Ketamine acts as an antagonist of N-methyl d-aspartate (NMDA) receptors in the central nervous system, thereby inducing a state of dissociative anaesthesia. At low doses, it has an analgesic and antihyperalgesic effect. Another of its properties is an affinity for opioid receptors. Although this does not produce an analgesic effect, it may have to do with its usefulness in reducing the need for opioids and preventing withdrawal syndrome1,2. Studies on the use of ketamine for surgical procedures have demonstrated a favourable haemodynamic profile, with minimal cardiovascular depression, which is an advantage in haemodynamically unstable patients, for example after cardiac surgery3–7. However, studies on the efficacy and safety of prolonged ketamine use are needed, as those available in the current literature are mostly retrospective and conducted in small samples of patients8–11. Other groups of patients, such as critically ill neurologic patients, could also benefit, as, while it has been hypothesised that ketamine can raise intracranial pressure, there is also recent evidence suggesting that it may have a beneficial effect in these patients by decreasing intracranial pressure12,13.

Propofol acts by modulating the neurotransmitter g-aminobutyric acid (GABA) through the binding to its receptor, producing a hypnotic and sedative effect. Since it is highly lipophilic, it has a rapid onset of action in the central nervous system, and the rapid redistribution from the central to the peripheral compartment causes a quick offset of its effects. It also has a hypotensive effect through the decrease of systemic vascular resistance and heart rate14. On account of these effects, it is not as widely used in haemodynamically unstable patients. However, due to the opposing characteristics of ketamine and propofol, their combined administration may improve the adverse event profile15,16.

The 2022 guidelines for sedation in the intensive care setting of the Society of Critical Care Medicine recommend ketamine as an adjuvant treatment in paediatric patients in whom customary drugs have achieved inadequate sedation, but continue to be more restrictive toward the use of propofol in children17. The aim of our study was to assess the efficacy and safety of the combined use of ketamine and propofol in continuous infusion as part of a standard protocol for prolonged analgesia and sedation in the PICU setting.

We designed a prospective observational study to assess the efficacy and safety of the combination of ketamine and propofol administered via continuous infusion in the PICU of a tertiary care hospital between September 2016 and January 2018. This combination is included in the analgesic and sedative drug rotation protocol introduced in 2012 with the aim of reducing withdrawal syndrome in critically ill patients that receive prolonged sedation18. The protocol is based on the use of different combination of analgesic and sedative drugs taking into account their pharmacokinetics and mechanisms of action, alternating opioid and non-opioid sedatives and benzodiazepine sedatives with sedatives from other groups19,20. It includes the administration of ketamine (dose of 1−2 mg/kg/h) and propofol (dose of 1−4 mg/kg/h) for 5 days. Fig. 1 presents the complete sequence. This protocol is used in every patient, although adaptations are allowed based on the judgment of the physician in charge of the patient. The study adhered to the principles of the Declaration of Helsinki of 1975, revised in 2000, and was approved by the Institutional Review Committee of our hospital under study code EVOLUCION PICU 1.2. We obtained the informed consent of the families for data collection for research purposes.

Figure 1.

Analgesic and sedative drug rotation protocol.

(0.12MB).

Thirty-two patients received ketamine and propofol via continuous infusion for more than 24 h in the period under study. The median age was 6 months (IQR, 2–48.5 months) and the median PRISM III score 8 (IQR, 3–12). Nineteen patients (59.4%) had a congenital heart defect. Table 1 presents other baseline characteristics of the sample.

Thirty patients (93.7%) received midazolam and 29 (90.6%) fentanyl in the 5 days preceding the introduction of ketamine and propofol. The median dose of midazolam was 2 μg/kg/minute (IQR, 2–3.2 μg/kg/minute; maximum, 6 μg/kg/minute), while the median dose of fentanyl was 2 μg/kg/h (IQR, 1−2.2 μg/kg/h; maximum, 5 μg/kg/h). The median dose of ketamine at treatment initiation was 1 mg/kg/h (IQR, 1–1.5 mg/kg/h). The median maximum dose for continuous infusion was 1.5 mg/kg/h (IQR, 1−2 mg/kg/h) and the median overall maximum dose (including boluses) was 1.6 mg/kg/h (1.2−2 mg/kg/h). The highest dose was 2 mg/kg/h and was used in 43.8% of cases. The median duration of infusion was 5 days (IQR, 3–5 days). The maximum duration was 8 days. The median duration of ketamine discontinuation was 2 h (IQR, 1−5 h), with a maximum of 24 h. There were no significant differences in the maximum dose of ketamine based on age (infants, 1.5 mg/kg/h vs age > 12 months, 1.5 mg/kg/h; P = .785), or between patients previously admitted to the PICU versus patients without a previous PICU stay (1.5 mg/kg/h vs 1.5 mg/kg/h; P = .468).

Per the hospital analgesia and sedation protocol, 32 patients (100%) received propofol in continuous infusion in combination with ketamine. The median maximum dose of propofol (including continuous infusion and boluses) was 3.2 mg/kg/h (IQR, 2.5–3.6 mg/kg/h). In 3 patients, propofol was switched to midazolam and in 1 by dexmedetomidine. One patient received midazolam and ketamine simultaneously. Twenty-four patients (74.2%) did not need concomitant pharmacotherapy during the treatment with ketamine and propofol. Eight (25.8%) received other treatments at the same time as intravenous ketamine and propofol for more than 24 h. In 2 patients, these concomitant treatments were already being administered at the time of treatment with fentanyl and midazolam before initiation of continuous infusion of ketamine and propofol, and were not suspended at that point. The other 6 patients (19%) required the introduction of adjuvant drugs after the initiation of continuous infusion of ketamine and propofol. These adjuvant treatments were: continuous infusion of metamizole (n = 3; 9%), midazolam (n = 1; 3%), dexmedetomidine (n = 4; 12.5%), continuous infusion of morphine (n = 1; 3%), inhaled sevoflurane (n = 1; 3%), fentanyl (n = 1; 3%) and remifentanil (n = 1; 3%). For metamizole, the maximum dose was 6.6 mg/kg/h, while for dexmedetomidine it was 1.2 μg/kg/h. The maximum dose of morphine was 20 μg/kg/h and the maximum dose of remifentanil 12 μg/kg/h.

The median scores in the MAPS, COMFORT-B and SOS scales prior to the introduction of combined ketamine and propofol and in the 5 days following treatment initiation can be found in the Supplemental material. There were no significant differences in the median scores in the MAPS and SOS before and after initiation of continuous infusion (Days 1−5) (MAPS 0–2 and SOS < 4) (Appendix B, Supplemental material). The level of analgesia did not vary with the introduction of ketamine and propofol (score: 1). The COMFORT-B score increased significantly comparing days 1–5 of treatment to the day before initiation of ketamine and propofol, but remained within 12–17 points (Supplemental material). In one patient, propofol and ketamine were discontinued because they did not achieve an adequate level of analgesia and sedation, switching to remifentanil and dexmedetomidine. Four patients had a SOS score greater than 4 in the first 48 h after discontinuation of ketamine and propofol (12.9%).

Table 2 presents the haemodynamic parameters measured during the administration of ketamine and propofol in continuous infusion. We did not find significant differences in the VIS before and after initiation of ketamine and propofol (P = .888). We found statistically significant differences between the MAP at baseline and at 1 h of infusion (64 mmHg compared to 60 mmHg; P = .023) and in the DBP at baseline and after 1 h of ketamine and propofol (50.5 compared to 48 mmHg; P = .023). We did not find differences between baseline and 12 h after initiation of ketamine and propofol (Table 2). One patient required noradrenaline 12 h after initiation of ketamine and propofol. None of the patients required antihypertensive drugs.

We did not find any differences in haemodynamic parameters or changes in the VIS in the 5 patients undergoing extracorporeal membrane oxygenation (ECMO) (Table 3).

The most frequent adverse event during ketamine and propofol infusion was bronchorrhoea (n = 12; 37.5%), although it was not clinically significant in most cases (n = 11; 91.6%). Three patients developed agitation. In one of them, both propofol and ketamine were discontinued and treatment with midazolam and fentanyl resumed, since propofol and ketamine are not effective for controlling agitation. An opioid infusion was initiated in the other 2 patients (morphine and remifentanil). There were no cases of nystagmus or arrythmia. We did not find an association between the maximum dose or the duration of ketamine and propofol infusion and the development of adverse events. The combination of ketamine and propofol was associated with a significant increase in triglyceride levels (from 127 mg/dL to 206 mg/dL; P < .001). There was also a mild increase in lactate levels (from 1 mg/dL to 1.8 mg/dL; P < .001). None of the patients developed rhabdomyolysis or hepatomegaly. There were no deaths during infusion of ketamine and propofol.

We present one of the few studies describing the prolonged use of ketamine and propofol administered in combination as continuous infusion in a PICU. Our study contributes evidence on the usefulness of combined ketamine and propofol for maintenance analgesia/sedation and included patients managed with ECMO and renal replacement therapy.

The doses of ketamine and propofol used in our unit were similar to those administered in previous studies8–11,15,25. The duration of ketamine infusion in previous studies ranged from 2 to 6 days, and it was of 5 days in our study1,2,8,26.

The combination of ketamine and propofol achieved an analgesic and sedative effect comparable to those achieved with previous treatments (fentanyl and midazolam). We considered the analgesic effect, assessed by means of validated scales, adequate (MAPS < 2). Although the score in the COMFORT-B scale increased with the introduction of the combination of drugs, the sedative effect remained in the range considered adequate (12–17)21. These results are consistent with those of other studies in the PICU setting that found adequate levels of analgesia and sedation using each of these drugs separately although there is still a dearth of data on their use in combination for prolonged analgesia1,2,8,27,28. It is worth highlighting that in 81% of cases, an adequate level of analgesia and sedation was achieved without needing to add any adjuvant drugs. The remaining 19% of patients received some other intravenous drug in continuous infusion in addition to propofol and ketamine. In these cases, it would be reasonable to consider that ketamine and propofol did not achieve the desired level of sedation. Fentanyl withdrawal may have played a significant role in these patients, as in most previous studies treatment with ketamine was initiated while maintaining the initial opioid treatment8,10,11, whereas in our study, opioid treatment was discontinued on initiating ketamine and propofol.

In one patient, the infusion of ketamine and propofol was discontinued on account of inadequate analgesia/sedation due to agitation. Agitation has been reported as an adverse effect of ketamine in procedural sedation, although it is more frequent with other drugs, such as benzodiazepines29.

Although the evidence in the paediatric population is limited, it seems that the haemodynamic impact of ketamine results from two opposing mechanisms: a direct negative inotropic effect and an indirect positive inotropic effect induced by the increase in the concentration of endogenous catecholamines30. Propofol produces a decrease in blood pressure through a decrease in vascular capacitance, myocardial contractility and autonomic control of cardiac output15. The combination of ketamine and propofol for procedural sedation in paediatric patients has proven safe, without significant decreases in heart rate or blood pressure, but studies assessing its effectiveness and safety for prolonged sedation are still lacking15,31,32. Few studies to date have included patients with congenital heart disease. It is important to consider potential haemodynamic changes when assessing the effects of ketamine and propofol in continuous infusion. A study on analgesia and sedation in paediatric patients undergoing cardiac catheterization found that the combination of ketamine and propofol reduced the cumulative dose of propofol and preserved the mean arterial pressure16.

We found a small but statistically significant decrease in the MAP and DBP in the first hour of ketamine and propofol infusion that normalised afterwards without requiring increased vasoactive drug support or fluid replacement therapy. This suggests that the use of combined ketamine and propofol in continuous infusion does not have significant haemodynamic repercussions, even in patients with cardiac disease.

In this study, none of the patients had serious adverse events associated with the administration of ketamine and propofol. Bronchorrhoea developed in more than one third of patients, an effect previously described in relation to ketamine, but it did not have a significant impact in respiratory function and did not require treatment with anticholinergic drugs. Other studies have had similar findings8,10,26. Previous clinical case series have reported propofol infusion syndrome in paediatric patients, chiefly with the use of doses much greater than 4 mg/kg/h. Propofol infusion syndrome is characterised by bradycardia, metabolic acidosis, hypertriglyceridemia, rhabdomyolysis and hepatomegaly33. None of the patients developed bradycardia, rhabdomyolysis or hepatomegaly, and the increases in the levels of triglycerides and lactate, while statistically significant, were not clinically relevant.

We did not detect other adverse effects described in the literature (intracranial or ocular hypertension). This limits the extrapolation of our findings to patients at risk of developing these adverse effects34–36.

Four patients (12.5%) developed withdrawal symptoms after discontinuation of ketamine and propofol. However, this must be interpreted with caution, given the concomitant use of other pharmaceuticals.

One of the limitations of our study was the small number of included patients. Still, it is one of the few studies with a prospective design, and the sample was larger compared to previous publications. It was an observational study that documented real-world analgesia and sedation practices in in our hospital. The observational design and lack of a control group are limitations that must be taken into account in the interpretation of the results.

It is also important to consider the use of other sedatives and analgesics in interpreting our findings. However, this is an unavoidable limitation, as in clinical practice adjuvant drugs are widely used in the sedation of critically ill children. Another limitation was the lack of data on the administration of analgesic and sedative drugs via the enteral route.

The combination of ketamine and propofol delivered through continuous infusion may be an appropriate treatment for paediatric patients admitted to intensive care units, as it achieves adequate levels of analgesia and sedation and allows discontinuation of opioids and benzodiazepines. Ketamine and propofol have a good safety profile and do not cause significant haemodynamic effects or adverse events. However, it is essential to obtain further data through prospective controlled trials conducted in larger samples.

This study did not receive specific financial support from any public, private or not-for-profit institutions.

The authors have no conflicts of interest to declare.