Background

Additives to local anesthetic mixture (LAM) during regional anesthesia had been tried for various surgical procedures as documented in recent studies (Li et al. 2016), to increase potency, extent, and improve onset of local anesthesia (Gentili et al. 2016; Pierce et al. 2016).

Fentanyl is considered as one of the most successful opioid analgesics. It is a synthetic, lipophilic phenylpiperidine opioid agonist with analgesic and anesthetic properties. Fentanyl selectively binds to the mu-receptor in the central nervous system (CNS) (Chen and Gupta 2014). Stimulation of the mu-subtype opioid receptor stimulates the exchange of GTP for GDP on the G-protein complex and subsequently inhibits adenylate cyclase, so decreasing in intracellular cAMP and leads to a reduction in the release of neurotransmitters (Celver et al. 2004). The analgesic effect of fentanyl is by two actions: it induces the opening of potassium channels and blocks the opening of N-type voltage-gated calcium channels, resulting in hyperpolarization and reduced neuronal excitability (Rodrigues et al. 2005; Kuhar et al. 2015).

Midazolam is a short-acting benzodiazepine derivative with anxiolytic, amnestic, hypnotic, anticonvulsant, and sedative properties (Griffin 3rd et al. 2013). Midazolam binds to the benzodiazepine receptor at the gamma-aminobutyric acid (GABA) receptor in CNS leading to increased opening of chloride channels, membrane hyperpolarization, and inhibitory effect of GABA in the CNS (Khom et al. 2006; Beltrán González et al. 2014). Midazolam may also interfere with the reuptake of GABA, thereby causing accumulation of GABA in the synaptic cleft (Baburin et al. 2008). The GABA receptors are crucial for induced sedation through α1-receptors and the anxiolytic effects through α2-receptors (Naguib et al. 1995; Newman et al. 2015) and have a role in the regulation of Ca channel activity at legends of peripheral benzodiazepine receptors and in CNS at nanomolar concentration leading to Ca-mediated increase in K conductance in CA1 cells in hippocampal slices with adenosine benzodiazepine actions via adenosine uptake inhibition or enhancing adenosine release. Neuraxial midazolam acts on the benzodiazepine receptors on the gray matter of the spinal cord, the highest concentration of which is found on the lamina II of the dorsal horn. The analgesic effect of neuraxial midazolam is caused by the spinal suppression of sensory functions and its antinociceptive effect mediated by GABAergic mechanisms (Edwards et al. 1990; Lötsch et al. 2013).

We hypothesized that the addition of either fentanyl or midazolam will affect the quality of the peribulbar block, prolonging its anesthetic and analgesic duration. This will later help in providing adequate anesthesia, analgesia, and comfort during lengthy ophthalmic operations. The aim of this study was to evaluate the analgesic and hemodynamic outcome of fentanyl or midazolam as an additive to local anesthetic mixture (LAM) for peribulbar anesthesia during posterior segment surgery.

Methods

The study was approved by the Local Ethics Committee and registered with the Pan African Clinical Trials Registry (PACTR201708002496243) and conducted at Anesthesia Department, Cairo University Hospitals, from January 2017 to August 2017. One hundred thirty-two ASA status I to III adult patients, aged 50–75 years scheduled for posterior segment surgery (intraocular foreign body and elective vitrectomy) were included in this prospective, randomized, double-blind trial. The patients signed written fully informed consent for study participation where 12 patients were excluded from the study.

Complete ophthalmological examination was done by the ophthalmologist to exclude complicated vitreous hemorrhage, diagnosis of any associated disorders as well as ophthalmic ultrasound and biometry was done for all cases to measure the axial length and diagnose posterior staphyloma if it was present. All patients were clinically evaluated for fitness for anesthesia and categorized according to ASA grade and patients who had comorbidities underwent preoperative control and continued their treatment during postoperative (PO) course. Patients with impaired orbital/periorbital sensation had history of abnormal bleeding or allergy to local anesthetics and had posterior staphyloma or with axial length > 30 mm were excluded from the study.

Patients were randomized using computer-generated numbers and were assigned randomly using sealed opaque envelopes prepared by a blinded assistant (consort flow diagram)

Patients were categorized into 3 groups (40 for each) according to the type of additive used with the LAM (2 ml of mepivacaine 3%, 1 ml of hyaluronidase containing 150 IU, 3 ml of bupivacaine 0.5%, 1 ml of lidocaine 2%). as follows:

  1. 1-

    Group C (control group): included patients assigned to receive 1 ml plain saline.

  2. 2-

    Group F (fentanyl group): 1 ml saline containing 25 μg fentanyl was used instead of plain saline.

  3. 3-

    Group M (midazolam group):1 ml saline containing 1 mg midazolam was used instead of plain saline.

Anesthetic procedure

No sedative premedication was administered in the preoperative period. In the operating room, an intravenous line was secured and standard monitoring (heart rate, electrocardiography, oxygen saturation, and noninvasive blood pressure) was applied to all the patients. The study solution for peribulbar block was prepared and performed by the anesthesiologist who was not involved in the study. Then, 3 ml of LAM assigned for each group was injected with needle 3-cm length and 27 G inserted at angle of 45° between the caruncle and medial canthal angle till the tip of the needle touched the ethmoid bone then the direction of the needle was changed to 90° with the hub of the needle at level of the iris. Another 3 ml of LAM was injected at the extreme inferotemporal border of the orbit with the same needle directed downward and medially below the globe. Light orbital compression was applied for 1 min; then eye was evaluated 1, 3, 5, and 10 min for the appearance of proptosis and chemosis. After satisfactory sensory and motor block, oxygen 4 L/min was delivered through a nasal cannula to the patient. Surgery was then allowed to proceed.

Study outcomes

Primary outcomes

The onset of eye block (eyelid and globe akinesia) was assessed and recorded throughout 15 min after LAM injection using a three-point scoring or ocular movement scale (0–complete akinesia–limited movement–normal movement) (Sarvela et al. 1993).

Secondary outcomes

  1. 1.

    Duration of the block was calculated since 15-min assessment after LAM injection till complete recovery.

  2. 2.

    Hemodynamic variables were assessed every 15 min during the entire procedure and every 30 min during the first two postoperative (PO) hours. Hypotension and bradycardia were defined as a 20% decrease in MAP and HR in relation to baseline measures.

  3. 3.

    PO analgesia was assessed by using visual analogue score (VAS) every hour for 6 h PO and was scored as 0 if no pain up to 10 which indicates intolerable pain sensation. PO analgesia was provided if VAS was > 3

Statistical analysis

Sample size calculated according to the standard nomogram for power calculation (Kraemer and Theimann 1987) defined a sample size of 36 patients per group giving the trial 80% power and is sufficient to detect a difference at the 5% significance level. Sample size and power were re-calculated and assured using Power and Sample Size Calculation Software program provided by the Department of Biostatistics, Vanderbilt University. Forty patients were included in each group to compensate for dropouts. Obtained data were presented as mean ± SD, ranges, numbers, and ratios. Results were analyzed using one-way ANOVA with post hoc Tukey HSD test and chi-square test (χ2 test). Statistical analysis was conducted using the SPSS (Version 15, 2006) for Windows statistical package. P value < 0.05 was considered statistically significant (Murphy and Myors 2003)..

Results

One hundred thirty-two patients scheduled for posterior segment surgical procedures were enrolled in the study. They were divided randomly into 3 equal groups (40 patients in each group); group C received LAM plus 1 ml plain saline, group F received LAM plus 25 μg fentanyl in 1 ml saline and group M received LAM plus 1 mg midazolam in 1 ml saline. Mean age of studied patients was 58.4 ± 6.6; range 50–75 years. Sixty-eight patients were males and 52 patients were females. Mean body mass index of studied patients was 31.4 ± 3.2; range 24.6–37.5 kg/m2. Forty patients ASA grade I, 64 patients ASA grade II, and 16 patients ASA grade III. Details of enrolment data of patients categorized according to the received local anesthetic mixture (LAM) are shown in Table 1.

Table 1 Enrolment data of studied patients categorized according to received LAM
Full size table

Injection of LAM was conducted successfully for all groups, patients of group F responded significantly (P < 0.05) faster than patients of the other groups with significantly (P < 0.05) fast onset of eyelid and globe akinesia. On the contrary, group M showed non-significantly (P > 0.05) fast onset of eyelid and globe akinesia compared to group C (Fig. 1).

Fig. 1
figure 1

Mean scoring of eyelid akinesia recorded in patients of studied group throughout 15 min after LAM injection

Full size image

Mean of total score of eyelid akinesia recorded throughout 15-min after LAM injection showed significantly (P < 0.05) progressing decrease in the three groups with significantly (P < 0.05) lower score in group F, but non-significantly (P > 0.05) in group M compared to group C with non-significantly (P > 0.05) lower score in group F compared to group M (Fig. 1). Similarly, mean of total score of globe akinesia recorded throughout 15 min after LAM injection showed significantly (p < 0.05) progressing decrease in the three groups with significantly (P < 0.05) lower score in group F, but non-significantly (P > 0.05) in group M compared to group C with non-significantly (P > 0.05) lower score in group F compared to group M (Fig. 2).

Fig. 2
figure 2

Mean scoring of globe akinesia recorded in patients of studied group throughout 15 min after LAM injection

Full size image

All surgeries were conducted uneventfully without intraoperative (IO) morbidities within a mean operative time of 96 ± 6.5; range 80–120 min with non-significant (P > 0.05) difference between studied groups despite being shorter in group F. Preoperative, intraoperative, and 2-h PO hemodynamic measures showed non-significant (P > 0.05) difference between studied groups; despite being lowest in group M (Tables 2 and 3).

Table 2 Intraoperative hemodynamic variables estimated in studied patients according to injected LAM
Full size table
Table 3 Postoperative hemodynamic variables estimated in studied patients categorized according to injected LAM
Full size table

Mean duration till complete recovery of eyelid and globe akinesia was significantly (P < 0.05) longer in groups F and M compared to group C with significantly (P < 0.05) longer duration in group F compared to groups M. Only 46 patients (38.3%) required PO rescue analgesia; 6 patients in groups C and M required it twice, while 40 patients among the three groups required it once with significantly (P < 0.05) lower frequency of patients required rescue analgesia in group F compared to groups C and M and non-significantly (P > 0.05) lower frequency in group M compared to group C. Moreover, the mean number of requests was significantly (P < 0.05) lower in group F compared to group C, but non-significantly (P > 0.05) lower than in group M with non-significantly (P > 0.05) lower mean number of requests in group M compared to group C (Table 4).

Table 4 Postoperative recovery data of studied patients categorized according to injected LAM
Full size table

Discussion

Our study showed that local anesthetic mixture (LAM) with additives, either fentanyl or midazolam, for peribulbar block (PBB) for posterior chamber surgery improved analgesic outcome manifested as significantly shorter duration till onset of motor block and longer duration of anesthesia that allowed completion of surgery despite the long operative time. Moreover, additives allowed significantly longer duration of postoperative (PO) analgesia and significantly lower consumption of rescue analgesia in comparison to control group received plain LAM.

As regards the differential effects of used additives, fentanyl in comparison to midazolam provided significantly faster anesthesia, longer duration of motor block, and is superior in terms of significantly prolonged PO analgesia with significantly lower consumption of rescue analgesia. Moreover, both additives allowed favorable hemodynamic outcomes manifested as stable blood pressure measures and heart rate comparable to the control group (Dopfmer et al. 1996).

These findings illustrated the beneficial effects of adding fentanyl or midazolam to LAM and go in hand with that reported for both additives during various forms of regional anesthesia; where various studies have demonstrated the presence of GABA receptors in peripheral nerves and the action of midazolam on GABA receptors is well established. Extra synaptic receptors for GABA are present on myelinated axons of peripheral nerves (Morris et al. 1983). Cairns et al. observed the presence of GABA receptors within the temporomandibular joint and its activation could decrease the transmission of nociceptive signals (Cairns et al. 1999).

Jarbo et al (2005) found that midazolam (50 mcg × kg−1) in combination with 30 ml of bupivacaine (0.5%) hastened the onset of sensory and motor block, and improved postoperative analgesia when used in brachial plexus block, without producing any adverse events.

Laiq et al (2008) found that bupivacaine (0.5%) in combination with midazolam (50 mcg × kg−1) quickened the onset as well as prolonged the duration of sensory and motor blockade of the brachial plexus for upper limb surgery. It improved postoperative analgesia without producing any adverse events compared to plain bupivacaine (0.5%) in equal volume. Trivedi and Patel (2010) reported a clinical study in 60 ASA I–II patients undergoing elective upper limb orthopedic surgeries. All the patients were divided into two groups (n = 30 each), group C (clonidine), and group M (midazolam) given injection bupivacaine 0.5% plain 20 ml and injection lignocaine 2% plain 10 ml in supraclavicular brachial plexus block. In group C, clonidine 150 mcg and in group M midazolam 5 mg were injected with local anesthetics. All the patients were observed for onset and duration of sensory and motor blockade, sedation score, and postoperative analgesia with visual analogue scale (VAS) score up to 12 h. They found no clinically significant difference in onset and duration of sensory and motor blockade among both the study groups. In group C more, sedation (score 2) was observed compared to group M (score 1) though the data is not statistically significant. Postoperative analgesia was more prolonged in group C (VAS score < or = 3 for 360 min) as compared to group M (VAS score < or = 3 for 300 min) Edwards et al. (1990) found that intrathecal injection of midazolam has produced segmental analgesic effect in both rats and humans. This analgesic effect is not caused by a local analgesic action of the drug and not accompanied by sedation, and this experiment suggested that the spinally mediated analgesia following intrathecal midazolam may be the result of a combination of the drug with spinal cord benzodiazepine receptors.

Moreover, Weigl et al. (2016) and Khezri et al. (2016) reported that addition of intrathecal fentanyl to spinal anesthesia is effective for intraoperative analgesia and decreases opioid consumption after cesarean section and Kumar et al. (2016) found that low-dose bupivacaine with fentanyl is superior to bupivacaine and butorphanol in terms of early PO recovery resulting in early discharge and better outcome of elderly patients with comorbidity. Also, Singh et al. (2016) found fentanyl as an adjuvant to intrathecal ropivacaine prolongs PO pain relief without increasing duration of motor blockade.

The effect of fentanyl could be mediated through a direct action on the peripheral opioid receptors, in the primary afferent tissues (dorsal roots) (Laduron 1984), or through centrally mediated opioid receptor analgesia after its uptake into the systemic circulation.

Karakaya et al (2001) found that addition of fentanyl to bupivacaine in brachial plexus axillary block approach enhanced and prolonged anesthesia and analgesia, increased duration of sensory and motor block and increased the duration of postoperative analgesia.

A synergism between opioids and local anesthetics has been reported (Nishikawa et al. 2000). It appears that opioids and local anesthetics act their action independently via different mechanisms. Local anesthetics block propagation and generation of neural action potentials by a selective effect on Na channels, whereas opioids act on the opioid receptors creating an increase in a K channel conductance. This action results in hyperpolarization of the nerve cell membrane and a decrease in excitability. Although Na channel block is proposed to be the primary mode of action, local anesthetic also has an effect on synaptic transmission. Li et al. (1995) showed that lidocaine inhibits both substance P binding and substance P-evoked increase in intracellular Ca.

Unfortunately, few studies evaluated these LAM additives for ophthalmic surgeries. However, the obtained results go hand in hand with multiple studies that used varied additives to LAM for peribulbar block (PBB) where Reah et al. (1998), Küçükyavuz and Arici ( 2002), Godarzi et al. (2011), Aissaoui et al. (2010), and Messeha and Elhesy (2015) found the addition of vecuronium (Reah et al. 1998), atracurium (Küçükyavuz and Arici 2002; Godarzi et al. 2011), rocuronium (Aissaoui et al. 2010; Messeha and Elhesy 2015) to LAM decreased the onset time of akinesia, improved the quality of globe and lid akinesia, reduced the need for supplementary injections, and provided better surgical conditions. Also, Madan et al. (2001) and Bharti et al. (2002) found that the addition of clonidine to LAM significantly increases the duration of anesthesia and analgesia after PBB with limited side effects. Moreover, Abo El Enin et al. (2009) found addition of fentanyl to LAM decrees the onset and increases the duration of akinesia and improves quality of PO pain in peribulbar block (PBB) and Channabasappa et al. (2013) found addition of dexmedetomidine to lidocaine and bupivacaine in PBB shortens the onset time and prolongs the duration of the block and postoperative analgesia. Also, Sinha et al. (2016) found addition of 50 mg of magnesium sulfate to the lidocaine-bupivacaine mixture for peribulbar block decreases the onset of akinesia without any obvious side effect.

Conclusion

The use of additives to the local anesthetic mixture during peribulbar block for posterior segment surgery provided satisfactory anesthetic outcome than local anesthetic mixture alone. Fentanyl was superior to midazolam as additive to local anesthetic mixture in terms of significantly faster onset and longer duration of block with significantly longer postoperative analgesia and lesser consumption of rescue analgesia. Moreover, both additives provided adjusted hemodynamic variables comparable to control group.

Study limitations

A limitation to our study was the limited sample size in each group. Further investigations are needed on a wider population sample in order to concur our results, to confirm their safety, and to support the absence of systemic complications. More researches may be suggested using different local anesthetics and different doses of midazolam and fentanyl to confirm our data.