A COMPARATIVE STUDY OF INTRAMUSCULAR KETAMINE AND A COMBINATION OF INTRAMUSCULAR DEXMEDETOMIDINE AND KETAMINE AS PREMEDICATION IN PAEDIATRIC ANESTHESIA

combina�on of ketamine and dexmedetomidine in children scheduled for surgery under general anaesthesia.


INTRODUCTION
Relieving pre and postopera ve anxiety is an important concern for a paediatric anaesthesiologist. Anxiety can lead to distress and makes the control of postopera ve pain 1 difficult. Forced induc on of anaesthesia may cause 2 personality and behavioral changes in children. Various routes of drug administra on are available for premedica on. Oral and sublingual prepara ons are painless, but are slow in onset, have poor bioavailability, and can be spi ed out. Rectal prepara ons can be uncomfortable. Whereas, nasal prepara ons can be 3 irrita ng to the pa ent. Intravenous (IV) administra on is painful and incites fear of subsequent contact with 4 healthcare professionals. Intramuscular (IM) medica ons cause slight pain at the injec on site, however easy administra on, good bioavailability, and not requiring intravenous access make it a fair op on for premedica on in a resource-constrained se ng. Among various op ons for intramuscular use, ketamine is an established premedicant. The primary mechanism of ac on of ketamine appears to be through its NMDA receptor antagonism on the central nervous system (CNS, and spinal cord). Ketamine produces a 5 'dissocia ve' anaesthe c state. Dexmedetomidine is a rela vely new selec ve alpha-2 adrenoceptor agonist and it provides seda on and anxiolysis via receptors within the locus ceruleus and analgesia via receptors in the spinal cord. It is devoid of respiratory depressant effect rendering it poten ally useful 1 for anaesthe c premedica on. Dexmedetomidine however, as a sole seda ve agent has not been uniformly successful for invasive procedures. To overcome some of the pi alls with dexmedetomidine as the sole agent, its 6 combina on with ketamine has been tried. This study was undertaken to evaluate pre-opera ve seda ve effect, anxiety level changes, ease of child-parent separa on of preopera ve intramuscular ketamine compared with intramuscular combina on of ketamine and dexmedetomidine in children scheduled for surgery under general anaesthesia.

METHODOLOGY
This clinical compara ve study was conducted in the elec ve theatre of BPKIHS, Dharan over the dura on of one year from 2017 to 2018. Ethical approval of this study was obtained from the BPKIHS Ins tu onal Review Commi ee (IRC no: IRC/1369/018) and informed wri en consent for the procedure was obtained from the parents of all the eligible pa ents and assent from the pa ents whenever possible. Every pa ent had the right to withdraw from the study at any me. Children with ASA physical status I and II, aged 2-10 years undergoing elec ve surgery under general anaesthesia were included. Children with history of allergy or hypersensi vity reac on to dexmedetomidine or ketamine and atropine, children with cardiac arrhythmia, congenital heart disease, ac ve upper respiratory tract infec on within 2 weeks, known lung disease or airway abnormali es, seizure disorder and mental retarda on were excluded. In addi on, children under medica ons that could provoke seizures, cause hypotension and bradycardia were also excluded from the study. All together, sixty children were involved 30 children in group A and 30 in group B. Group A children received 3mg /kg ketamine and Group B children received combina on of 2mg/kg undiluted ketamine and 1mcg/kg undiluted dexmedetomidine intramuscularly 20 minutes prior to planned induc on of anaesthesia via 26 G needle into the gluteal muscle. All eligible children were evaluated and their parents explained about the nature of the study during preanaesthe c evalua on in the evening prior to the day of the surgery. Informed wri en consent was obtained. The children were kept nil per oral according to standard paediatric fas ng guidelines and premedica on was omi ed. Baseline systolic and diastolic blood pressure together with heart rate was recorded on the same day in respec ve wards where the child was admi ed. On the day of surgery in the preopera ve holding area, each child was assessed for level of seda on using Richmond Seda on Scale, by the inves gator responsible for observing the outcome. The baseline pulse rate, oxygen satura on (Spo ) 2 were noted, and the study solu on was administered. Pulse and Spo and the preopera ve seda on score was 2 evaluated every 5 min a er the study drug was given ll induc on of anaesthesia. Separa on from the parent was evaluated by the separa on score described by Venham LL, Gaulin-Kremer E, Munster E, Bengston-Audia D, Cohan J. 7. Availability of oxygen, suc on apparatus, self-infla ng resuscita on bag, mask, equipment for intuba on and drugs for resuscita on were ensured at all mes. A er observing for 20 min in the preopera ve holding area, the child was brought into the opera on theatre for induc on of anaesthesia and pa ent monitor was a ached. HR, SBP, DBP, MAP and Spo was recorded at 1 min, 3 min and 5 min 2 a er induc on. Then every 5 min un l the end of surgery. Intravenous cannula on was performed before the induc on of anaesthesia in children with sa sfactory acceptance score. A four-point evalua on system was used to evaluate the acceptance of the intravenous cannula 8. An acceptance score of 3 or 4 was taken as sa sfactory. Anaesthesia in children with inadequate seda on was induced inhala onally with sevoflurane in oxygen and their peripheral access was secured by expert hands. A er securing the peripheral venous access, injec on glycopyrrolate 4mcg/kg was administered IV in all pa ents. In children accep ng intravenous cannula on, induc on of anaesthesia was achieved with intravenous bolus of injec on propofol 2-4mg/kg and fentanyl 2mcg/kg. A er achieving adequate depth of anaesthesia, an appropriately sized laryngeal mask airway (LMA) was inserted using standard technique. Anaesthesia was maintained with 1.0-2.0 MAC end-dal concentra on of sevoflurane in 100% oxygen and intravenous bolus of fentanyl was administered for analgesia as required. Hydra on was provided with lactated Ringer's solu on was used based on hourly fluid requirements, deficits caused by restric on of food and  fluids, third-space losses according to the surgical procedure. On comple on of the surgery, sevoflurane was discon nued and LMA was removed. The requirement of propofol for induc on, fentanyl for maintaining analgesia and percentage of sevoflurane required for maintenance was noted. Hypotension was defined as a decrease of systolic blood pressure of more than 20% from the baseline readings and was treated with intravenous fluid bolus of 10ml/kg. Similarly, bradycardia was defined as HR less than 20% of the baseline readings and was treated with IV glycopyrrolate 4mcg/kg. Any complica on was managed with the standard hospital protocol. The child was then transferred to the recovery room, where a nurse monitored the child for any complica ons. Nausea was defined as a subjec vely unpleasant sensa on associated with awareness of the urge to vomit, whereas vomi ng was defined as the forceful expulsion of gastric contents from the mouth. The main outcome variables were seda on score (Richmond Agita on Seda on Scale), IV cannula acceptance (IV cannula Acceptance Score), and parental separa on (Separa on Score). Secondary outcome variables were hemodynamic response during induc on; incidence of hypotension and bradycardia; post-opera ve nausea, vomi ng and shivering. The sample size es ma on was based on a previous study 9 which showed 85% pa ents being adequately sedated in the combina on group(intranasal ketamine and dexmedetomidine) and 66% in the other group(ketamine). With a power of 80%, alpha value 0.05 and the percentage of children sedated in each group was taken in considera on and sample size of 23 was calculated to be 23 for each group. To account for the possible drop outs and correct bias the sample size was rounded off at 30 and 30 pa ents were taken in either group. Propor on, percentage, mean, median, S.D., and interquar le range were calculated for descrip ve analysis. Chi-square and independent t-test were applied depending on the data type.

RESULT
All 60 pa ents completed the study. The demographic parameters including age, sex, height, and weight of the pa ents in both the groups were sta s cally comparable.
The anxiety level between the two groups was comparable before the administra on of premedica on (p = 0.447). A er premedica on, there was no significant difference in the anxiety level between the two groups except at 0 min 1 a er premedica on.
IV cannula acceptance score was significantly higher in Group A compared to group B (p= 0.001). However, the separa on score was significantly higher in group B (p= 0.007). The heart rates at all the other observa on points were sta s cally comparable among the groups (p> 0.05). The baseline mean systolic blood pressure was comparable between the two groups (p> 0.05). However, a er induc on the systolic blood pressure was significantly higher in group A pa ents at 1 min, 10 min and 15 min. The mean systolic pressure was comparable at 5 min a er induc on (p> 0.05). Similarly, mean diastolic blood pressure at 1min and 3 min a er induc on was significantly higher in group A than in group B (p 0.05). Therea er, the mean diastolic pressure was comparable between the groups (p> 0.05). MAP was significantly higher in group A than in group B (p= 0.04) 3 min post induc on of anaesthesia. Average mean arterial pressure was comparable between the two groups (p> 0.05) at all the other points of measurement.

Original Research Ar cle
The mean oxygen satura on values were comparable between the two groups at all me points of measurement (p>0.05). Sevoflurane requirement was significantly reduced in group B as compared to group A and was sta s cally significant (p<0.05) at all me points except 1 min following induc on of anaesthesia. There was no incidence of postopera ve nausea and vomi ng in any of the groups. Shivering was observed in 2 cases of group A and 2 cases of group B (Table 4)

DISCUSSION
The present study did not find any significant difference between the level of seda on in the children undergoing surgery under general anesthesia who received intramuscular ketamine alone. Similar effects, in terms of seda on can be elucidated by the seda ve proper es of both ketamine as well as dexmedetomidine. However, parental separa on was significantly be er in the combina on group as compared to ketamine alone. Expectedly, ketamine induced delirium could have been a enuated by dexmedetomidine thereby providing a superior parental separa on score observed in the combina on group. Conversely, anxiolysis in terms of ease of IV cannula on was superior in the ketamine group. This can be construed by the strong analgesic proper es of ketamine and a higher dose used in the group. Previous studies on oral ketamine and intranasal dexmedetomidine have found that the rate of successful venous cannula on was 47%, 68% and 80% with dexmedetomidine, ketamine and the combina on groups 10 respec vely. The rate of sa sfactory parental separa on was comparable among the groups. These findings are contrary to the findings of our study where the IV acceptance was be er with the ketamine group whereas the combina on group had superior outcomes in terms of parental separa on. This discrepancy between their findings and ours may also have been as a result of different routes of drug administra on and me allowed for premedica on. Yuen VM, Hui TW, Irwin MG, Yuen MK studied intranasal dexmedetomidine in a dose of 0.5 mcg/kg and 1 mcg/kg and oral midazolam 0.5mg/kg for pre-medica on in children and concluded that intranasal dexmedetomidine produces more seda on than oral midazolam, but with acceptable 1 coopera on. It is difficult to compare our findings with theirs as we used intramuscular route, a combina on of dexmedetomidine and ketamine.
André P Schmidt found that children receiving clonidine or dexmedetomidine preopera vely have similar levels of anxiety and seda on postopera vely as those receiving midazolam on premedica on with oral midazolam 0.5mg/kg, oral clonidine 4mcg/kg, or transmucosal 1 1 dexmedetomidine 1mcg/kg. Our findings are similar to the findings of both the above studies in elici ng the anxioly c effect of dexmedetomidine. The fact that the me allowed for premedica on in our study was significantly less, most likely accounts for the delayed seda on in the combina on 1,11 group of our study. Also, we used intramuscular route for premedica on, which differs from other routes in terms of bioavailability and further, there are no studies sugges ng the equivalent dosing of dexmedetomidine through various routes. Both intranasal dexmedetomidine and ketamine have been found equally efficacious as a premedicant in children undergoing MRI, with comparable reported anesthesiolgist's 12 sa sfac on. It is difficult to compare this study with ours as our study differs from theirs in many respects including use of a different dose and route of drug administra on and absence of a placebo group in our study. Further, our study did not compare anesthesiologists' acceptance. Another study comparing the analgo-seda ve effects of oral dexmedetomidine and ketamine reported a delayed onset of seda on with 3mcg /kg and 4mcg /kg of dexmedetomidine rela ve to ketamine 8mg/kg and 13 5mcg/kg of dexmedetomidine. Our study differs from this study not only in terms of route of administra on but also in dose of the drug used to premedicate the children. Dexmedetomidine alone was not used in our study. Furthermore, we used intramuscular route and a lower dose of the drugs. Another study reported that a faster seda on with ketamine at 10 min with comparable results at 20 and 30 min 14 respec vely. In the study, the effects of intranasal dexmedetomidine 3mcg/kg and ketamine 7 mg/kg was compared for procedural seda on in school aged children undergoing MRI. Again, the route of drug administra on and drug dosage used were different from our study. This study used much higher dose of both study drugs as compared to our study. All the above three studies however, found c o m p a r a b l e s e d a o n b e t w e e n k e t a m i n e a n d 12,14 dexmedetomidine, which is similar our own findings. We found comparable seda on scores at all me points except at 10 min sugges ng a delayed onset of seda on in the combina on group. Which is concordant with the findings 1 5 of Mason KP, Lubisch NB, Robinson F, Roskos R and Ibrahim M.14 The delayed onset of seda on in our study can be a ributed at least par ally to the lower dose of dexmedetomidine that we used and also the less me that we allowed for seda on. Though the route of drug administered in their study differed from our study, seda on and anxiolysis were comparable, sugges ng intramuscular route being an equally effec ve alterna ve for premedica on in children in our resource constrained se ng where atomisers may not be readily available for intranasal administra on of drugs and where securing an intravenous cannula can be difficult and trauma sing to a struggling child. Further, there is a paucity of research comparing ketamine and combina on of ketamine and dexmedetomidine as pediatric premedicants. Expectedly, we observed increased heart rate in the children premedicated with only ketamine. In both the groups the heart rate decreased from the baseline and was the least a er 20 min of premedica on. However, episodes of bradycardia requiring management with atropine were not encountered. This finding is in contrary to the findings of the study by Scheinin H, Jaakola ML, Sjövall S, Ali-Melkkilä T, Kaukinen S, Turunen J. 16 where bradycardia was observed more frequently in dexmedetomidine pa ents; 20% in those pa ents who received intramuscular injec on of 2.5mcg/kg dexmedetomidine administered 60 min before induc on of anesthesia and 33% in those pa ents who received intramuscular dexmedetomidine 2.5 mcg/kg and intravenous fentanyl 1.5mcg/kg 60 min before induc on of 14 anesthesia. Ibrahim also observed bradycardia in two pa ents premedicated with dexmedetomidine (HR <20% of baseline) but it did not require treatment with atropine. The combina on of ketamine with dexmedetomine and use of glycopyrrolate prior to induc on of anesthesia may have counteracted the bradycardia caused by dexmedetomidine in our study. In the same line with our observa ons, Yuen VM, Hui TW, 1 Irwin MG, Yuen MK also observed significant reduc on of heart rate from baseline at 45 and 60 min a er administra on of dexmedetomidine 0.5mcg/kg and 1mcg/kg which however did not any require treatment. Gyanesh P, Haldar R, Srivastava D, Agrawal PM, Tiwari AK, 1 2 Singh PK also found that the heart rate and blood pressure remained similar throughout the procedure and for 3 hr a er the comple on of MRI in both ketamine and dexmedetomidine group while in the study conducted by 15 Mason KP, Lubisch NB, Robinson F, Roskos R none of the pa ents given intramuscular dexmedetomidine had bradycardia, hypertension, or oxygen desatura on. In our study, baseline MAP and DBP were raised in ketamine treated children than in children treated with the combina on of ketamine and dexmedetomidine. We a ribute this finding to the cardiodepressant ac on of dexmedetomidine which may have a enuated the rise in blood pressure caused by ketamine. Furthermore, we observed a lesser altera on in hemodynamics in children premedicated with ketamine and dexmedetomidine combina on as compared to ketamine alone. The findings of Ibrahim 14 and Gyanesh P, Haldar R, Srivastava D, Agrawal 1 2 PM, Tiwari AK, Singh PK were in line with our findings. Also, children premedicated with combina on of ketamine and dexmedetomidine had lesser percent of sevoflurane required than those with ketamine alone. It was observed to be sta s cally significant 3 min a er induc on of anesthesia and therea er. Postopera ve shivering was observed only in 4 cases, of which 2 were in the ketamine and 2 in the combina on group. No other adverse events were observed in our study. Dexmedetomidine is considered a weak an eme c but ketamine is a known emetogenic therefore, our study showed no incidence of nausea and vomi ng in children premedicated with combina on of ketamine and dexmedetomidine. Also, the children received intravenous ondansetron which may have a enuated nausea and vomi ng postopera vely in both the groups.

CONCLUSION
In conclusion, combina on of intramuscular ketamine 2mg/kg and dexmedetomidine 1mcg/kg is a good anxioly c and seda ve comparable with intramuscular ketamine 3mg/kg with superior anxiolysis in terms of parental separa on in children and comparable adverse events as compared to ketamine alone. However, intramuscular ketamine has a be er anxiolysis in terms of IV cannula on.

LIMITATIONS OF THE STUDY
The major limita on in our study is the short me between administra on of premedicant prior to induc on of anaesthesia. There was only 20 min before inducing general anesthesia a er premedica on considering the me constraint and delay in star ng the surgery. Had a longer period been allowed, results for seda on could have been different. Similarly, we used intramuscular route which though effec ve, is not ideal in children. Further we used a quasi experimental design.