Skip to main content

Comparison of sufentanil with sufentanil plus magnesium sulphate for sedation in the intensive care unit using bispectral index



In intensive care unit patients we assessed, using bispectral index (BIS) monitoring, whether the addition of magnesium sulphate infusion could decrease the sufentanil infusion required to maintain sedation.

Patients and methods

A total of 30 adult patients who were expected to require machanical ventilation for 6 hours in the intensive care unit were randomly assigned to receive either sufentanil infusion or sufentanil plus magnesium infusion. We monitored BIS levels continously. BIS levels in the range 61–88 are required to maintain a state of sedation, and in both groups BIS levels were kept within this range by increasing or decreasing the sufentanil infusion. Hourly consumption of sufentanil was monitored. Cardiovascular, respiratory and biochemical data were recorded.


There was no significant difference between the groups with respect to cardiovascular, respiratory and biochemical parameters. Magnesium infusion, when added to sufentanil infusion, decreased the consumption of sufentanil at all times accept during the first hour (P < 0.001). There was no significant difference in BIS values between the groups (P > 0.05).


This is the first clinical study to demonstrate that magnesium sulphate infusion decreases sufentanil requirements. Because of the limited number of patients included and the short period of observation, our findings must be confirmed by larger clinical trials of magnesium infusion titrated to achieve prespecified levels of sedation. Furthermore, randomized clinical studies are needed to determine the effects of magnesium infusion on opioids.


Inadequate sedation may adversely affect morbidity and even mortality in the intensive care unit (ICU), and the search for the ideal sedative agent therefore continues [1, 2]. Sedation is an important part of therapy for critically ill patients in the ICU. It reduces anxiety and stress, facilitates sleep, prevents injuries and accidental removal of catheters, reduces resistance to mechanical ventilation, and decreases oxygen consumption in severe head injury [2]. Under-sedation may result in hypertension, tachycardia, discomfort and resistance to mechanical ventilation. Hypotension, bradycardia, coma, respiratory depression, immunosuppression, paralytic ileus and renal failure are among the risks associated with over-sedation [2, 3]. Use of combinations of hypnotics and opiates for sedation has become common.

Bispectral electroencephalographic analysis was recently developed, and the effects of many anaesthetic agents on levels of sedation have been studied using bispectral index (BIS) monitoring. These reports suggest that BIS levels correlate well with depth of sedation [4, 5]. Only a few investigators have studied the validity of BIS monitoring in ICU patients, but a correlation between BIS levels and sedation scales in ICU patients has been reported for various agents [6, 7].

Sufentanil allows rapid emergence from sedation and less respiratory depression than occurs with other sedative agents, and it maintains haemodynamic stability [8]. It is an effective analgesic and its rapid onset of action allows simple titration, according to individual patient tolerance, to the the desired clinical end-point. Furthermore, its short duration of action means that the desired analgesic effect will continue for as long as the infusion is continued, and any undesirable side effects will be short-lived following discontinuation [9]. In addition, mechanically assisted spontaneous ventilation modes can safely be used under continuous sedation with sufentanil [10].

Magnesium sulphate is involved in several processes, including hormone receptor binding, gating of calcium channels, transmembrane ion flux and regulation of adenylate cyclase, muscle contraction, neuronal activity, control of vasomotor tone, cardiac excitability and neurotransmitter release. In many of its actions it has been likened to a physiological calcium channel blocker [11].

In the present study, conducted in ICU patients, we assessed (using BIS monitoring) whether the addition of magnesium sulphate infusion could decrease the sufentanil infusion required to maintain sedation.

Patients and method

Patient population and study design

The study was approved by the regional committee on medical research ethics. Written informed consent was obtained directly from the patients whenever possible, or from their next of kin.

A total of 30 adult patients (trauma, general surgical and medical) requiring mechanical ventilatory support were enrolled in this prospective analysis. The initial severity of illness was determined using Acute Physiology and Chronic Health Evaluation (APACHE) II [12] scores. Patients with overt disease affecting the brain (e.g. head trauma, intracranial haemorrhage, meningitis or stroke) and those receiving neuromuscular blocking agents were excluded.

Patients were ventilated mechanically with oxygen enriched air to achieve acceptable blood gas levels. If required, patients underwent surgical procedures before the start of the study. No invasive surgery was performed during the 24 hour study period. Ventilator settings, level of positive end-expiratory pressure and fractional inspired oxygen were kept constant during sufentanil and magnesium infusion.

Antibiotic treatment was adjusted in accordance with results of bacteriological culture, such as culture of blood or of samples taken from different sites of the body. No inotropic agent was administered during the study.


The study was prospective, randomized, double blind, and placebo-controlled. Randomization was conducted according to a computer directed, permutated block design. In accordance with the double blind protocol, drug solutions and infusions were administered to patients by a nurse who had no knowledge of the study protocol. The BIS was monitored using a BIS Monitor Model A-2000 (Aspect Medical Systems Inc., Newton, MA, USA). After placement of the BIS electrode (Aspect Medical Systems Inc., Natick, MA, USA) above the bridge of the nose, over the temple area, and between the eye and hairline, the monitor undergoes automic impedence testing to ensure acceptable signal reception. When an inadequate signal is sensed the display's colour pattern changes, allowing easy differentiation between true and spurious values. Electrodes were repositioned or replaced if impedances increased to the extent that electroencephalographic evaluation was impaired. The degree of sedation was measured continously using BIS monitoring. Patients were maintained at BIS levels in the range 61–88, which are associated with a sedated state.

At the start of the study all patients received 1 μg/kg sufentanil by intravenous bolus. Immediately after, groups 1 and 2 received an intravenous sufentanil infusion for 6 hours, but group 2 also received an intravenous infusion of magnesium sulphate (2 g/hour) for 6 hours via a dedicated central venous catheter. Average BIS values were kept in the range 61–88 by decreasing or increasing sufentanil infusion in both groups, and hourly consumption of sufentanil was monitored. Evidence of hypocalcaemia was sought using clinical signs (Chvostek's and Trousseau's) and measurement of total serum calcium concentrations. Effects of elevated magnesium include flushing, loss of tendon reflexes, respiratory arrest, and prolongation of the PR interval and the QRS complex; these were also evaluated.

The sedative infusion was discontinued when cardiovascular and respiratory adverse events were identified, defined as a change in arterial pressure of more than 40% from baseline, bradycardia to less than 50 beats/min, or tachyarrhythmia. No other sedative or analgesic agents were given, and no patient received spinal or epidural analgesia in the ICU.


All patients underwent placement of arterial catheters and central venous catheters via the subclavian vein. Arterial blood samples were drawn for measurements of pH, oxygen and carbon dioxide tensions, and arterial oxygen saturation (Medica Easy BloodGas; Massachusetts, USA). Central venous pressure, mean arterial pressure, heart rate and nasopharyngeal temperature were continuously monitored (SpaceLabs Inc., Redmond, USA). Sodium, potassium, calcium (Ilyre, Ion Selective Electro Analyzer; LISpA, Milan, Italy), magnesium levels (Merek Mega, Darmstadt, Germany), and bilirubin, alanine aminotransferase and creatinine (Vitalab Flexor, Dieren, The Netherlands) were determined at baseline (15 min before start of the study), immediately after sedative infusion and 24 hours after sedative infusion.


Demographic data; haemodynamic, biochemical and arterial blood gas changes; sufentanil dose; and changes in BIS were analyzed using independent samples t tests. Data were expressed as means ± standard deviation. P < 0.05 was considered statistically significant.


Patient characteristics

Clinical and demographic characteristics of the patients are summarized in Table 1. Of the 30 patients included in the study, 15 received sufentanil (group 1) and 15 received sufentanil plus magnesium (group 2). Eight patients died while they were hospitalized in the ICU for reasons related to infection (four in group 1 and four in group 2). Baseline APACHE II scores (12.3 ± 4.3 and 13.4 ± 4.6 in groups 1 and 2, respectively) and central venous pressure (5.9 ± 2.4 mmHg and 5.6 ± 2.7 mmHg) were similar between groups (P > 0.05). The sufentanil infusion and the sufentanil plus magnesium infusion were well tolerated by all patients, and no adverse effects were noted.

Table 1 Clinical characteristics of the patients included

Haemodynamic parameters and oxygen transport variables

There was no significant difference between the groups with respect to pH, oxygen or carbon dioxide tension, ratio of arterial oxygen tension to fractional inspired oxygen, and arterial oxygen saturation (P > 0.05). No significant change in heart rate or mean arterial pressure was found in either group (Table 2). There was no significant difference between groups in biochemical parameters or temperature (P > 0.05).

Table 2 Hemodynamic parameters, and calcium and magnesium levels


Magnesium sulphate infusion, when added to sufentanil infusion, decreased the consumption of sufentanil at all times except during the first hour (P < 0.001; Fig. 1). There was no significant difference between the groups in terms of BIS values (P > 0.05; Fig. 2).

Figure 1
figure 1

Hourly consumption of sufentanil. Group 1 (n = 15) received sufentanil infusion alone, whereas group 2 (n = 15) received sufentanil plus magnesium sulphate. Data are expressed as means ± standard deviation. *P < 0.001, group 1 versus group 2.

Figure 2
figure 2

Hourly bispectral index (BIS) values. No statistically significant differences were found between the groups (P > 0.05). Group 1 (n = 15) received sufentanil infusion alone, whereas group 2 (n = 15) received sufentanil plus magnesium sulphate. Data are expressed as means ± standard deviation.


The overall hospital mortality rate was similar in both groups (26.6%). All of those who died did so while being mechanically ventilated. Mean survival time was 7 ± 3 days in the group receiving sufentanil plus magnesium sulphate and 7 ± 4 days in the group receiving sufentanil alone. The duration of mechanical ventilation and the number of ventilator free days (calculated as the number of days a patient was alive and without mechanical ventilation at 28 days) were similar between the groups. In the sufentanil and sufentanil plus magnesium groups, ventilation duration was 6 ± 2 and 7 ± 3 days and number of ventilator free days was 9 ± 3 and 8 ± 4 days, respectively (P > 0.05).

Side effects

No side effects were noted during or after administration of sufentanil infusion and sufentanil plus magnesium infusion.


Sedatives are common adjuncts in the treatment of anxiety and agitation. The causes of anxiety in critically ill patients are multifactorial and probably secondary to the continuous noise (making it impossible to communicate), continuous ambient lighting and excessive stimulation (inadequate analgesia, frequent vital sign monitoring, repositioning, lack of mobility and room temperature) that are characteristic of the ICU setting. Efforts to reduce anxiety, including frequent reorientation, maintenance of patient comfort, provision of adequate analgesia and optimization of the environment, may be supplemented with sedatives [1, 2].

The BIS of the electroencephalogram is an empirical, statistically derived variable that provides information about the interaction of brain cortical and subcortical regions. Sigl and Chamoun [13] reported technical details and rationale for the use of BIS. First, BIS provides information regarding interaction between cortical and subcortical areas, which changes with increasing amounts of hypnotic drugs. Second, the BIS is an empirical, statistically derived measurement that was developed by analyzing a large database of electroencephalograms from individuals who had received hypnotic agents. Third, the BIS measures a state of the brain and not the concentration of a particuler drug. Fourth, interpretation of the BIS is based on the assumption that sedation is intended to produce a state of sleep that includes a lack of awareness and a lack of recall, whereas analgesia is intended to produce a state of reduced pain perception manifested by decreased autonomic responses to noxious stimuli. Finally, in general a BIS score of 100 reflects the awake state, 80 reflects some sedation, 60 reflects a moderate level of hypnosis, and 40 reflects a deep hypnotic state [14]. In the present study BIS values were maintained in the range 61–88.

BIS is a processed electroencephalographic measurement that correlates with the sedative properties of single anaesthetic agents. Use of combinations of hypnotics and opiates to achieve sedation has become common. BIS is useful in the ICU for monitoring sedation (and preventing over-sedation) and for shortening the duration of ICU stay, and a consequent decrease in hospital costs associated with its use may be anticipated [15]. However, several questions have been raised regarding the use of BIS in the ICU concerning the widespread use of opioids (which reduces the validity of BIS) [15, 16], interpretation difficulties in neurological diseases [17] and controversial studies that demonstrated decreased correlation between clinical sedation scores and BIS [1719]. For example, a study conducted in a paediatric ICU [20] demonstrated that opioids do provide some degree of sedation.

Kroll and List [21] found analgesia and sedation with sufentanil to be satisfactory in critically ill patients. At a dosage range of 0.75–1.0 μg/kg per hour, this drug can safely be given to patients undergoing controlled mechanical ventilation. Wappler and coworkers [22] found that continuous infusion of sufentanil (1 μg/kg per hour was given initially) for analgesia and sedation is suitable for intensive care patients with a short stay in the ICU. Prause and colleagues [10] found that critically ill patients under continuous sedation with sufentanil (median 0.44 μg/kg per hour) exhibit a statistically significant rise in arterial carbon dioxide tension, but this respiratory depression is only slight and has no clinical significance. Mechanically assisted spontaneous ventilation modes can safely be used under continuous sedation with sufentanil. In the present study all patients initially received 1 μg/kg sufentanil by intravenous bolus.

There is a suggested role for magnesium in almost every physiological system. Key underlying mechanisms of action are those of calcium antogonism via calcium channels, regulation of energy transfer, and membrane sealing or stabilization [23, 24]. This has led to several studies on the central and peripheral nervous systems, and the cardiovascular, respiratory, endocrine and reproductive systems. Magnesium's action as an anticonvulsant is secondary to antagonism at N-methyl-D-aspartate receptors [25]. Stimulation of this subgroup of glutamate receptors is known to lead to excitatory postsynaptic potentials, causing seizures. Magnesium has successfully been used as an anticonvulsant in eclampsia [26].

In addition, magnesium is known to have a marked antiadrenergic effect. This is mediated by a variety of mechanisms, of which the most important is probably calcium channel blockade. Calcium plays a fundamental role in stimulus–response coupling of catecholamine release from the adrenal medulla and adrenergic nerve terminals, and its role in adrenal catecholamine release is well described [11]. In tetanus, magnesium has been used to treat both muscle spasm and autonomic dysfunction, which leads to large increases in catecholamine release [27].

Calcium channel blockers have antinociceptive effects in animals and potentiate the analgesic effects of morphine in patients with chronic pain [28, 29]. As mentioned above, magnesium is also an antagonist of N-methyl-D-aspartate receptor ion channels, and this may explain part of its analgesic activity [30]. The analgesic effect of magnesium has been demontrated both in human and in animal studies. Its effect on decreasing perioperative analgesic [31] and anaesthetic [32] requirements was demonstrated in various studies. In the present study, under highlights of magnesium's effects, when used for sedation/analgesia in ICU, we aimed to investigate the influence on sufentanil dose. Attygalle and Rodrigo [27] administered 2–3 g/hour magnesium in tetanus. In the present study we used a similiar dose of 2 g/hour magnesium.

Magnesium is utilized in the control of spasms in eclampsia and the safety of the therapeutic range (2–4 mmol/l) is well established, because areflexia only occurs at levels above 4 mmol/l and muscle paralysis above 6 mmol/l [27]. Magnesium does not cause sedation at serum concentrations below 8 mmol/l as long as ventilation is adequate (because it does not easily cross the blood–brain barrier) [27]. Attygalle and Rodrigo [27] administered 2–3 g/hour magnesium in tetanus. Those investigators found that magnesium sulphate can be used as the sole agent for the control of spasms in tetanus without the need for sedation and artificial ventilation. In our study serum magnesium concentrations did not increase to above 5 mmol/l but these concentrations increased the effectiveness of sufentanil when the agents were administered together.

However, the patients studied here were not postoperative patients and were not extubated at the end of the study; we were therefore unable to evaluate the effect of infusion of sufentanil plus magnesium on extubation criteria. We aimed to determine the effect of sufentanil plus magnesium infusion over a short period of time. Further studies in postoperative patients are needed to determine effects of sufentanil plus magnesium infusion on extubation criteria.

This is the first clinical study to demonstrate that magnesium infusion decreases sufentanil requirements. Because of the limited number of patients included and the short period of observation, our findings must be confirmed by larger clinical trials of magnesium infusion titrated to achieve prespecified levels of sedation. Furthermore, randomized clinical studies are needed to determine the effects of magnesium infusion on opioids.

Key messages

  • We assessed whether the addition of magnesium infusion in ICU patients decreased the sufentanil requirements using bispectral index

  • This clinical study demonstrated that magnesium infusion decreased sufentanil requirements



Acute Physiology and Chronic Health Evaluation


bispectral index


intensive care unit.


  1. Venn RM, Grounds RM: Comparison between dexmedetomidine and propofol for sedation in the intensive care unit. Br J Anaesth 2001, 87: 684-690. 10.1093/bja/87.5.684

    Article  CAS  Google Scholar 

  2. Riess ML, Graefe UA, Goeters C, Aken HV, Bone HG: Sedation assessment in critically ill patients with bispectral index. Eur J Anaesthesiol 2002, 19: 18-22.

    Article  CAS  Google Scholar 

  3. Burns AM, Shelly MP, Park GRI: The use of sedative agents in critically ill patients. Drugs 1992, 43: 507-515.

    Article  CAS  Google Scholar 

  4. Liu J, Singh H, White PF: Electroencephalogram bispectral analysis predicts the depth of midazolam-induced sedation. Anesthesiology 1996, 84: 64-69. 10.1097/00000542-199601000-00007

    Article  CAS  Google Scholar 

  5. Glass PS, Bloom M, Kearse L, Rosow C, Sebel P, Manberg P: Bispectral analysis measures sedation and memory effects of propofol, midazolam, isoflurane, and alfentanil in healthy volunteers. Anesthesiology 1997, 86: 836-847. 10.1097/00000542-199704000-00014

    Article  CAS  Google Scholar 

  6. Freye E, Neruda B, Falke K: EEG power spectra and evoked potentials during alfentanil/midazolam analgesia-sedation in intensive care patients [in German]. Anasthesiol Intensivmed Notfallmed Schmerzther 1991, 26: 384-388.

    Article  CAS  Google Scholar 

  7. Campbell ML, Bizek KS, Thill M: Patient responses during rapid terminal weaning from mechanical ventilation: a prospective study. Crit Care Med 1999, 27: 73-77. 10.1097/00003246-199901000-00029

    Article  CAS  Google Scholar 

  8. Flacke JW, Bloor BC, Kripke BJ, Flacke WE, Warneck CM, Van Etten AP, Wong DH, Katz RL: Comparison of morphine, meperidine, fentanyl, and sufentanil in balanced anesthesia:a double-blind study. Anesth Analg 1985, 64: 897-910.

    CAS  Google Scholar 

  9. Palmer CM: Continuous intrathecal sufentanil for postoperative analgesia. Anesth Analg 2001, 92: 244-245.

    Article  CAS  Google Scholar 

  10. Prause A, Wappler F, Scholz J, Bause H, Schulte am Esch J: Respiratory depression under long-term sedation with sufentanil, midazolam and clonidine has no clinical significance. Intensive Care Med 2000, 26: 1454-1461. 10.1007/s001340000637

    Article  CAS  Google Scholar 

  11. Fawcett WJ, Haxby EJ, Male DA: Magnesium: physiology and pharmacology. Br J Anaesth 1999, 83: 302-320.

    Article  CAS  Google Scholar 

  12. Knaus WA, Draper EA, Wagner DP, Zimmerman JE: APACHE II: a severity of disease classification system. Crit Care Med 1985, 13: 818-829.

    Article  CAS  Google Scholar 

  13. Sigl JC, Chamoun NC: An introduction to bispectral analysis for the EEG. J Clin Monit 1994, 10: 392-404.

    Article  CAS  Google Scholar 

  14. Shapiro BA: Bispectral index: better information for sedation in the intensive care unit? Crit Care Med 1999, 27: 1663-1664. 10.1097/00003246-199908000-00056

    Article  CAS  Google Scholar 

  15. Frenzel D, Greim CA, Sommer C, Bauerle K, Roewer N: Is the bispectral index appropriate for monitoring the sedation level of mechanically ventilated surgical ICU patients? Intensive Care Med 2002, 28: 178-183. 10.1007/s00134-001-1183-4

    Article  Google Scholar 

  16. Singh H: Bispectral index (BIS) monitoring during propofol-induced sedation and anaesthesia. Eur J Anaesthesiol 1999, 16: 31-36. 10.1046/j.1365-2346.1999.00420.x

    Article  CAS  Google Scholar 

  17. O'Connor M, Kress JP, Pohlmann A, Tung A, Hall J: Pitfalls of monitoring sedation in the ICU with the bispectral index. Anesthesiology 1998, 89: A461. 10.1097/00000542-199809080-00085

    Article  Google Scholar 

  18. Simmons LE, Riker RR, Prato BS, Fraser GL: Assessing sedation during intensive care unit mechanical ventilation with the bispectral index and the sedation-agitation scale. Crit Care Med 1999, 27: 1499-1504. 10.1097/00003246-199908000-00016

    Article  CAS  Google Scholar 

  19. De Deyne C, Struys M, Decruyenaere J, Creupelandt J, Hoste E, Colardyn F: Use of continuous bispectral EEG monitoring to assess depth of sedation in ICU patients. Intensive Care Med 1998, 24: 1294-1298. 10.1007/s001340050765

    Article  CAS  Google Scholar 

  20. Berkenbosch JW, Fichter CR, Tobias JD: The correlation of the bispectral index monitor with clinical sedation scores during mechanical ventilation in the pediatric intensive care unit. Anesth Analg 2002, 94: 506-511.

    Article  Google Scholar 

  21. Kroll W, List WF: Is sufentanil suitable for long-term sedation of a critically ill patient? Anaesthesist 1992, 41: 271-275.

    CAS  Google Scholar 

  22. Wappler F, Scholz J, Prause A, Mollenberg O, Bause H, Schulte am Esch J: Level concept of analgesic dosing in intensive care medicine with sufentanil. Anasthesiol Intensivmed Notfallmed Schmerzther 1998, 33: 8-26.

    CAS  Google Scholar 

  23. Iserı LT, French JH: Magnesium: nature's physiologic calcium blocker [editorial]. Am Heart J 1984, 108: 188-194.

    Article  Google Scholar 

  24. Ryan MF: The role of magnesium in biochemistry: an overview. Ann Clin Biochem 1991, 28: 19-26.

    Article  Google Scholar 

  25. Mason BA, Standley CA, Irtenkauf SM, Bardicef M, Cotton DB: Magnesium is more efficacious than phenytoin in reducing N-methyl-D-aspartate seizures in rats. Am J Obstet Gynecol 1994, 171: 999-1002.

    Article  CAS  Google Scholar 

  26. Begum R, Begum A, Bullough CH, Johanson RB: Reducing maternal mortality from eclampsia, using magnesium sulphate. Eur J Obstet Gynecol Reprod Biol 2000, 92: 223-224. 10.1016/S0301-2115(99)00274-2

    Article  CAS  Google Scholar 

  27. Attygalle D, Rodrigo N: Magnesium sulphate for control of spasm in severe tetanus. Can we avoid sedation and artificial ventilation? Anaesthesia 1997, 52: 956-962. 10.1111/j.1365-2044.1997.206-az0340.x

    Article  CAS  Google Scholar 

  28. Miranda HF, Bustamante D, Kramer V, Pelissier T, Saavedra H, Paeile C, Fernandez E, Pinardi G: Antinociceptive Ca2+channel blockers. Eur J Pharmacol 1992, 217: 137-141. 10.1016/0014-2999(92)90833-P

    Article  CAS  Google Scholar 

  29. Wong CH, Dey P, Yarmush J, Wu WH, Zbuzek VK: Nifedipine-induced analgesia after epidural after epidural injection in rats. Anesth Analg 1994, 79: 303-306.

    Article  CAS  Google Scholar 

  30. Ascher P, Nowak L: Electrophysiology studies of NMDA receptors. Trends Neurosci 1987, 10: 284-288. 10.1016/0166-2236(87)90174-3

    Article  CAS  Google Scholar 

  31. Koinig H, Wallner T, Marhofer P, Andel H, Horauf K, Mayer N: Magnesium sulfate reduces intra and postoperative analgesic requirements. Anesth Analg 1988, 87: 206-210.

    Google Scholar 

  32. Telci L, Esen F, Akcora D, Erden T, Canbolat AT, Akpir K: Evaluation of effects of magnesium sulphate in reducing intraoperative anaesthetic requirements. Br J Anaesth 2002, 89: 594-598. 10.1093/bja/aef238

    Article  CAS  Google Scholar 

Download references

Author information

Authors and Affiliations


Corresponding author

Correspondence to Dilek Memiş.

Additional information

Competing interests

None declared.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Memiş, D., Turan, A., Karamanlıoğlu, B. et al. Comparison of sufentanil with sufentanil plus magnesium sulphate for sedation in the intensive care unit using bispectral index. Crit Care 7, R123 (2003).

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • DOI: