Indian Pediatrics 2001; 38: 967-972  

Midazolam Sedation in Mechanically Ventilated Newborns: A Double Blind Randomized Placebo Controlled Trial

Vishal Arya, Siddarth Ramji

From the Neonatal Division, Department of Pediatrics, Maulana Azad Medical College, New Delhi 110 002, India.

Correspondence to: S. Ramji, Professor, Neonatal Division, Department of Pediatrics, Maulana Azad Medical College, New Delhi 110 002, India. Email: [email protected] in

Objective: To determine efficacy of midazolam as a sedative in mechanically ventilated newborns. Design: Double blind randomized placebo controlled trial. Setting: Neonatal Unit of Tertiary Hospital. Outcome: Sedation over 48 h of observation. Methods: Neonates with birth weight <2000 g who were mechanically ventilated within 7 days of life were randomly assigned to midazolam and placebo group. Midazolam and pacebo were administered as bolus (0.2 mg/kg) followed by continuous infusion (0.06 mg/kg/h). Both groups received morphine infusion (10 µg/kg/h). Sedation score was noted at 6 hourly intervals for 48 hours. Hemodynamic variables, ventilatory variables, complications and side effects of treatments were also recorded. Results: Thirty-three neonates were enrolled (17 in midazolam, 16 in placebo group). The groups were comparable for birth weights and gestation. The midazolam group had significantly better sedation from 18-24 hours after enrollment compared to placebo group. At 48 h there were no significant differences in proportion of infants with adequate sedation between midazolam and placebo group. The two groups were comparable with respect to heart rate, perfusion, ventilatory indices and blood gas parameters. None of the infants were noted to have hypotension on loading with midazolam or placebo. Seizures were noted in 2 neonates in placebo group 24 hours after enrollment (insignificant statistically). Conclusion: Sedation provided by continuous infusion of midazolam and morphine appears to be comparable to morphine alone in newborn babies on mechanical ventilation, with no significant adverse effects. The course of mechanical ventilation is not influenced by use of midazolam.

Key words: Mechanical ventilation, Midazolam, Newborn, Sedation.

Intensive care can be a painful and stressful experience for newborns that induce behavioral, hormonal and metabolic change. Continuous infusion of sedatives has been extended to ventilated babies to control agitation and help them adapt to ventilation. Currently opiod-benzodizepine combination is used to sedate newborns, but its use is limited by prolonged variable elimination of these drugs, leading to prolonged sedation. Mida-zolam having a short half-life produces a rapid and consistent effect with rapid elimination at discontinuation of therapy, thus facilitating early extubation. However, experience with midazolam as a sedative during mechanical ventilation in newborns is limited(1,2). The present study was designed to test the hypothesis that midazolam is an effective sedative in mechanically ventilated low birth weight newborns.

Neonates delivered at a tertiary teaching hospital were enrolled into a randomized placebo controlled study. All neonates with birth weight <2000 g who required mechanical ventilation during first seven days of life were included. Neonates were excluded if they had encephalopathy, birth asphyxia, major 

encephalopathy, birth asphyxia, major malformation, and were born to mothers who had received benzodiazepine prior to delivery. The primary outcome variable was adequacy of sedation (as judged by sedation score) during 48 hours after initiating midazolam infusion.

Sample size: To detect a 50% difference (control arm 40%, trial arm 90%) in adequacy of sedation between midazolam and placebo group with 80% power and a probability of 5%, it was estimated that 15 subjects would need to be enrolled into each limb of the study.

Eligible neonates were randomized to midazolam or placebo group by computer generated random numbers placed in opaque envelops. Midazolam (FulsedTM, Ranbaxy Laboratories Ltd.) and placebo (0.9% saline) were supplied in similar color and volume vials pre-coded as drug A and drug B. The code was broken only at the end of study after the analysis. Midazolam (1 mg/ml solution) therapy was initiated by administering a bolus of 0.2 ml/kg followed by continuous infusion of 0.06 ml/kg/h (0.06 mg/kg/h)(3). A similar volume of placebo was administered to the control group. Infants in both groups received morphine by continuous infusion at a dose of 10 µg/kg/h.

Behavior Scale(4): The adequacy of sedation was assessed on a 5 item behavior scale, which scores 0 or 1 for each item: Facial expression: calm and relaxed (0) or pronounced (1); Sucking: absent (0) or strong and rhythmic with pacifying effect (l); Continuous motor activity: normal (0) or agitated (1); Excitability, responsiveness to stimulation; normal (0) or tremulous clonic movements (1); Excessive flexion (fingers and toes): absent (0) or present and constant (1). The sedation score was determined before drug therapy and 6 hourly thereafter till next 48 hours by one of the investigators blinded to the infant’s group allocation. Infants with a total score of 0 or 1 were considered to have adequate sedation.

Data Collection: Heart rate, blood pressure and ventilatory requirements (FiO2, PIP, PEEP, bpm) were recorded continuously. Arterial blood gas was done regularly as indicated by clinical condition. Mean daily values calcula-ted for each patient were used for analysis. The patients were also monitored for ventilatory complications, i.e., air leak and intraventricular hemorrhage. Side effects of treatment, which were monitored, included epileptiform move--ments, hemodynamic instability (hypotension, tachycardia, oliguria) which would require volume expansion and/or vasoactive drugs.

Statistical Analysis: The continuous variables were analyzed using student-t test and Mann Whitney-U test (for non-parametric variables). Proportions were analyzed by Chi-square or Fischer exact test. A probability of 5% was considered significant. Patients who had to be withdrawn from the study prior to its completion both due to death or side effects of drugs were retained in the original group and analyzed on an intention to treat basis.

The institutional ethics committee approved the study and informed consent was obtained from parents prior to infant’s inclusion in the study.

There were 33 neonates eligible for the study; 17 in the midazolam and 16 in the placebo group. Table I provides the character-istics of neonates enrolled into the study. The groups were comparable with respect to birth weight, gestation, sex, ages at enrolment, indications for ventilation, ventilation charac-teristics and oxygenation indices

Three patients each in midazolam and placebo group did not complete 24 hours of follow up after enrollment. Additionally, 4 patients in midazolam group and 4 in placebo 

group did not complete 48 hours of follow up. The losses in the two groups were similar and differences were insignificant. Median age (range) at withdrawal from study was 27 (5-36) and 26.5 (12-44) hours (p >0.05) respect-ively, in midazolam and placebo groups. Reasons for withdrawal were deaths (6 in midazolam and 7 in placebo group) and cessation of mechanical ventilation (1 in midazolam group). The loss to follow-up group had comparable baseline characteristics (i.e., birthweight, gestation, sex, oxygen index and AaDO2 gradient) to those who completed the study.

The details of adequacy of sedation in the two groups are provided in Table II The midazolam group had significantly better sedation from 18 hours of initiation of infusion. This difference persisted as long as infusion was continued. The numbers of babies with adequate sedation were also significantly higher in the midazolam group compared to those receiving placebo at 24, 30 and 36 hours.

The groups were also comparable for hemodynamic variables during the study period. The midazolam and placebo groups had comparable heart rates (per min) at baseline (143.7 ± 14.3 vs 151.2 ± 20.9), post bolus (141.7 ± 16.1 vs 151.9 ± 23.9) and during the next 48 hours. The groups were also comparable for their perfusion status and urine output during the 48 h of observation. After bolus of midazolam or placebo none developed hypotension. Epileptiform move-ments were observed in 2 neonates in the placebo group 24 h after enrolment into the trial. These infants were neurologically normal and investigations failed to detect a possible cause for these seizures. Measures of oxygenation, ventilatory parameters and acid base status remained similar between the two groups throughout the study. In addition, the treatment did not influence the duration of ventilation (Table I). Pneumothorax or intra-cranial haemorrhage were not observed in any of the enrolled neonates.

Table I__Characteristics of Patients Enrolled into the Study

None of the differences between the groups were statistically significant (p >0.05).

(AaDO2 : Alveolar-arterial oxygen gradient).

Table II__Adequacy of Sedation in Midazolam and Placebo Groups

* p <0.05

Intensive care is a painful experience for the newborn. Analgesia and sedation is recommended routinely in neonatal intensive care, especially in ventilated newborns. Anxiety and pain is difficult to distinguish and their precise measurement in the newborn is even more difficult(5). In the present study, behavioral assessment was used to demonstrate the sedative effect of midazolam in ventilated newborns. Opiod benzodiazepine combination is most commonly used to sedate newborns, but its use is limited by prolonged variable elimination of these drugs, leading to prolonged sedation. Midazolam produces a rapid and consistent effect because of its short half-life. Midazolam has been used to sedate newborns undergoing anesthesia, after surgery and during short procedures like CT scan or MR imaging.

Jacqz et al.(1) conducted the first placebo controlled trial of midazolam in mechanically ventilated neonates wherein they randomized 24 to the midazolam arm and 20 to the placebo arm. Sedation was assessed on a behavior scale similar to that used in the present study. A significantly higher proportion of babies were assessed to be in a state of adequate sedation in the midazolam group than the placebo group throughout the period of the study of 5 days. Hemodynamic instability requiring vasoactive drugs or albumin were observed in 30% of midazolam and 33.3% of placebo group. Measures of oxygenation and ventilation remained similar between the two groups. Jacqz et al. did not observe any correlation between sedation scores and the plasma levels of midazolam. In another recent multicentric randomized placebo controlled trial, Anand et al(2) randomized 67 preterm neonates (24-32 weeks, <72 h of age) on mechanical ventilation to either morphine, midazolam or placebo infusion groups. It was observed that pain during endotracheal suction was 

significantly less in morphine and midazolbm groups compared to the placebo group. The results of the present study are similar to the earlier observations that midazolam produces effective sedation when used together with morphine in ventilated neonates but only marginally better than morphine alone.

Hypotension has been reported as a serious adverse effect of midazolam particularly in very premature neonates. Jacqz et al.(3), observed 4 cases of hypotension following bolus of midazolam , in 3 of whom Fentanyl was coadministered. Burtin et al.(6), reported hypotension in 6 babies after bolus of midazolam or during cotreatment with Fentanyl. In the present study, none of the patients developed hypotension after midazolam administration, which were similar to the observations of Jacqz et al.(1). It appears likely that risk of hypotension with midazolam is more likely when Fentanyl is co-administered for sedation. Oliguria and epileptiform movements are the other reported side effects of midazolam. Oliguria and epileptiform movements were not observed as a side effect of midazolam in the present study. Jacqz et al.(1) also did not observe epilepti-form movements or oliguria in their study. The present study did not demonstrate any beneficial effect of midazolam on the course of disease (as measured by ventilatory indices). These observations are similar to earlier studies(1).

In children, plasma concentrations of midazolam required to obtain adequate sedation depend on the age of the child and the indication for sedation, but usually range from 100-400 µg/L(7). In the present study the dose used was based on the pharmacokinetic data from previous studies on critically ill neonates(3). Although plasma concentration of midazolam could not be measured in the present study, the doses used achieved clinically adequate sedation in newborn infants.

The results of the present study suggest that midazolam when used in conjunction with morphine, is a safe drug, which can be used for sedation in mechanically ventilated low birth weight babies. However, the benefit of midazolam over morphine alone appears to be minimal.

Contributors: SR designed the study, analyzed the data and critically reviewed the manuscript. VA monitored the patients and drafted the manuscript.

Funding: Ranbaxy Laboratories Ltd. provided the drugs and placebo for the study.

Competing interests: None stated.

1. Jacqz AE, Daoud P, Burtin P, Desplenques L, Beufil F. Placebo controlled trial of midazolam sedation in mechanically ventilated newborn babies. Lancet 1994; 344: 646-650.

2. Anand KJ, Barton BA, Mclntosh N, Lagercrantz H, Pelausa E, Young TE, et al. Analgesia and sedation in preterm neonates who require ventilatory support: Results from the NOPAIN trial. Neonatal outcome and prolonged analgesia in neonates. Arch Pediatr Adolesc Med 1999; 153: 331-338.

3. Jaczq AE, Daoud P, Burtin P, Maherzi S, Beaufils F. Pharmacokinetics of midazolam during continuous infusion in critically ill neonates. Eur J Clin Pharmol 1992; 42: 329-332.

4. Barrier G. Attia J, Mayer MN, Amiel TCl, Shnider SM. Measurement of postoperative pain and narcotic administration in infants using a new clinical scoring system. Inten Care Med 1989; 15: 537-539.

5. Anand KJS, Hickey PR. Pain and its effects in the human neonate and fetus. N Eng J Med 1987; 317: 1321-1329.

6. Burtin P, Daoud P, Jacqz E, Mussat P, Moriette G. Hypotension with midazolam and fenta- nyl in the newborn. Lancet 1991; 337: 1545-1546.

7. Hartwig S, Roth B, Theisohn M. Clinical experience with continuous intravenous sedation using midazolam and fentayl in pediatric intensive care unit. Eur J Pediatr 1991; 150: 784-788.

.