Introduction
For successful functional endoscopic sinus surgery (FESS), with good surgical outcomes and no major complications, surgeons need to have a good knowledge of nasal and paranasal sinus surgical anatomy, and excellent operative field visualisation. Intra-operative bleeding is the main reason for a poor quality surgical field.
While haemostasis for head and neck surgery relies on vessel ligation and extensive cauterisation, haemostasis for FESS differs in that it relies on vasoconstriction and clotting mechanisms. Controlled hypotension is used to minimise intra-operative blood loss during surgery and improve surgical field quality.Reference Boezaart, van der Merwe and Coetzee1 Various agents have been used for this purpose, including vasodilators (e.g. sodium nitroprussideReference Degoute, Ray, Manchon, Dubreuil and Banssillon2 and nitroglycerineReference Degoute, Dubreuil, Ray, Guitton, Manchon and Banssillon3), beta-adrenergic antagonists (e.g. propranolol and esmolol),Reference Degoute4 high doses of potent inhaled anaesthetics (e.g. isoflurane)Reference Pavlin, Colley, Weymuller, Van Norman, Gunn and Koerschgen5 and magnesium sulphate.Reference Elsharnouby and Elsharnouby6 Major disadvantages caused by these agents are reflex tachycardia, tachyphylaxis and rebound hypertension, which may enhance venous bleeding and reduce visibility. High doses of inhalation anaesthetics may prolong the hospital stay.Reference Dal, Celiker, Ozer, Basgul, Salman and Aypar7 Other drawbacks are cyanide toxicity due to sodium nitroprussideReference Amaranath and Kellermeyer8 and myocardial depression due to esmolol.Reference Degoute, Ray, Manchon, Dubreuil and Banssillon2
Dexmedetomidine is a potent, highly selective alpha-2 adrenoceptor agonist, with a receptor affinity eight times higher than that of clonidine.Reference Coursin, Coursin and Maccioli9 Central nervous system stimulation of postsynaptic alpha-2 receptors leads to the inhibition of sympathetic activity, which decreases noradrenaline release, blood pressure and heart rate. It also causes sedation, analgesia and vasoconstrictive effects.Reference Gerlach and Dasta10 The hypotensive action and hemodynamic stability thereby reduce intra-operative bleeding. In addition, dexmedetomidine has no effects on the respiration system,Reference Carollo, Nossaman and Ramadhyani11 and its use decreases the requirement for other anaesthetics and analgesics. Dexmedetomidine is a more effective sedative and analgesic agent than clonidine. The most common adverse effects of dexmedetomidine are hypotension and bradycardia, which are usually related to the rate of administration and dosage.Reference Easley12
This study aimed to systematically review randomised, controlled trials that investigated dexmedetomidine and its effectiveness on operative field quality for FESS.
Materials and methods
Electronic systematic searches for randomised, controlled trials were conducted, without language, publication year or publication status restrictions. The date of the last search was 9 May 2014. A literature search was performed using Ovid Medline and Embase databases, with the following Medical Subject Heading terms and key words: ‘endoscopic sinus surgery’ in combination with ‘dexmedetomidine’, ‘Precedex’, ‘Dexdor’ and ‘Dexdomitor’.
Randomised, controlled trials that fulfilled the following inclusion criteria were included: (1) studies that comprised an adult population aged 18 years or older; (2) studies that involved patients undergoing FESS; (3) studies that involved a comparison of the use of dexmedetomidine with placebo, no treatment or other anaesthetic agents; and (4) studies that reported on the quality of the operative field. References of included studies and additional sources were searched to identify any missing published or unpublished trials.
When studies were sufficiently homogeneous, data were pooled for meta-analysis. The primary outcomes were operative field quality and intra-operative bleeding. Secondary outcomes were operative time and adverse events. Standardised mean difference and 95 per cent confidence intervals (CIs) were used for the continuous data. Risk ratio and 95 per cent CIs were used for the dichotomous data. A random effect model was used. Statistical assessments were performed using Review Manager (‘RevMan’) software, version 5.1.6.13 The significance of any discrepancies in estimates of treatment effects from different trials was assessed by means of Cochran's Q test for heterogeneity and by a measure of the I2 statistic. I2 values of less than 40 per cent, 40–60 per cent, or greater than 60 per cent represent low, moderate or substantial heterogeneity respectively.Reference Higgins and Green14 The quality of included studies was assessed by evaluating risks of bias, guided by the Cochrane Handbook for Systematic Reviews of Interventions.Reference Higgins and Green14
Results
Study selection
A total of 794 articles were retrieved from Ovid Medline and 159 were retrieved from Embase; 729 of these were removed in first-level title screening because they were duplicate or irrelevant studies. Eight studies were removed after screening of the abstracts. One additional study was excluded as it was presented at a conference without publication, and essential data could not be obtained.Reference Malhotra, Kumar, Arora and Gupta15 Five studies were included in the review.Reference Gao, Xu, Wang, Lv, Yang and Shang16–Reference Shams, El Bahnasawe, Abu-Samra and El-Masry20 Characteristics of the five included studies are displayed in Table I. A flow chart of study retrieval and selection is shown in Figure 1.
Flow chart of study retrieval and selection.
Characteristics of included studies
Pts = patients; FESS = functional endoscopic sinus surgery
Participants
There were 254 participants in total in the included studies. The mean patient age (reported in four of the trialsReference Guven, Demiraran, Sezen, Kepek and Iskender17–Reference Shams, El Bahnasawe, Abu-Samra and El-Masry20) was 26.3 years; 56.1 per cent of the patients (reported in four of the trialsReference Gao, Xu, Wang, Lv, Yang and Shang16–Reference Lee, Kim, Park, Jeon, Kim and Joo19) were men.
Comparisons and effects of interventions
Two studies compared dexmedetomidine with saline.Reference Guven, Demiraran, Sezen, Kepek and Iskender17, Reference Ke and Pen18 One study compared dexmedetomidine plus sevoflurane with sevoflurane alone.Reference Gao, Xu, Wang, Lv, Yang and Shang16 Two studies compared dexmedetomidine with other anaesthetic drugs.Reference Lee, Kim, Park, Jeon, Kim and Joo19, Reference Shams, El Bahnasawe, Abu-Samra and El-Masry20
Four studiesReference Gao, Xu, Wang, Lv, Yang and Shang16, Reference Guven, Demiraran, Sezen, Kepek and Iskender17, Reference Lee, Kim, Park, Jeon, Kim and Joo19, Reference Shams, El Bahnasawe, Abu-Samra and El-Masry20 scored the quality of the surgical field using a predefined six-point scale, as per Fromme et al.,Reference Fromme, MacKenzie, Gould, Lund and Offord21 as follows: 0 = no bleeding; 1 = slight bleeding with no suctioning of blood required; 2 = slight bleeding with occasional suctioning required (surgical field is not threatened); 3 = slight bleeding with frequent suctioning required and bleeding that threatens the surgical field seconds after suction is removed; 4 = moderate bleeding with frequent suctioning required and bleeding that threatens the surgical field directly after suction is removed; and 5 = severe bleeding with constant suctioning required and bleeding that appears faster than can be removed by suction, with the surgical field severely threatened.
Intra-operative bleeding was reported in three studies.Reference Ke and Pen18–Reference Shams, El Bahnasawe, Abu-Samra and El-Masry20 Two studies reported the volume of blood loss,Reference Lee, Kim, Park, Jeon, Kim and Joo19, Reference Shams, El Bahnasawe, Abu-Samra and El-Masry20 while the other reported the number of cases with nasal bloody exudate.Reference Ke and Pen18 The operative time was reported in three studies.Reference Guven, Demiraran, Sezen, Kepek and Iskender17, Reference Lee, Kim, Park, Jeon, Kim and Joo19, Reference Shams, El Bahnasawe, Abu-Samra and El-Masry20 Adverse events were reported in four studies.Reference Gao, Xu, Wang, Lv, Yang and Shang16, Reference Guven, Demiraran, Sezen, Kepek and Iskender17, Reference Lee, Kim, Park, Jeon, Kim and Joo19, Reference Shams, El Bahnasawe, Abu-Samra and El-Masry20
Dexmedetomidine versus placebo
Two studies compared the use of dexmedetomidine with saline.Reference Guven, Demiraran, Sezen, Kepek and Iskender17, Reference Ke and Pen18 The benefit of dexmedetomidine over placebo was shown in both studies; however, data could not be pooled for meta-analysis.
Guven and colleaguesReference Guven, Demiraran, Sezen, Kepek and Iskender17 assessed the surgical field using the Fromme et al. scale.Reference Fromme, MacKenzie, Gould, Lund and Offord21 Surgical field visibility was better in the dexmedetomidine group (scores of 1.4 ± 1.27 vs 3.15 ± 0.74; p = 0.019).Reference Guven, Demiraran, Sezen, Kepek and Iskender17 The operative time was similar between groups (92.25 ± 27.21 minutes vs 90.75 ± 19.34 minutes; p = 0.13).
Ke and PenReference Ke and Pen18 reported the number of patients with significant bleeding. The results favoured the use of dexmedetomidine, as none of the patients in the dexmedetomidine group required packing.Reference Ke and Pen18 Sixty per cent of the dexmedetomidine group, compared to 90 per cent of the saline group, had intra-operative bleeding (p < 0.05). Operative time was not assessed.
Dexmedetomidine versus no treatment
One trial compared dexmedetomidine plus sevoflurane with sevoflurane alone.Reference Gao, Xu, Wang, Lv, Yang and Shang16 The operative field quality score was significantly better in the dexmedetomidine plus sevoflurane group (score of 1.25 ± 0.4) than in the sevoflurane alone group (score of 2.34 ± 0.6) (p < 0.05). Operative time was not assessed.
Dexmedetomidine versus esmolol
One trial compared dexmedetomidine with esmolol.Reference Shams, El Bahnasawe, Abu-Samra and El-Masry20 Shams and colleagues concluded that both drugs enabled hypotensive anaesthesia and provided a good quality operative field for FESS, with similar surgical field scores (2.81 ± 0.74 vs 2.53 ± 0.93; p = 0.18) and intra-operative blood loss (214.69 ± 158.78 ml vs 186.47 ± 176.98 ml; p = 0.50). The operative time was similar in both groups (61.19 ± 26.6 minutes vs 65.12 ± 23.05 minutes; p = 0.52).
Dexmedetomidine versus remifentanil
One trial compared dexmedetomidine with remifentanil.Reference Lee, Kim, Park, Jeon, Kim and Joo19 Lee and colleagues concluded that both dexmedetomidine and remifentanil were effective in providing a good quality operative field, with similar surgical field scores (2.0 ± 0.5 vs 2.0 ± 0.5; p > 0.05) and similar intra-operative blood loss (130.6 ± 26.8 ml vs 131.4 ± 22.5 ml; p > 0.05). The operative time was similar between the two groups (88.1 ± 14.3 minutes vs 90.0 ± 13.6 minutes; p > 0.05).
Heart rate and mean arterial pressure
Heart rate and mean arterial pressure data for the dexmedetomidine groups and control groups of the individual studies are displayed in Table II. When compared to the placebo or ‘no treatment’ groups, dexmedetomidine significantly decreased heart rate and mean arterial pressure. When compared to esmolol or remifentanil, there was no significant difference between groups. When data were pooled for meta-analysis, the pooled results significantly favoured dexmedetomidine over placebo in terms of the control of heart rate (mean difference −4.56 bpm; 95 per cent CI −5.87, −3.26) and mean arterial pressure (mean difference −6.25 mmHg; 95 per cent CI −9.80, −2.70) (Figure 2).
Forest plots comparing dexmedetomidine with placebo groups for (a) heart rate and (b) mean arterial pressure. SD = standard deviation; IV = inverse variance; CI = confidence interval
Heart rate and mean arterial pressure in dexmedetomidine and control groups
Bpm = beat per minute; SD = standard deviation
Adverse events
Adverse events were reported in four studies.Reference Gao, Xu, Wang, Lv, Yang and Shang16, Reference Guven, Demiraran, Sezen, Kepek and Iskender17, Reference Lee, Kim, Park, Jeon, Kim and Joo19, Reference Shams, El Bahnasawe, Abu-Samra and El-Masry20 All events were minor and successfully managed. When compared to the placebo group, Guven et al. reported fewer events in the dexmedetomidine group for nausea, vomiting and tachycardia, but a greater number of hypotension events.Reference Guven, Demiraran, Sezen, Kepek and Iskender17 When compared to the no treatment group, Gao et al. reported a significantly lower number of events in the dexmedetomidine group for delirium, pain, chill and respiratory inhibition.Reference Gao, Xu, Wang, Lv, Yang and Shang16 When compared to esmolol and sevoflurane, there were no significant differences in the number of adverse events reported in any of the dexmedetomidine groups.Reference Lee, Kim, Park, Jeon, Kim and Joo19, Reference Shams, El Bahnasawe, Abu-Samra and El-Masry20
Quality of included studies
Two studies could not be assessed because of language difficulties.Reference Gao, Xu, Wang, Lv, Yang and Shang16, Reference Ke and Pen18 All three studies assessed for quality had a low risk of randomisation and allocation concealment.Reference Guven, Demiraran, Sezen, Kepek and Iskender17, Reference Lee, Kim, Park, Jeon, Kim and Joo19, Reference Shams, El Bahnasawe, Abu-Samra and El-Masry20 Two studies were double blind.Reference Guven, Demiraran, Sezen, Kepek and Iskender17, Reference Lee, Kim, Park, Jeon, Kim and Joo19 Shams and colleagues performed a single-blind study; however, the anaesthetist in that study (who was not blinded) assessed other objective outcomes that were not the outcomes of interest.Reference Shams, El Bahnasawe, Abu-Samra and El-Masry20 Intra-operative bleeding and operative field quality were assessed by the surgeon who was blinded to the agent used.Reference Shams, El Bahnasawe, Abu-Samra and El-Masry20 There was no drop-out in any studies because all studies investigated intra-operative outcomes. All studies had a low risk of reporting bias. A summary of the risks of bias for each study are displayed in Figure 3.
Summary of risks of bias for each relevant study.
Discussion
Dexmedetomidine has been shown to be beneficial in decreasing intra-operative blood loss and improving the quality of the operative field for FESS. This is because dexmedetomidine is a potent alpha-2 adrenoceptor agonist and its action inhibits noradrenaline release. This causes decreased heart rate and lower blood pressure, with vasoconstrictive effects. This hemodynamic stability leads to better visualisation of the surgical field. In contrast to alpha-2 adrenoceptor agonists, inhalational anaesthetics decrease blood pressure, causing vasodilatory effects and resulting in reflex tachycardia. Simply decreasing mean arterial pressure does not always bring about improved visibility, particularly when patients have increased heart rate and vasodilatation in nasal mucosa. Jacobi and colleagues reported that controlled hypotension did not improve surgical field conditions or intra-operative blood loss when sodium nitroprusside was used in FESS, although mean blood pressure was around 65 to 75 mmHg throughout the surgery.Reference Jacobi, Bohm, Rickauer, Jacobi and Hemmerling22
The findings revealed that dexmedetomidine use did not reduce operative time compared to placebo. Although the visibility of the operative field does improve, this may not be sufficient to affect the speed of surgery. Another possibility is that operative time correlates with the extension of FESS. ‘Full-house’ FESS (entire paranasal surgery) may not proceed as initially planned if the visibility of the operative field is compromised. Surgeons may instead perform limited FESS or terminate the procedure in order to prevent further complications.
Esmolol is a beta-adrenergic antagonist that produces negative chronotropic and ionotropic effects, resulting in a decrease of cardiac output. Once cardiac output is decreased, arterial blood pressure is also decreased, and a stable course of controlled hypotension and an excellent operative field can be achieved.Reference Boezaart, van der Merwe and Coetzee1, Reference Degoute, Ray, Manchon, Dubreuil and Banssillon2 Remifentanil hydrochloride is a potent, short-acting synthetic opioid analgesic drug. It provides higher surgical field visibility because of its ability to decrease heart rate, cardiac output and blood pressure.Reference Degoute, Ray, Manchon, Dubreuil and Banssillon2, Reference Eberhart, Folz, Wulf and Geldner23, Reference Manola, De Luca, Moscillo and Mastella24 When compared to esmolol and remifentanil, dexmedetomidine was reported as similarly effective for improving visualisation.
Better quality of the operative field in the dexmedetomidine group than the saline group was also reported by Malhotra and colleagues (p = 0.011; assessed using the Fromme et al. scaleReference Fromme, MacKenzie, Gould, Lund and Offord21).Reference Malhotra, Kumar, Arora and Gupta15 However, that study, presented at the European Society of Anaesthesiology Congress, has only been published in abstract form, and data could not be obtained; hence, the study was not included. Operative time was not assessed.
Two additional randomised trials assessed the effect of dexmedetomidine on intra-operative bleeding and improved operative field quality for septoplasty.Reference Ayoglu, Yapakci, Ugur, Uzun, Altunkaya and Ozer25, Reference Durmus, But, Dogan, Yucel, Miman and Ersoy26 Similar findings of the beneficial effects of dexmedetomidine were reported.
Midazolam is commonly used as a sedative for sinonasal surgery performed under local anaesthesia. There have been two randomised trials comparing midazolam with dexmedetomidine.Reference Balaraju, Pathan, Thomas, Kantha, Sheetal and Yogesh27, Reference Ragab, El Shamaa and Ibrahim28 Although the aim was to investigate the sedative effect, dexmedetomidine was shown by these two studies to have decreased intra-operative bleeding and improve the operative field quality as a result of the hemodynamic stability it produced.
Major adverse events of dexmedetomidine use have not been reported. However, dexmedetomidine may induce bradycardia and hypotension. It is recommended that patients given dexmedetomidine are closely monitored. Bradycardia of less than 50 beats per minute and mean arterial pressure of less than 50 mmHg should be corrected.
One limitation of this systematic review is that there are co-interventions of other anaesthetic drugs and rescue medicine. These confounders may have affected the outcomes. However, these confounders were given to both arms of the studies. Furthermore, group allocation was random and the outcome assessors were blinded. In addition, the studies included were of good quality, with no bias.
Acknowledgement
The authors would like to thank Dr Chitsuda Wacharasindhu for kindly interpreting two Chinese articles.
