Observations on role of benzodiazepines in babies admitted to SCBU.

Arrowe Park Hospital, The Wirral, Merseyside, UK, 1996.

Dr James Robertson, Paediatrician, Special Care Baby Unit, Arrowe Park Hospital.

BACKGROUND: "The Wirral had a large drug problem with approximately 5000 addicts at any one time out of a total population of 350,000. This gives us a rate of about 15,000/1,000,000.

over the last few years htere has been an increasing tendency for polyuse. Prior to Mid-late 1993, the main drugs of misuse were heroin and methadone, but after this time there has been an increasing tendency to use benzodiazepines instead of heroin. This has led to two changes:

  1. Increased fertility due to decrease in heroin use.
  2. A change in the withdrawal problem.

It is already known that heroin causes a withdrawal that usually starts within 2 - 3 days of birth and is usually over within 2 weeks.

Methadone causes a slightly longer withdrawal that starts within the first few days of life, but then would last for up to 4 weeks.

In the years 1994 - 1995, we noted that a proportion of our babies [in SCBU] were experiencing a withdrawal that was lasting for up to 50 to 70 days and this withdrawal did not start for 10 days.

We suspected that this was caused by the benzodiazepines as pharmacologically they are not metabolised until 10 days of life in neonates (as we have no pure benzodiazepine users, it is not possible to prove this, but there have been a few case reports of prolonged withdrawal in mothers known to use benzodiazepines on top of methadone in the literature). This has led us to create a hypothetical graph to explain the different withdrawals to mothers.

It is also known that not all drug babies will withdraw and it is not possible to predict which will. It is however, accepted that the more of a drug that a mother is taking, the more chance that baby has of withdrawing and we felt that the more drugs different drugs a mother was on, the more likely the baby was to withdraw also.

METHOD: In 1994 and 1995, the policy at the Drug Unit was to try to decrease both the methadone and the benzodiazepines on alternate weeks, such that by delivery, very few mums were off either drug. However, with the change in pattern of withdrawal and the increased frequency, this obstetric practice was altered at the end of 1995, beginning of 1996, when we explained to mothers that they could stay on the same dose of methadone (and occasionally slightly increase it), if they stopped their benzodiazepines. We created an order of stopping drugs as follows:

  1. Alcohol.
  2. Cocaine.
  3. Benzodiazepines.
  4. Heroin.
  5. Methadone.

Alcohol was top of the list because it is known to cause permanent damage and the American literature seems to suggest that cocaine can cause potential damage - we felt after these drugs, benzodiazepines were causing prolonged withdrawal.

RESULTS: In 1994, we knew of a total of 25 mums who were abusing drugs while pregnant, with a mixture of heroin, methadone and benzodiazepines. A total of 17 babies were admitted to SCBU for a total of 487 days. Three of these babies were in SCBU for more than 60 days. These longer withdrawals were very difficult to treat with our usual management of chloral hydrate and morphine/chlorpromazine.

In 1995, we knew of 54 mums who were abusing a similar mixture of drugs as above. We admitted 22 babies to SCBU for a total 504 days. Again, there were several babies with prolonged withdrawal.

In 1996, after our change in obstetric management (as discussed above), there were--- pregnancies, of which only 13 were admitted to SCBU for aq total of 162 days (excluding one baby with prolonged withdrawal from amphetamines and diabetes insipidus).

1994 1995 1996
NO OF BABIES 25 54 ---
No admitted 17/25 22/54 13/
Average LOS(days) 28 23 12

The P value, when comparing the admission rates of 1994 and 1995 combined with 1996 using a Yates corrected Chi square test is <0.001. To further confirm that we had achieved a change in the number of mother who had stopped taking benzodiazepines, we reviewed urine tests from 1994 - 1997. Unfortunately as this was done retrospectively, we could not review all urines. however we only accepted results which showed that thte mothers had come off benzodiazepines and we were not happy with a verbal report in this regard. This showed that in 1994/1995, out of the 21 sets of case notes reviewed, only 4 were off benzodiazepines at the time of delivery. However, in 1996-1997, of 37 sets of case notes reviewed, 26 were not taking benzodiazepines, 21 were weaned off and 11 stayed on them. It should be noted that the failure group included 2 that delivered prematurely, before we were able to get them off benzodiazepines and several others came in 'as-booked' very late, non-attenders, or went from regular or infrequent users.

1994/1995 1996/1997

This again is analysed by a Yated corrected Chi square with a p value of 0.002 and if we exclude the two premature babies, the value is 0.001.

DISCUSSION: On the Wirral a problem was perceived with the introduction of benzodiazepines as a drug of abuse which created a prolonged length of stay[in SCBU] and also an increased frequency of withdrawal, such, that in 1994-1995 on average 50% of babies were being admitted to our Special Care Baby Unit. Towards the end of 1995-1996 we changed our practice so that we prioritised the stopping of benzodiazepines during pregnancy, which was backed up by frequent urine specimens. We managed to successfully get about 80% of our mothers off benzodiazepines and this decreased the admission rate to SCBU from 50% down to 13%. These results are evidence that the neonatal management of drug withdrawal is best started in pregnancy and requires monitoring of maternal drug use to prioritise which drugs should be stopped in poly drug use and which order, as suggested above.

By changing our practice during pregnancy, not only have we decreased our admission rate without changing any other practices, but also decreased the length of the more severe withdrawal.

Prioritising which drug to decrease first in poly-drug use will have a much greater effect than decreasing them all.

Reproduced with kind permission
(C) Dr James Robertson, Arrowe Park Hospital, Merseyside.