Sedative
PREGNANCY RECOMMENDATION: Limited Human Data—Animal Data Suggest Low Risk
BREASTFEEDING RECOMMENDATION: Compatible*
*Potential Toxicity If Combined with Other CNS Depressants
PREGNANCY SUMMARY
Midazolam is a short-acting benzodiazepine used for anesthetic induction. No reports have been located that describe the use of midazolam in humans during the 1st or 2nd trimesters. Use immediately near birth has resulted in adverse neonatal neurobehavior.
FETAL RISK SUMMARY
Reproduction studies in rats and rabbits at 10 and 5 times the human dose, respectively, found no evidence of teratogenicity in either species or impairment of fertility in rats (1).
Midazolam crosses the human placenta, but this transfer, at least after oral and IM use, appears to be slower than that experienced with other benzodiazepines, such as diazepam, oxazepam, or lorazepam (2). In 13 patients given 15 mg of midazolam orally a mean of 11.4 hours (range 10.5–12.4 hours) before cesarean section, only 1 had measurable levels of the drug at the time of surgery. Maternal venous level was 12 ng/mL and cord venous level was 7 ng/mL. No patient had detectable levels of the drug in the amniotic fluid. A second group of patients (N = 11) was administered 15 mg of midazolam orally a mean of 34.3 minutes (15–60 minutes) before cesarean section. The mean serum concentrations in maternal venous, umbilical venous, and umbilical arterial blood were 12.7, 8.4, and 5.7 ng/mL, respectively. The cord venous:maternal venous and cord arterial:maternal venous ratios were 0.74 and 0.45, respectively. Six patients in a third group were administered midazolam, 0.05 mg/kg IM, 18–45 minutes (mean 30.5 minutes) before cesarean section. Drug levels from the same sampling sites and ratios obtained in the second group were measured in this group, with results of 40.0, 21.7, and 12.8 ng/mL, respectively, and 0.56 and 0.32, respectively. None of the 1- and 5-minute Apgar scores of the 30 infants was less than 7, and no adverse effects attributable to midazolam were observed in the newborns (2).
The placental transfer of midazolam and its metabolite, α-hydroxymidazolam, was described in a 1989 reference (3). (See reference 7 for the clinical and physiologic condition of the newborns.) Twenty women were given 0.03 mg/kg of midazolam IV for anesthesia induction before cesarean section. The mean concentrations of midazolam in the mothers’ serum and cord blood were 339 and 318 ng/mL (ratio 0.66), respectively. Similar measurements of the metabolite produced values of 22 and 5 (ratio 0.28), respectively. The elimination half-life of midazolam in the newborn infants was 6.3 hours (3).
Plasma levels of midazolam were measured at frequent intervals up to 2 hours after a 5-mg IV dose administered to two groups of pregnant patients in a study published in 1985 (4). Levels were significantly higher in 12 patients in early labor than in 8 women undergoing elective cesarean section. No reason was given for the significant difference. Data on the exposed newborns were not given.
Midazolam, 0.2 mg/kg (N = 26), or thiopental, 3.5 mg/kg (N = 26), was combined with succinylcholine for rapid-sequence IV induction before cesarean section in a study examining the effects of these agents on the newborn (5). Five of the newborns exposed to midazolam required tracheal intubation compared with one in the thiopental group, a significant difference (p <0.05). The authors concluded that thiopental was superior to midazolam for this procedure.
A 1989 study compared the effects of midazolam 0.3 mg/kg (N = 20) with thiopental 4 mg/kg (N = 20) in mothers undergoing induction of anesthesia before cesarean section (6). The only difference found between the groups was a significantly higher (mean 7.9 mmHg) diastolic blood pressure in the midazolam group during, but not after, induction. Characteristics of the newborns from the midazolam and thiopental groups (both N = 19) were compared in the second part of this study (7). Oxygen via face mask was required in five midazolam-exposed newborns compared with three in the thiopental group. Respiratory depression on day 1 and hypoglycemia and jaundice on day 3 were also observed in the midazolam group. Moreover, three statistically significant (p <0.05) neurobehavioral adverse effects, from the 19 tested, consisting of body temperature, general body tone, and arm recoil, were noted after midazolam exposure during the first 2 hours after birth. Other parameters that were inferior in the midazolam group, but that did not reach statistical significance, were palmar grasp, resistance against pull, and startle reaction (7). Although these differences do not meet all of the criteria for “floppy infant syndrome” (see Diazepam), they do indicate that the use of midazolam before cesarean section has a depressant effect on the newborn that is greater than that observed with thiopental.
BREASTFEEDING SUMMARY
Midazolam is excreted into breast milk. In a study published in 1990, 12 women in the immediate postpartum period took 15 mg midazolam orally at night for 5 nights (8). No measurable concentrations of midazolam or the metabolite, hydroxy-midazolam, were detected (<10 nmol/L) in milk samples collected a mean 7 hours (range 6–8 hours) after drug intake during the 5-day period. One mother, however, who was accidentally given a second dose (total dose 30 mg) did have a milk concentration of 30 nmol/L (milk:plasma ratio 0.20) at 7 hours. The authors estimated that the exposure of the infant would be nil in early breast milk if nursing was held for 4 hours after a 15-mg dose. Two women also were studied at 2–3 months postpartum. In six paired milk and serum samples collected up to 6 hours after a 15-mg dose, the mean milk:plasma ratio was 0.15. Based on an average milk concentration of 10 nmol/L, a nursing infant would ingest an estimated 0.33 mcg of midazolam and 0.34 mcg of metabolite per 100 mL of milk if nursed within 4–6 hours of the maternal dose (8).
The effects of exposure to benzodiazepines during breastfeeding were reported in a 2012 study (9). In a 15-month period, spanning 2010–2011, 296 women called the Motherisk Program in Toronto, Ontario, seeking advice on the use of these drugs during lactation and 124 consented to the study. The most commonly used benzodiazepines were lorazepam (52%), clonazepam (18%), and midazolam (15%). Neonatal sedation was reported in only two infants. Midazolam was not used by either mother. There was no significant difference between the characteristics of these 2 and the 122 that reported no sedation in terms of maternal age, gestational age at birth, daily amount of time nursing, amount of time infant slept each day, and the benzodiazepine dose (mg/kg/day). The only difference was in the number of CNS depressants that the mothers were taking: 3.5 vs. 1.7 (p = 0.0056). The investigators concluded that their results supported the recommendation that the use of benzodiazepines was not a reason to avoid breastfeeding (9).
The American Academy of Pediatrics classifies midazolam as a drug for which the effect on a nursing infant is unknown but may be of concern if exposure is prolonged (10).
References
1.Product information. Versed. Roche Laboratories, 1997.
2.Kanto J, Sjovall S, Erkkola R, Himberg J-J, Kangas L. Placental transfer and maternal midazolam kinetics. Clin Pharmacol Ther 1983;33:786–91.
3.Bach V, Carl P, Ravlo O, Crawford ME, Jensen AG, Mikkelsen BO, Crevoisier C, Heizmann P, Fattinger K. A randomized comparison between midazolam and thiopental for elective cesarean section anesthesia. III. Placental transfer and elimination in neonates. Anesth Analg 1989;68:238–42.
4.Wilson CM, Dundee JW, Moore J, Collier PS, Mathews HLM, Thompson EM. A comparison of plasma midazolam levels in non-pregnant and pregnant women at parturition. Br J Clin Pharmacol 1985;20:256P–7P.
5.Bland BAR, Lawes EG, Duncan PW, Warnell I, Downing JW. Comparison of midazolam and thiopental for rapid sequence anesthetic induction for elective cesarean section. Anesth Analg 1987;66:1165–8.
6.Crawford ME, Carl P, Bach V, Ravlo O, Mikkelsen BO, Werner M. A randomized comparison between midazolam and thiopental for elective cesarean section anesthesia. I. Mothers. Anesth Analg 1989;68:229–33.
7.Ravlo O, Carl P, Crawford ME, Bach V, Mikkelsen BO, Nielsen HK. A randomized comparison between midazolam and thiopental for elective cesarean section anesthesia: II. Neonates. Anesth Analg 1989;68:234–7.
8.Matheson I, Lunde PKM, Bredesen JE. Midazolam and nitrazepam in the maternity ward: milk concentrations and clinical effects. Br J Clin Pharmac 1990;30:787–93.
9.Kelly LE, Poon S, Madadi P, Koren G. Neonatal benzodiazepines exposure during breastfeeding. J Pediatr 2012;161:448–51.
10.Committee on Drugs, American Academy of Pediatrics. The transfer of drugs and other chemicals into human milk. Pediatrics 2001;108:776–89.