentanyl resulted in haemodynamic stability at skin incision but not at the time of skin closure (Collins et al. 1985). In slightly older babies a dose of 25–50 µg/kg resulted in haemodynamic stability during sternotomy (Duncan et al. 2000). Remifentanil is rarely used in this context; 1 µg/kg/min suppresses the haemodynamic and hormonal stress response (Weale et al. 2004). Traditionally the author used a dose of at least 50 µg/kg fentanyl given as a short infusion over 5–10 minutes, always preceded by a muscle relaxant, in order to avoid rigor with the inability to ventilate the patient (Fahnenstich et al. 2000). For the second intervention, this patient erroneously did not receive any fentanyl at all because the three-way stopcock was turned to the wrong position, but nevertheless haemodynamic stability during surgery was maintained. How can this be explained? A very long elimination half-time of between 6 and 32 hours has been reported in preterm babies with a gestational age between 23 and 38 weeks (Collins et al. 1985). Therefore, 47 hours after the extremely high dose of 120 µg/kg fentanyl, plasma levels may still have been sufficient to blunt the haemodynamic stress response.
Second, there is a debate on the necessity of amnesia and hypnosis as part of the anaesthetic regime in this age group. There is no doubt that profound analgesia and the absence of dyspnoea are essential. But, as long as no fear about the future exists, at least up to the age when stranger anxiety is common, sleep and amnesia are probably not necessarily needed and a pure opioid regime is defensible. If an awake regional technique were used, no one would request sleep and amnesia. This is further underlined by the fact that even at early school age, where intraoperative awareness is a relatively common complication of general anaesthesia (Davidson et al. 2011), being accidentally awake during surgery does not provoke anxiety and long-term sequelae to the extent that it does in adults (Phelan et al. 2009). As long as children live in their ‘magic world’ and unexplained perceptions are a normal part of their lives, they are protected against the negative consequences of being awake. A hypnotic agent is therefore not essential for neonatal anaesthesia; nonetheless, it has to be admitted that traces of sevoflurane markedly increase haemodynamic stability at the time of skin incision.
This case of major surgery in a small patient also prompts a discussion on the need for placing central venous and arterial catheters. Reliable venous access is essential. The question will always be what we can get with an acceptable risk in a 640 g baby. The author aimed for at least one line with backflow; in this case it was a surgically implanted Broviac catheter into the internal jugular vein, placed before beginning laparotomy. Because of his own negative experiences, the author strongly recommends doing this in two steps: first the Broviac catheter, then remove the drapes and connect the line, before positioning and preparing for laparotomy.
Summary and Recommendations
Failed drug administration is an inherent risk of total intravenous anaesthesia. Meticulous care is needed to avoid all technical drawbacks.
A pharmacodynamic feedback system, e.g. EEG-based monitoring, should be standard practice in children above 1 year of age.
This case also illustrates the unsolved debate about the optimal anaesthetic regime in very tiny babies. A high-dose opioid regime is often used e.g. for duct ligation.
Vascular access with backflow, which allows taking blood samples, is recommended for major surgery in small babies.
References
Case
Many years ago, a 5-year-old boy, weighing 17 kg, was scheduled for an occipital craniotomy and medulloblastoma resection in the prone position. Some days earlier, propofol-based sedation for an MRI had been uneventful. After an inhalational induction with sevoflurane a total intravenous technique with propofol and remifentanil was chosen in order to allow reliable registration of the evoked potentials; these were traditionally recorded by a specialized neuroanaesthetist in this institution. The further instrumentation, beside peripheral venous access and a size 5.0 cuffed nasotracheal tube, included a radial arterial line, a central venous catheter via the right internal jugular vein, a BIS monitor and a bladder catheter. The anaesthetic course was uneventful over 8 hours, haemodynamic stability was maintained, and no vasoactive support was needed. At the end of the case the body temperature was 36.5 °C. The regularly taken blood gases were within the normal range. But shortly after arrival in the paediatric intensive care unit (PICU) a metabolic acidosis with an elevated lactate concentration of 4.8 mmol/l was evident; the lactate concentration peaked soon afterwards at 7 mmol/l and was accompanied by a creatine kinase (CK) of 480 U/l. Over the next few hours the values gradually returned to normal. The presumptive diagnosis of a propofol infusion syndrome was made. Propofol had been infused at a rate of 10 mg/kg/h over 6 hours.
Some weeks later radiation therapy was initiated, with daily sessions over several weeks. Now the question arose: was it safe, with such a history, to administer daily propofol?
Discussion
This case shows that propofol infusion syndrome (PRIS) is always a threat when propofol is administered over a prolonged time. PRIS was initially described in the early 1990s, with a high mortality after prolonged administration of propofol to children in the intensive care setting (Parke et al. 1992). But PRIS also occurs in the context of anaesthesia of duration as short as 150 minutes (Kill et al. 2003, Mehta et al. 1999).
The pathophysiology is only partly understood; mitochondrial toxicity is clearly a prime suspect (Krajcova et al. 2015). At the beginning of the century, based on a case report (Wolf et al. 2001), impaired oxidation of fatty acids was postulated (Vasile et al. 2003). The mitochondrial oxidative metabolism is blocked; propofol impedes the electron flow through the respiratory chain and interacts with coenzyme Q (Vanlander et al. 2015); propofol is thus mimicking a ‘pseudohypoxic state’. This lack of energy at a cellular level induces rhabdomyolysis, hyperkalaemia and cardiac pump failure with bradycardia and a Brugada type ECG. In addition, the cardiac toxicity is enhanced by high concentrations of free fatty acids. The body has recourse to anaerobic metabolism, and therefore metabolic acidosis and increasing lactate levels are early signs of PRIS (Koch et al. 2004). At an early stage these changes are reversible when the propofol administration is stopped or reduced (Koch et al. 2004); as it progresses, fatalities are common. A great number of cases have been reported, affecting both children and adults, during intensive care and during anaesthesia. PRIS can occur even at doses below 4 mg/kg/h (Krajcova et al. 2015).
For several reasons the occurrence of PRIS is more obvious in children. First, for pharmacokinetic reasons children need very high doses of propofol to achieve a sufficient depth of sedation or anaesthesia. Second, the high metabolic rate combined with the limited glycogen stores induces an early recurrence to fatty acid oxidation. Third, children are usually otherwise healthy, and mortality of cardiac origin should rarely occur in a child sedated and ventilated because of a respiratory infection. So the singularity of the symptoms of PRIS makes diagnosis relatively easy; whereas the elderly multimorbid adult patient often has many other reasons for circulatory failure, and cardiac death and the diagnosis of PRIS is probably often overlooked.
PRIS is an established entity in adults; in otherwise healthy subjects 10% mortality was caused by propofol in head-injured patients (Cremer et al. 2001) or patients with refractory status epilepticus (Iyer et al. 2009). With high doses of propofol, 35% of these patients showed signs of PRIS. PRIS is not just a rare idiosyncratic event; it is more likely a dose-dependent phenomenon which could even affect a majority of our patients, although some individuals may be more sensitive.
In some reported cases the propofol infusion was stopped for several hours, and then restarted at a low infusion rate; early after re-exposure, PRIS developed. Therefore, in the case described here, the safety of daily propofol administration over several weeks for radiation therapy was questioned. As there was no answer in the scientific literature, the pragmatic approach was to proceed with propofol for these daily short anaesthetics and to measure blood gas and lactate at the end of the first week, which showed ensuring normal values.
Propofol is in wide clinical use in paediatric anaesthesia. Pain at injection, hypotension and bradycardia are typical complications and are regularly seen. PRIS is a rarer event, but, including this case, the author has seen three cases of PRIS during his professional career. In addition, several times rumours about severe or even lethal cases reached him. A grass-green discoloration of the urine is another rare side effect of propofol, but it seems to be benign and not to indicate a patient at increased risk of developing PRIS. The author’s recommendations for the clinical use of propofol are summarized in Fig. 6.5.
