Jones v Taunton And Somerset NHS Foundation Trust

In Bolam v Friern Hospital Management Committee (Footnote: ³)McNair J set out the classic test as follows:

In Maynard v West Midlands RHA (Footnote: ⁴)Lord Scarman said:

In Bolitho v City and Hackney Health Authority (Footnote: ⁵): Lord Browne-Wilkinson explained and refined the Bolam test in this way:

In C v North Cumbria University Hospitals NHS Trust (Footnote: ⁶)Green J, as he then was, gave a helpful analysis of the case law on breach of duty. He said^:

It was agreed that for the purposes of this hearing the court could read and take into account Mrs Harris’ witness statement. There are perhaps only four paragraphs which are of relevance at this stage. I will therefore produce them in full as follows:

I have two other children; Judy ‘Jude’ Harris aged 37, and Kimberley ‘Kim’ Harris aged 27. Neither of my other children suffer from celebral palsy or any other serious illness. When I gave birth to Judy I had an episiotomy to help with the birth. When I gave birth to Kimberley I had an episiotomy and a suction cup was also used.

The first two trimesters of my pregnancy were unremarkable. During my third trimester I experienced a number of ‘false alarms’ where I would believe I was going into labour. On each occasion I would attend Musgrove Park Hospital, Taunton. The false alarms occurred over a period of several weeks.

Following the last of my false alarms I was treated by a nurse who told me I would have to go the full term (emphasis added). The nurse provided me with an injection which I was told would stop the onset of false alarms. I did not recall the name of the drug, however I have since been advised that it was called Nifedipine. I cannot remember the exact date of the injection, though I do recall it was several weeks before I eventually gave birth to Luc. It was during the third trimester and I now understand that I was given two doses.

I remember after I was given the injection I was not on the planet, I am able to recall that the bed was tipped up to a near vertical position I understand that this was to restore my blood pressure to an appropriate level.

I did not experience any further false alarms following the injections…”

I now propose to set out some of the key medical records some of which were attested to by Doctor Bett in her evidence.

The 25^th November 1995 notes refer to a protocol in connection with Nifedipine. No 1995 protocol has been located. There is this evidence:

In the experts’ joint statement, it is agreed that the 1995 protocol was likely to have been the same or similar to the 1997 protocol which has been disclosed. The Defendant suggests that caution should be exercised before deciding that the precise terms of the 1997 document applied to events in 1995. There is no evidence that it was identical. Indeed, as will appear from the evidence, it seems that it was not identical and may well have differed materially from the 1997.

Final submissions were originally listed to be heard commencing at 12 noon on 8^th May 2019. At that point Mr Moon QC asked for 15 minutes because of a matter which had very recently come to his attention. This was granted. The court sat again at about 12:20 p.m. It was shown three emails passing between Anne Persey and Robert Fox. Anne Persey was writing in her capacity as legal services manager of the Defendant. Mr Fox was, in 1995, a consultant obstetrician and gynaecologist at Musgrove Park Hospital.

After time had been given for consideration, the Claimant indicated that they may wish to apply to strike out the Defence for abuse of process or to adjourn the trial. This application was then made and listed to be heard on 22^nd May 2019. Evidence and skeletons were exchanged and further disclosure given. In the event, the application was withdrawn on Monday 20^th May 2019 and final submissions took place on 22^nd May.

The position after hearing the evidence is that it appears that in 1995 there was some document akin to a protocol which recommended Nifedipine. On the balance of probabilities Doctor Bett followed this ‘protocol’ in prescribing Nifedipine. This she described in her evidence summarised below. It was clearly not the same as the 1997 protocol.

There was a suggestion that I might draw an adverse inference due to the failure to call Mr Bidgood or Mr Fox about the 1995 Protocol. However, on analysis this appeared to be more a concern that I might draw an inference favourable to the Defendant based on whatever 1995 document there was. In the circumstances this does not arise.

The 1997 protocol was referred to in evidence. Its central contents are:

“PRETERM LABOUR

…

Introduction

…

The diagnosis of preterm labour is difficult. Early symptoms may be very subtle and the cervix may dilate with minimal contractions if infection is present.

Preterm labour has several causes. The management differs greatly according to the cause and so it is imperative to consider carefully why a woman has gone into labour.

PRINCIPALS (sic) OF MANAGEMENT

Encourage women with any symptoms suggestive of preterm labour to present early.

Admit all women presenting with symptoms to delivery suite for assessment.

Establish or refute diagnosis of labour (this may take hours/days).

…

Consider need for tocolysis.

…

STANDARD CARE

On Admission to Delivery Suite

In addition to any standard management of labour:

1.Review of symptoms	Contractions, mucus show, SRoM & vaginal discharge Abdominal pain & APH. Fetal movements.
2.Institute monitoring	Pulse, BP, temperature. CTG
3.Physical examination	…. Abdominal palpation for tenderness …. Cervix for effacement/dilatation/presentation ….
6. Insert venflon	If APH or suspect occult abruption.
	If instituting tocolytic therapy.
	….

Consider need for corticosteroids to enhance fetal lung maturity.

Doctor Bett was on an obstetric rotation from August 1995 to February 1996. She was therefore half way through this rotation.

Doctor Bett said that she first saw Mrs Harris on 2^nd November 1995. The medical notes for the patient, including those relating to Mrs Harris’ previous pregnancies, would have been available to Doctor Bett at some point. The usual position was that Doctor Bett would look back to an earlier summary page written by a doctor when Mrs Harris was first clerked in during the 1995 pregnancy. If there was no such summary somebody would have to go back through the previous notes.

Doctor Bett was taken in cross-examination through entries relating to a previous pregnancy. These entries date from 7^th September 1990 to 8^th January 1991. In summary they show that Mrs Harris was seen as an outpatient on 7^th December 1990, where it is recorded that she had social difficulties and a poorly supportive husband. She was miserable and had abdominal discomfort, though there were good fetal movements. She was seen again on 27^th December 1990 with mild contractions approximately every five minutes. Some two hours later the contractions had completely settled with no signs of labour.

On 3^rd January 1991 Mrs Harris was re-admitted. At 6 a.m. she was recorded as ‘still contracting irregularly.’ At 7 a.m. the notes say ‘painful contractions every 1:3-5. Later she was contracting once every five minutes with moderate to strong contractions. At 18:00 hours she was seen by the consultant and it was decided that she was not in labour and sent home. On 8^th January 1991 there is a record of the os uteri being 3cm dilated at 22:00 hours. She was transferred from the labour ward with a possibility that she had been labouring since early a.m. Contractions had almost disappeared by the evening. The next morning she had tightening once every five minutes, the CTG appeared reactive. At 12:45 p.m. Mrs Harris had some tightenings. CTG monitoring suggested contracting mild to moderate. The notes record a slightly blood- stained show at midnight. The following morning she continued to complain of contractions once every four minutes. These were mild on palpation but very uncomfortable for Mrs Harris. At 11:20 on 9^th January 1991 the contractions were becoming stronger in a regular pattern, varying in strength. The CTG was reactive. In the afternoon the notes state that Mrs Harris was getting irregular tightenings. It was determined that she was not in labour. Her uterus was soft. She went home.

Doctor Bett was taken through the medical notes for this date. She accepted that the presentation was similar to the 1990/1991 presentation described above. She said, however, that each time somebody comes in as a patient the history has to be taken and they have to be assessed at that period of time. If a patient complains of contractions, then an examination has to be done so as to assess the situation. On 2^nd November 1995 Doctor Bett concluded that Mrs Harris was not in labour despite the history of contractions at quite a high rate.

Again, in relation to 9^th November 1995 midwife and doctor’s note, Doctor Bett accepted that there was a similar pattern to that in 1990/1991 and on 2^nd November 1995. This was a pattern of complaints of abdominal pain and contractions which turned out not to be labour.

Doctor Bett prescribed betamethasone which was given at 20:40. It appears that she saw Mrs Harris shortly before that time. She would have been able to see the notes from 9^th November 1995 and the midwifery note of the 25^th November 1995 admission at 18:40. She would have either looked at the note or discussed the matter with the midwife. She said that Mrs Harris was complaining of pain similar to the 2^nd November 1995 and 9^th November 1995. She said it is hard to assess, but if a patient says the pain is worsening that has to be considered and a examination has to be done. The diagram of the location of the pain which she made at about 20:40 hours is effectively the same as that she made on 2^nd November 1995. It would have been difficult to confirm if Mrs Harris had in fact had a show but that would have to be borne in mind. The possible show the day before would be relevant, because a show is a mucous plug which comes away early on in labour.

Apart from the possible show, another difference from 2^nd November 1995 and the examination at about 20:40 on 25^th November 1995 was that on the first occasion there was no radiation of the pain. On the second occasion pain was radiating from both sides to the pubic area. Doctor Bett thought this was consistent with a contraction.

Doctor Bett’s examination at about 20:40 on 25^th November 1995 included her palpating Mrs Harris’ abdomen. It was tender+ - i.e. it was tender but not excruciatingly so. The palpation of the fetal head showed that the head had partially descended but not engaged. She could not feel the head or membranes. 1cm dilation of the cervix represents early dilatation. Her note of multips os recorded a specific appearance. Doctor Bett did not know why she did not note it on 2^nd November 1995. The amnistix tests amniotic fluid. The negative result shows that the waters had not broken. At that point Doctor Bett’s impression was that Mrs Harris was in early labour. If Mrs Harris had not insisted on leaving, Doctor Bett would have advised her to stay in hospital, having received the steroid, so she could be monitored.

The note at 23:56 hours is a midwife note. It appears from the writing that it was a change of midwife from earlier. It was now Mrs Dinsdale. The presentation was, according to Doctor Bett, consistent with the familiar pattern from Mrs Harris. She said

Doctor Bett was asked a number of questions about her note at midnight on 25^th November 1995. She said that she asked the midwife to examine the cervix. She did not examine it herself. She did not record that she had palpated the abdomen on that occasion. She would work together with the midwife. It appears that Doctor Bett was called and saw Mrs Harris with the midwife who had seen her at 23:56.

One of the things to be looked for is progression of dilatation of the cervix. Doctor Bett accepted that she would have been best placed to decide if there had been a change in the cervix, as she had been the person who had examined Mrs Harris earlier that evening. The note at midnight records what she was told and the plan she made. She did not know what she was expecting dilatation to be. From the history she had and the examination of the midwife then, if dilatation was less than 4cm, the protocol suggested prescription of Nifedipine. Doctor Bett accepted that if there had been further dilatation of the cervix that would help confirm preterm labour. She was working to the protocol. She said they were trying to prevent the labour. With contractions and pain every two minutes it seemed as though there was preterm labour.

Doctor Bett was taken to the midwife’s examination note at 00:41. She accepted that the midwife did not report any dilatation of the cervix. When Doctor Bett wrote subsequently in her note ‘M/W: soft 1 cm dilated pp-4cm’, she accepted that the dilatation was not in the midwifery note. Doctor Bett said she would have discussed the matter with the midwife and recorded what the midwife told her after her examination. She would have known there was no progression in the dilatation. Mrs Harris was complaining of continuing pain. She was working on the basis of the increase in pain and contractions every two minutes. Looking at the midwifery note, effacement of the cervix would be relevant, as would the fact that the cervix admitted a finger.

Doctor Bett did not write down a diagnosis of preterm labour. She accepted that if she was going to make a diagnosis of preterm labour she should have written it down, along with anything relevant to it. She believed that she would have made a diagnosis of preterm labour because the protocol was to prevent such labour progressing.

On the drug chart the second entry for betamethasone appears out of chronology. This is because she would have written it up as the second dose to be taken when she prescribed the first dose at 20:40. Then there are three entries for Nifedipine. Doctor Bett would have written those in at the same time. The entries state:

“26/11/95 – if Cx dilation <4 cms Nifedipine 10mg s/l (Footnote: ¹⁵)

26/11/95 30 minutes later Nifedipine 10mg s/l

26/11/95 60 minutes later Nifedipine 10mg s/l

-

Call if BP less then 100/50 mmHg

-

If contractions cease then no further Nifedipine.”

It appears the midwife has written against the third prescription:

“hypotensive and not contracting so not given.”

Doctor Bett was asked whether it was right for her to make the decision to prescribe Nifedipine. She said that as a junior SHO she felt that she was part of a good team and well supported with midwives and registrars to guide her. Some registrars may have come themselves to see Mrs Harris. She, Doctor Bett, relayed the information to the registrar on duty and went with what he said at the time.

Doctor Bett did not know that Nifedipine was unlicensed. She prescribed it because it was in the protocol. The protocol would be kept in a folder at the nurses’ station on the ward. Doctor Bett would have checked it. She would not have asked the patient to consent to the administration of Nifedipine.

Doctor Bett would have prescribed the Nifedipine be given sub-lingually presumably because that was in the protocol.

Later Doctor Bett was called because Mrs Harris became hypotensive. Therefore, Doctor Bett instructed that the Nifedipine be stopped. She was aware hypotension was an effect of Nifedipine. The notes show the first time Mrs Harris was seen by a registrar was by a Doctor Meates who came on after 8 a.m.

Doctor Bett did not remember the protocol. She did not require venflon but, had it been in the protocol, which she was following, she would have ensured this was done. Similarly, if the protocol had said that Nifedipine should be given orally she would have ensured it would be given orally. That is why she says that the protocol must have stated that Nifedipine be given sub-lingually.

In general terms the protocol would have said that the drug was there to reduce contractions and prevent labour. If contractions ceased then the protocol would have said that the drug be stopped.

Doctor Bett made it clear that she would follow very carefully what the protocol had said. Therefore, where the 1997 protocol states: ‘if contractions reduce substantially, repeat Nifedipine (10mg orally) every 4 hours for 48 hours’ – that is not consistent with what she wrote. Therefore it would not have been in the protocol she was following. However, the regime for three doses suggested in the 1997 protocol would appear to accord with what she was following at the time.

Where the 1997 protocol suggests more than 3cms dilation as a contra-indication, although logically this may equate to less than 4cms, Doctor Bett felt that the protocol she was following would have said less than 4cms.

Finally, having prescribed the Nifedipine, Doctor Bett was expecting the midwife to manage the contractions. She interpreted the midwife note at 01:20 that Mrs Harris was complaining of the abdomen still being painful (in waves) as meaning that Mrs Harris was still getting waves of pain i.e. contractions.

In paragraphs 13 and 14 of his witness statement Mr Emovon says:

“13.

Doctor Bett has recorded that Nifedipine should be given as per protocol. More than 22 years after the events I cannot now recall what protocol this is referring to, but it must have existed otherwise it would not have been agreed. I can confirm however, it would have been my standard practice to prescribe Nifedipine or other tocolysis to a mother who is 30 weeks in preterm labour with a view to suppressing labour. This would have been in line with my training and general practice at this time.

14.

I understand it is alleged that the second dose of Nifedipine should not have been given at 01:35 on 26^th November 1995. I see from the notes that I discussed the case with Doctor Bett at midnight but it is not noted whether we also discussed what the next steps in the mother’s care should be. I would have expected Doctor Bett to have called me if the clinical picture significantly changed. Otherwise, I would have expected her to continue to treat the mother as per our protocol and this must have included the second dose of Nifedipine if this was indicated. Had Doctor Bett called me to discuss giving a second dose of Nifedipine I am likely to have agreed that this was indicated because at 01:20 she has noted ‘abdomen still being painful (in waves)’. As there was evidence that Ms Harris was still experiencing contractions, I would have recommended continuing with Nifedipine to try and suppress the labour.”

Mr Emovon said of the note which referred to Mrs Harris’ ‘abdomen still being painful (in waves)’, that ‘in waves’ conveyed to him severity of pain and suggested that it had become more frequent. He said that if a midwife had said this to him as a registrar he would accept what the midwife said, the midwife being a trained professional. Indeed, he said he would prefer a midwife’s examination to that of a freshly qualified SHO.

Nevertheless, it would be a matter for the doctor to decide if a second dose was indicated. His opinion, reading the notes, was that contractions were not improving and they were getting worse. Therefore, a second dose was indicated. Somebody i.e. a doctor or midwife would examine the patient to check for contractions. They would examine the abdomen, but also take into account the patient saying that the contractions were getting worse in terms of frequency. He would look at a CTG trace but said that this did not always show contractions. He would not necessarily have expected Doctor Betts to call him to give authorisation for the second dose. If the protocol was there and the situation was getting worse, the SHO would not have to call. He would have expected Doctor Bett to call him if the clinical picture had changed significantly. Although the blood pressure later dropped, within a minute it had picked up from 46 to 54. He did not think that episode of hypotension was significant. If Doctor Bett was comfortable following the protocol, even if the plan changed, then that was a matter for her.

Mr Emovon was asked whether he would have expected to have been told about the second dose having been given and that tocolysis was later stopped. His response was that if Mrs Harris had progressed to labour, there was nothing he could do for her. It would then be a matter for the midwives to deliver the baby unless the patient needed a caesarean.

In terms of the prescription of tocolysis, Mr Emovon had said in his statement that he did not agree that Mrs Harris was not in preterm labour at the time. She was recorded as experiencing pains every two minutes which were worsening. There was cervical effacement and the cervix was 1cm dilated. In the circumstances, there was evidence that she was in preterm labour although the CTG trace indicated no uterine activity. Having been given the information by Doctor Bett, he must have agreed that she should administer Nifedipine as per the protocol. The information he had at the time was that there was pain every two minutes worsening and the cervix was 1cm dilated. In addition (Footnote: ¹⁶) there was the information that the cervix was ‘soft’. To Mr Emovon this meant effacing of the cervix, which happens when a woman is going into labour. The pp of 4cm is the distance between the head to the ischial spines. All that information indicated strongly to him that Mrs Harris was in preterm labour. In addition Mrs Harris had come back to the hospital saying the pains were worsening. He was aware of her previous admissions. He said that as a doctor one has to make a judgment as to what is going on at the relevant point. On this occasion she had come back within a few hours of being seen earlier that day in hospital. This he regarded as significant because it was a short time.

As to his own position, Mr Emovon said he could have been anywhere in the hospital. He was covering a number of areas. He might have been in theatre. Had he gone he would have examined the cervix and palpated the abdomen for evidence of contractions. He would have looked at the CTG. The fact that it did not show contractions would not particularly matter. He would have palpated the abdomen for regular contractions. He accepted there could be other causes of pain, including urinary tract infection. He said that one of the problems is determining if there is early labour or not. A clinician should err on the side of caution and give tocolytics. It is all a matter of judgment. However, when he was informed on the night, the impression was that Mrs Harris had gone on into preterm labour.

As to the prescription of the particular tocolytic, Mr Emovon said that Ritodrine was another possibility. Nifedipine was the drug they used at the time. He went with the first line treatment set out in the protocol unless there were contra-indications. He cannot remember when Nifedipine became the first line treatment. He was a registrar and followed the protocol. During his working history the protocols depended on the hospital and he followed the protocol for the hospital in which he worked. He also relied on the protocol in respect of the administration of the tocolysis.

After 26^th November 1995 Mrs Harris delivered at term. Mr Emovon said he did not know whether that meant (a) she was in preterm labour on 25^th/26^th November 1995 and this was inhibited by the Nifedipine or (b) she was not in preterm labour on that night.

He said he did not know whether Mrs Harris was multips os or not. If the cervix is dilated then it is dilated. If she had been multips os, then he probably would have given the drug earlier because she would be more likely to deliver.

In Mr Hare’s CV he summarises his clinical experience. Prior to being a consultant, he had a special interest in genital infection and also developed an interest in colposcopy, premalignant disease and day care surgery. Between 1976 and 1983 he was a consultant in Cambridge. His CV does not mention particular interests during that period. Between 1983 and 1998 he was a consultant in a district general hospital in Huntingdon. He there maintained his special interest in colposcopy day surgery (especially pregnancy termination and sterilisation) and infection. He did not have any special interest in tocolysis. He has published nothing on tocolysis and has not carried out any research on tocolytic drugs. He managed a labour ward for 15 years and therefore had a lot of experience of tocolysis, but he accepted that Profesor Thornton knows much more about research into tocolysis than he does.

In 2000 he was a co-author of a book prepared for lawyers, not doctors. Under the heading ‘Management of Preterm Labour’ he did not mention calcium blockers such as Nifedipine. He said that in 2000 he was aware there was some movement to use calcium blockers and away from using Ritodrine. He was not aware of the use of calcium blockers, for example in Musgrove Park Hospital or Walsgrave Hospital in Coventry in the 1990s.

In paragraph 31 of his Report Mr Hare addressed the question ‘should the fact Nifedipine was not licensed for use in pregnancy have influenced this decision?’ He describes this as a major consideration and brings into evidence the 2017 data sheet for Nifedipine. He said that when he prepared his Report he was dealing with a 1995 case and the use of the 2017 data sheet was to show that as far as the manufacturers were concerned the position had not changed. He said he was aware of the NICE Guidelines which, with qualifications, recommended in 2015 the use of Nifedipine in preterm labour despite the fact it was not licensed for this use. He did not refer to the NICE Guidelines in his Report. He said he didn’t think it was necessary to do so, given that he had referred to the 2002 RCOG Guidance. He thought that was adequate. He said he could have included the NICE Guidance but did not think it necessary. It could have been regarded as relevant, like many other documents. He did not consider he was in breach of the statement of truth of an expert (particularly at paragraphs 3, 7 and 8).

Although it will become apparent that I do not accept a number of elements of Mr Hare’s evidence, I do not accept that he did not comply with his duties as an expert. He was an honest and respectable witness.

Professor Thornton’s doctorate, awarded in 1989, was for work on Oxytocin. Oxytocin is a hormone active in labour which affects contractions. He was then awarded an MRC Clinician Scientist in 1992 at Cambridge. He said that very few of these are given in obstetrics. It is an award where clinical work is combined with research.

In 1998 he was appointed Professor of Obstetrics at Warwick University. His particular interests were in early labour which included dealing with tocolysis. He led the speciality on behalf of the RCOG. His work consisted not only of how the uterus contracts, but also on labour and how to help women in childbirth. This covered many aspects of childbirth including tocolysis. He oversaw a lot of documents which came out of the RCOG. His role in respect of a number of documents was an oversight role.

As Professor of Obstetrics one of the things Professor Thornton had done was to found and become Chair of the Clinical Study Group for preterm birth. He brought together a number of obstetricians from the UK who were interested in preterm labour and birth. He was responsible for setting up the group. It led to recognition by the National

Listed as a clinical interest of Professor Thornton is ‘preterm labour’. This involves a large proportion of his research time and complements his clinical interest in the management of preterm labour. He frequently lectures on this at national and international meetings. He has written numerous guidelines and chapters for clinical texts. He set up a specialist preterm labour clinic at UHCW which received regional and national referrals. He now undertakes clinical obstetrics at Barts NHS Trust where he assists the development of a specialist preterm labour clinic.

Professor Thornton lists co-authorship of over 130 publications. In addition, he has been responsible for a number of plenary presentations. In particular:

Professor Thornton said he was invited around the world to talk about current and recommended treatment of preterm labour. He said that he hoped that his views are taken seriously in relation to preterm labour and tocolysis. There is always discussion in the area. He has spent much of his working life dealing with preterm labour.

In clinical practice Professor Thornton developed and led a specialist antenatal clinic for women at high risk of preterm labour. This attracted regional and national referrals. There were three protagonists, Professor Shennon from London, Professor Jane Norman from Edinburgh and Professor Thornton in the Midlands (Footnote: ¹⁷). The three Professors set up preterm labour clinics and worked together to provide high quality care to women with preterm labour problems.

Criticisms were made of Professor Thornton as an expert witness. While I accept his evidence for the most part, there are certain matters whish require careful consideration

In the joint statement the experts agree that preterm labour is labour before 37 weeks gestation. Threatened preterm labour would be preterm contractions with or without cervical dilation. The same would be true for the term ‘suspected preterm labour’. As for ‘early preterm labour’ this could mean labour in early pregnancy or early in the process of labour. The experts assumed the latter; a reasonable definition therefore would be labour before the signs and/or symptoms are fully established or early in the process of established labour. All these terms were in common or regular use in 1995. The term ‘preterm labour’ covers all the other above terms. In the literature and in the evidence, terms were used interchangeably, though Professor Thornton sought to keep to the term threatened preterm labour. I have generally tried to do the same; however, unless stated to the contrary the expressions should be regarded as co-terminous.

The experts also agree that in 1995 the use of tocolysis was reasonable/justified for threatened or suspected preterm labour. However, administering tocolysis involves administering a drug with side effects. It is common ground that it should not be given as a prophylactic.

Mr Hare did not disagree with the statements in two books which are very similar. One extract states (Footnote: ¹⁸):

“The term threatened preterm labour is often used to describe pregnancies complicated by episodes of clinically significant uterine activity but without cervical change.”

Mr Hare’s opinion was that if the court decides that the notes in the present case record significant uterine contractions (i.e. not Braxton Hicks contractions) then the criteria for threatened preterm labour were satisfied.

In the joint report Professor Thornton says that the administration of tocolysis was definitely justified because Mrs Harris had preterm contractions and no contraindications. Mr Hare says that the diagnostic threshold had not been reached.

It is agreed that the diagnostic threshold for (threatened) premature labour at which tocolysis should be instituted can be difficult. Mr Hare’s opinion on paper in summary is:

Mr Hare was extensively cross-examined. I shall attempt to condense his evidence in cross examination in this way:

Mr Hare accepted that the note recording that Mrs Harris had perhaps had a show was relevant in that a show can be a sign that labour is about to begin.

Doctor Bett’s examination note at about 20:40 hours of 25^th November 1995 recording that the cervix was 1cm dilated, was soft and about 1cm long was the subject of discussion:

In relation to contractions:

Dealing with Mrs Harris’ history of preterm contractions in earlier pregnancies, Mr Hare said that this was more or less neutral. He did not accept that there was a distinction because the earlier pregnancy complaints were at 35-36 weeks and not 31

In relation to CTG tracing:

Mr Hare had referred in the joint statement to seven pieces of literature. It was put to him that in those papers the CTG was not used to diagnose threatened preterm labour, but rather to determine whether the women were eligible for the trial. He said that that begged the question. Mr Moon QC took him to two papers namely:

It is of importance to note that a diagnosis may have to be made under some time pressure because the experts agree that there may be a point after which the process of labour is irreversible. Therefore, failure to act may render treatment ineffective. Professor Thornton in his report said ‘it is considered that once labour is fully established, it is unlikely that the process can be significantly delayed. For this reason, threatened preterm labour is usually diagnosed on relatively soft criteria.’ In the textbook (Footnote: ²⁷) the authors say:

“…because of the need for early management of suspected preterm labour, the diagnosis is commonly made in clinical practice before the above criteria are met, and hence the reported incidence of threatened and actual preterm labour may be open to question.”

Professor Thornton added in evidence that the risks of not treating with tocolysis are high. A baby born at 30 weeks has a 4-5% chance of dying. If a baby is born at less than 32 weeks, the risk of brain injury is high. Therefore, when thinking whether to give tocolysis, the decision is heavily weighted in favour of treatment, and that is standard practice. Such is the perceived advantage that some US studies will not do placebo controlled trials on the basis that it is unethical not to treat with tocolysis in threatened preterm labour. Professor Thornton agreed that tocolysis should not be used prophylactically. He said one should look at the symptoms and signs and investigations (if applicable) in each individual case to see if tocolytic treatment was or was not appropriate. However, it is not possible to identify in advance which women will progress from threatened preterm labour into established labour.

Although the above window of opportunity is important and requires action rather than prevarication, that does not detract from the fact that there is a diagnostic threshold which should be reached before prescribing tocolysis. The diagnosis may be difficult, and doctors acting non-negligently may mistakenly diagnose threatened preterm labour and properly prescribe tocolysis. Nevertheless, as Professor Thornton himself wrote in the 1995 Yearbook of the RCOG: “…any possible improvement in fetal outcome must be offset by the risks of exposing the mother and fetus to the hazards of treatment. This is particularly important in preterm labour since uterine activity often spontaneously abates. The judicious use of tocolysis is thus of paramount importance and these drugs must be administered in clinical practice with the same rigour that is required in research.” (Footnote: ²⁸) Part of the problem is that there was, and is, difficulty in proving that tocolysis actually improves outcome, such that it is also reasonable not to use them³¹. Careful consideration should therefore be given before prescribing tocolysis.

There is thus a difficult assessment to be made between using care before prescribing and not missing the opportunity to prescribe. In the present case, however, if Mrs Harris was, or was reasonably thought to be, having true, early labour uterine contractions at a rate of about 1 in every two minutes or thereabouts, then the administration of tocolysis was not in breach of duty.

Professor Thornton was asked about certain matters following the 2^nd dose of Nifedipine (Footnote: ²⁹).

In relation to whether Mrs Harris had been experiencing true contractions at 11.56 and thereafter until the Nifedipin was stopped:

One of the matters to be addressed is that Mrs Dinsdale, the midwife who made the notes at 23.56 and onwards on 25^th /26^th November 1995 has not provided a witness statement and has not given evidence. Therefore her notes have had to be interpreted by others. The Claimant submits that I should draw inferences adverse to the Defendant in accordance with Wisniewski v Central Manchester HA (Footnote: ³²)

In Wisniewski, the Court of Appeal said this:

“(1)

In certain circumstances a court may be entitled to draw adverse inferences from the absence or silence of a witness who might be expected to have material evidence to give on an issue in an action.

(2)

If a court is willing to draw such inferences they may go to strengthen the evidence adduced on that issue by the other party or to weaken the evidence, if any, adduced by the party who might reasonably have been expected to call the witness.

(3)

There must, however, have been some evidence, however weak, adduced by the former on the matter in question before the court is entitled to draw the desired inference: in other words, there must be a case to answer on that issue.

(4)

If the reason for the witness's absence or silence satisfies the court then no such adverse inference may be drawn. If, on the other hand, there is some credible explanation given, even if it is not wholly satisfactory, the potentially detrimental effect of his/her absence or silence may be reduced or nullified.”

Wisnieski was recently considered by the Court of Appeal in Manzi v King’s College Hospital NHS Foundation Trust (Footnote: ³³). At [30]-[31] the Court said:

“30.

There are three aspects to the claimant's submissions that demonstrate the difficulty that she has on this issue.

First, Wisniewski is not authority for the proposition that there is an obligation to draw an adverse inference where the four principles are engaged. As the first principle adequately makes plain, there is a discretion i.e. "the court is entitled [emphasis added] to draw adverse inferences". An appellate court will be hesitant to interfere with the exercise of such a discretion given that it is being exercised in the knowledge of all the nuances of evidence that are in the knowledge of the judge who receives that evidence. Second, the judge in this case did not conclude that an absent witness had to be central to the case, he merely and correctly identified that the doctor in Wisniewski was central to that claim as the person who had failed to defend his clinical judgment. By comparison, the judge decided that Dr Hooper's role and hence evidence was tangential for the reasons I have summarised…. Third, there was an explanation for absence and that was a decision on proportionality grounds taken by the defendant i.e. this was not a case where a defendant or witness deliberately prevents or avoids the admission of evidence that would undermine their case.

31.

There is also a further difficulty that the claimant must face. On 21 August 2015 Master Roberts gave case management directions. The claimant sought a direction for disclosure of information about Dr Hooper but did not seek an order that she file and serve a witness statement. They could have asked for the latter. If the claimant was of the view that Dr Hooper's evidence was as important to her case as is now asserted and that an adverse inference would be appropriate in Dr Hooper's absence, they could have asked for a direction which contained the warning that an adverse inference may be drawn if the evidence was not provided. Even without such a direction, the claimant could have made arrangements to obtain evidence from Dr Hooper themselves.”

What is the factual position in this case regarding Mrs Dinsdale? Witness statements were served by the Defendant on 19^th April 2018. A year later, and shortly before trial, the Claimant’s solicitors wrote on 17^th April 2019 referring to the clinical notes and asking for the full name of the midwife, whether she was in receipt of an NHS pension and about the efforts that the Defendant had made to contact her. The Defendant responded on 25 April 2019 saying that the midwife was called Mrs Dinsdale and she had left the Trust in 1997. They had contact details for her, she was no longer practising as a midwife, and she had not responded to the solicitors’ request that she review the records and provide a witness statement. They were aware she does work for the NHS but had no knowledge of her pension arrangements

In considering whether I should draw an adverse inference, the following factors are relevant:

In this discussion section, I will take into account the preceding evidence, refer to further evidence from Professor Thornton, and come to conclusions as to whether it

2^nd November 1995: The complaint was of abdo pains/contractions which had been irregular, but had become more regular. No contractions were detected. There was no tension/tenderness on palpation. The cervix was closed. Professor Thornton said he would be a little concerned about the possibility of threatened preterm labour, but this note gave him a different feeling from the notes on 25^th November (Footnote: ³⁶).

9^th November 1995: Mrs Harris was admitted via the GP, with ?preterm labour. Professor Thornton said that if a GP was in doubt about abdominal pain, then s/he would want to exclude preterm labour. The presenting complaint was abdominal pain. His interpretation of the pain being “sometimes as frequent as 1:5” was that it did not tell us anything about frequency and it suggested that it was not always 1:5. It is not known if Mrs Harris volunteered this information or if she was asked (e.g.) whether the pain was constant and, if not, how often it came and went. The report to the SHO was abdo pains during the evening which had settled; Mrs Harris had slept and then the pains came again in the morning. There were no palpable tightenings. Although there was a record of diarrhoea the previous day, which made Professor Thornton less concerned about possible labour, he agreed that there was no definitive diagnosis of the cause of pain.

Professor Thornton said that these two admissions would have increased his sensitivity to the possibility of preterm labour. A woman who has a history of complaining of contractions may give rise to an increased possibility of preterm labour.

25^th November 1995, 1^st admission: at 18.40 the presenting complaint to the midwife was of backache and abdominal pain since the preceding day, and of a possible show. On palpation the midwife did not find any tension/contractions. Later (probably shortly prior to 20.40):

It is of note that Doctor Bett’s note was that her impression was that Mrs Harris was in early labour. She strongly advised Mrs Harris that she should not discharge herself as Doctor Bett wanted to keep her under observation, presumably because she was worried that she might be threatening preterm labour. A steroid, Betamethasane, was prescribed. It is steroids which are given to improve lung development in the premature foetus. However, they take a day or two to work. Tocolysis stops the contractions, so giving the steroids time to work. Usually the two drugs are prescribed at the same time. Prescribing a steroid at about 20.40 would presumably be as a precaution given the impression of early labour. Professor Thornton said that the threshold for giving a steroid is lower than for prescribing tocolysis.

25^th November 1995 2^nd admission: the midwife’s note at 23.56: On this admission Mrs Harris complained of contractions occurring 1:2. The midwife (Mrs Dinsdale) palpated the abdomen and found that it tensed for short periods, quite frequently, though it did not seem as often as once every 2 minutes. Mrs Harris was very uncomfortable. The tension was felt suprapubically. What if any difference was there from the position earlier that evening and on the two previous admissions earlier in November? Professor Thornton said:

I shall pause at this stage of the narrative to make certain findings. They are these:

Doctor Bett was consulted at midnight. Apart from recording the fresh history and the fact of discussion with the Registrar, her note merely sets out the regime for prescribing Nifedipine, subject to the midwife confirming that the cervix was less than 4cm dilated. Though Doctor Bett does not note a diagnosis of preterm labour or threatened preterm labour, she said in evidence that she was trying to prevent labour and it seemed with the contractions/pain every 2 minutes that there was preterm labour. It also appears that she had diagnosed it at this point, as she decided (so long as the cervix was found to be < 4cm dilated) to prescribe Nifedipine. This was Professor Thornton’s understanding.

At midnight there was no further cervical examination by Doctor Bett. No progress in dilatation is recorded by the midwife or Doctor Bett. This appears to be because, in Doctor Bett’s opinion, the die was cast for the Nifedipine administration, assuming < 4 cm dilatation. (Footnote: ⁴⁰)

At 0041 the abdominal CTG was commenced. The midwife made some findings on vaginal examination. Nifedipine 10 mg was then administered at 0100 following Doctor Bett’s prescription.

I therefore find that Mrs Harris was probably having true, early labour uterine, not Braxton Hicks, contractions, and that there were other indicators such as the possible show (amplified as ‘pink mucousy loss’), and a soft, 1 cm long, 1 cm dilated cervix (Footnote: ⁴²). Therefore it was reasonable to prescribe tocolysis. Although Doctor Bett did not record a diagnosis of threatened preterm labour, she did in fact make this diagnosis. It was reasonable to make this diagnosis.

As mentioned, both experts agree that if preterm labour/threatened preterm labour was a proper diagnosis, then tocolysis was indicated to allow steroids to operate so as substantially to reduce the risk to the fetus. Mr Hare says that that if a tocolytic drug was indicated, that drug in 1995 should have been Ritodrine. The Defendant accepts that it would not have been a breach of duty to use Ritodrine. Mr Hare says that it was a breach of duty to prescribe Nifedipine.

The two drugs are of different types. Ritodrine is a member of a group variously described in the literature as Beta-sympathomimetics/beta-agonists of Betamimetics. Nifedipine is a calcium channel blocker. (Footnote: ⁴³)

Mr Hare accepted that it is probable that the Defendant’s 1995 protocol (Footnote: ⁴⁴) recommended Nifedipine. He accepted that if it did recommend Nifedipine then it probably discouraged the drug which he would have prescribed, namely Ritodrine. In those circumstances it would not be Doctor Bett, the SHO, who would be criticised for prescribing Nifedipine. The central complaint is that the protocol/recipe was negligent in its recommendation. He said that when Protocols began to be introduced in the 1990s, in a District General Hospital it was usually a consultant who would write the Protocol on a particular topic. The consultant would do the research before writing the Protocol. The Protocol had to contain a balanced view of the evidence at the time, especially if it was recommending the use of a new drug. Mr Hare said he did not regard the 1997 Protocol (or the 1995 Protocol assuming it similarly recommended using Nifedipine as the tocolytic drug of choice) as a balanced view, taking account of the available literature at the time (Footnote: ⁴⁵).

In summary it was Mr Hare’s opinion that in November 1995 no reasonable body of clinicians in the UK should have been using Nifedipine as tocolysis.

Professor Thornton’s opinion was different. He said that in 1995 Ritodrine was the treatment of choice, but, because of its side effects, there was a move to introduce other drugs. By that date it was acceptable to use Ritodrine or Nifedipine (and some others).

Professor Thornton was cross-examined on certain extracts in his report. In summary these were:

“Brief History of Tocolysis

………….

12.

As nifedipine increased in popularity in the 1990’s, ritodrine was essentially phased out. The Royal College Guidelines from 2002 state that if a tocolytic drug is used ritodrine no longer seemed the best choice…

Oxytocin Antagonist

13……………Thus, in the 1990’s the only licenced treatment was ritodrine which was considered to be associated with more adverse effects than nifedipine……..

General comments on case

16………..She was given nifedipine which was standard practice at the time. Although ritodrine was licenced for use as a tocolytic, it was associated with marked maternal side effects and was falling into disrepute……I consider that nifedipine should have been used in preference to ritodrine at this time given the data available. Indeed I consider that a significant proportion of obstetricians would have used nifedipine in this situation. Given the major competitor for nifedipine (atosiban) was not licenced until 2000, nifedipine was the first choice tocolytic at the time. I therefore consider that the management of Mrs Harris with nifedipine was appropriate and consistent with standard practice……

Comments on amended particulars of claim (undated)

….

(ii)

it was contrary to the 1997 Protocol to use tocolysis continually

………

The administration of nifedipine was favoured over the use of ritodrine given the marked maternal side effects of the latter. Therefore, even if the protocol had been available and it suggested that ritodrine should be used, I consider that it would have been outdated and nifedipine should have been administered as was widely undertaken at the time…..

(c)

Despite the fact that the Defendant was not part of (or conducting its own) clinical trial

……….

Atosiban was not licenced until 2000 so this was not a viable alternative. It follows that the only alternatives in 1995 were to give ritodrine (associated with marked side effects), nifedipine (which required further evaluation), or no treatment which was considered (and still considered by some) not to be ethical. Thus, nidfedipine was the only sensible option at the time despite there being no high quality evidence of efficacy or improved outcomes…

(d)

(ii) Failed, negligently and in breach of the 1997 Protocol, to start an intravenous infusion running before instituting Nifedipine treatment….

…..

In 1995 it was considered that maternal fluids were relatively contraindicated in preterm labour. This was because ritdorine was falling into disrepute and there had been a number of serious maternal side effects (including maternal deaths) reported. It was considered that this was in part due to pulmonary oedema which was exacerbated by administration of maternal fluids….”

It was put to Professor Thornton in cross-examination that (a) he was suggesting in the report that Nifedipine was the only sensible option, (b) he was exaggerating by saying that in effect Ritodrine had been supplanted by Nifedipine by the mid-1990s. His response was that Ritodrine was falling into disrepute (Footnote: ⁴⁶), but that in 1995 Ritodrine was a reasonable treatment to use, though Nifedipine had been introduced into a number of units at that time. Ritodrine was not universally in disrepute by 1995, but by the end of the decade it was. In the earlier sections of his report⁵⁰ he talked about the history and difficulties of Nifedipine, the later extracts were from sections dealing with Mrs Harris’ treatment. In the Defendant’s unit in 1995 Nifedipine was the choice. What could not happen was that different drugs would be used by different consultants and registrars in the same unit. There needed to be conformity of treatment in a unit. This was important to reduce the risk of mistakes. Nationally it was reasonable to use a number of drugs in late 1995.

Some sections of Professor Thornton’s report might be understood to be suggesting that Ritodrine should not have been used in 1995. However, that was not what he said he had meant. Also, that might reflect his personal view of the how he perceived matters, given that he is likely to have been in the avant garde. His oral evidence in summary was that both Ritodrine and Nifedipine were acceptable tocolytics to use as at 1995. That fits with a statement in paragraph 11 of his report where he says: “The timing of this trial⁵¹ indicates that there was clinical equipoise of the use of nifedipine and ritodrine at that time”. I therefore do not accept any substantial criticism of Professor Thornton in this regard.

For the sake of clarity I have set out in the Appendix to this judgment, relevant extracts from the literature adduced in evidence. It is there itemised and referenced. In the body of the judgment I will refer to a piece of literature by its item number in the Appendix.

In the 1980s, because of increasing concern about the side effects of Ritodrine, (the then [and still] only drug licensed for tocolysis), new drugs, including Nifedipine, were receiving clinical evaluation. There were also concerns about the efficacy of Ritodrine. Two United States papers (items 1 & 2) discussed these issues and the 1990 review paper (item 2) said that calcium channel blocking agents represented an apparently powerful class of tocolytic drugs, though concern over their effect on the fetus and newborn (Footnote: ⁴⁷), as well as their unproven efficacy, should limit their clinical use pending further investigation. (Footnote: ⁴⁸)

Item 3 (1991) compared Nifedipine and Ritrodine. Mr Hare said this was an interesting contribution to the debate, but did not accept that this one small trial demonstrated progress, nor that studies were increasingly supporting Nifedipine in preference to Ritodrine.

In July 1992 a paper was published in the New England Journal of Medicine by the Canadian Pre-term Labor Investigators Group. This indicated that Ritodrine had no beneficial effects on perinatal mortality but was associated with increased maternal morbidity – particularly the potential to cause pulmonary oedema in the mother. Shortly after this publication a further two fatal cases of complications related to pulmonary oedema were reported by the Committee on Safety of Medicines (CSM) (Footnote: ⁴⁹)

Mr Hare accepted that the New England Journal of Medicine is a pre-eminent Journal and that this was an important article. Mr Hare further accepted that, although Ritodrine was still in the textbooks and Nifedipine was not, there was an increasing opinion that Nifedipione needed to be investigated. His opinion was that there was an increasing feeling among specialists in tocolysis that Nifedipine mightbe, not was, a reasonable alternative. He said that the research basis was not nearly adequate.

In December 1993 an article (Item 5) was published in the British Journal of Obstetrics and Gynaecology. It did not suggest further research and strongly suggested that Nifedipine could be used as tocolysis. Its conclusion is that Nifedipine is as effective as Ritodrine in suppressing preterm labour and its use is associated with less frequent side effects. This journal is the one most commonly read by obstetricians in the UK and has what is described as a ‘high impact factor’ as a journal. Mr Hare believes that the article contains a number of flaws, but he accepted that a reasonable clinician reading

In April 1994 a further article (Item 6) was published in the journal Obstetrics and Gynaecology. This is one of the two most respected American journals, with a very high impact factor. It is a review article, relying on and bringing together the evidence from 90 references. In his report Mr Hare had described the article as poorly written and over optimistic. This included 3 criticisms: In summary: (a) the previously published trials are dealt with in simplified form; (b) the paper wrongly combined evidence from the use of Nifedipine in the treatment of pregnancy induced hypotension with that from its use in normotensive women in preterm labour, so as to provide evidence of its safety for mother and fetus, and (c) it underplayed the significance of the danger to the fetus from animal experiments, attributing this to anaesthesia, poor hydration and dose disparity. Mr Hare’s opinion was that, notwithstanding this paper, and previous papers, Nifedipine should still have been regarded as an untried drug and should not have been in use.

Nevertheless, Mr Hare accepted that the concession he made about a reasonable clinician reading this journal at the time (Footnote: ⁵⁰) was applicable. In this paper, also, there has been no subsequent literature seriously undermining its conclusions. The implication of this article is that a reasonable body of clinicians in 1995 could use Nifedipine as tocolysis.

Item 7 is a paper written in 1997 by Papatsonis et al. It was suggested to Mr Hare that Papatsonis is a leading authority in the field. Mr Hare said he had come under criticism and that the (subsequent) Cochrane review in which he was involved had come under criticism because Papatsonis was a researcher. Mr Hare had not referred to Item 7 in his report. He said he did not consider a 1997 paper relevant to the position in 1995. It was pointed out that in paragraph 34 of his report, he had referred to later opinions and advice on Nifedipine, stating: “There is still controversy concerning the use of the drug.” He said he did not refer to it there as it was a research paper, not a review paper; he had referred to the Cochrane review.

Mr Hare said that he was aware that Item 7 was the largest ever study comparing Ritodrine and Nifedipine. The study compared patients in 3 hospitals in Holland between February 1992 and February 1995. All patients in the trial were given Nifedipine. It interpreted the literature as Nifedipine having fewer side effects than Ritodrine and concurred with this.

In his 2^nd report, commenting on Item 7, to which Professor Thornton had referred, Mr Hare said that he could not accept that “obstetricians working in an English District General Hospital in 1995 should base their practice on a research project that has just been completed in two hospitals in the Netherlands and has yet to be published”. He said in evidence that when he wrote this, he thought it possible that other reputable hospitals in the UK and US were using Nifedipine, but he did not know. It was put to him that, apart from Musgrove Park and Walsgrave Hospital in Coventry, a hospital in Denver was using it. He said that if that was the case, it was relevant but did not make it correct.

In his reports Mr Hare had relied upon two textbooks for the position relating to prescribing tocolysis in the mid-1990s. These were Dewhurst (fifth edition 1995 - Item 8) and Turnbull (second edition 1995 - Item 9). Mr Hare agreed in cross-examination that textbooks can lag behind the latest research. The section in Dewhurst on Ritodrine is out of date, the last date of any literature referred to is 1978. There is no reference to any of the literature on Nifedipine which had been produced during the 1980s and early 1990s. Turnbull refers to some of the more recent literature after 1990 on Nifedipine.

In 1994 the RCOG produced Guidelines on the use of Ritodrine (Item 10). This followed the Canadian Pre-term Labor Investigators Group Study and two reported fatal cases of complications related to pulmonary oedema. These Guidelines discuss only Ritodrine (Footnote: ⁵²). There is no discussion of other groups of drugs (e.g. Nifedipine) which might be used to inhibit pre-term labour. Mr Hare said that RCOG Guidelines are very highly thought of and generally accepted as policies. He said they should be available on every ward. Professor Thornton said that this document was a warning about the use of Ritodrine, as the Introduction makes clear. It was not a review of all possible tocolytic drugs.

The Drug and Therapeutics Bulletin (1992 - Item 11) gave examples of when unlicensed use of drugs would seem justified. These included where (a) the licensed indications do not reflect current knowledge, (b) the indications listed do not include well proven uses of the drug and (c) the licensed indications are over restrictive. (Footnote: ⁵³)

Another textbook, Enkin et al., (1995 - Item 12) says there are not enough data on Nifedipine to justify its use outside the context of well-designed and carefully monitored randomised trials. Mr Hare endorsed this viewpoint, despite that fact that there is no reference in the textbook to any of the literature on Nifedipine in the book. He said it was a guide, a summary. Mr Hare was asked what a doctor who had read (i) the literature in high quality journals that Nifedipine was recommended and (ii) the summary in Enkin should do. He said the doctor should rely on evidence-based medicine.

In 1997 the RCOG published another Guideline on Beta-Agonists (Item 13). It made a reference to calcium antagonists as ‘adjuncts to beta-agonists’ and said that they had not been shown to have the desired effects and that the available data did not justify their use. The only reference for this statement was Enkin et al. (Item 12). According to Professor Thornton, what was being there discussed was Nifedipine in addition to Ritodrine, the desired effect being to attempt to reduce cardiovascular side effects. This, he said, is a completely different situation from evaluating the advantages/disadvantages of the two drugs as tocolytics.

The RCOG 2002 Clinical Guideline (Item 14) acknowledges the use of Nifedipine (Footnote: ⁵⁴). Although it questions the evidence base for substantive benefit for the baby from tocolysis, if a tocolytic agent is used Nifedipine is suggested as preferable to Ritodrine. Mr Hare said that by 2002 the use of Nifedipine was creeping in and that the responsibility for the prescription of it was down to the prescribing doctor. However

In 2013 the GMC published Guidelines (which did not exist in 1995) on prescribing unlicensed medicines. Extracts are set out in the Appendix, at Item 15.

According to Mr Hare, the RCOG Yearbook is a vehicle for publishing major literature in the preceding year and developments in practice. The Yearbook is marketed principally at consultants. Professor Thornton was cross-examined about a section in the 1995 Yearbook (Item 16) of which he was a co-author. It was put to him that what he said in that document was the polar opposite of his report (Footnote: ⁵⁵). He disagreed and made the following points:

In re-examination Professor Thornton was asked what an obstetrician should do in 1995 faced with the evidence, including the yearbook and two new pieces of literature (items 5 and 6) recently published in high impact journals. His response was that an obstetrician reading the Yearbook would know that Nifedipine was not recommended

In his report Professor Thornton had asserted that the administration of Nifedipine was widely undertaken in 1995. In evidence he said that he was in Cambridge in 1995. Cambridge was using Ritodrine. He described Cambridge as tending to be very conservative. His feeling was that in 1995 others were using Nifedipine. A proportion of units did favour Nifedipine in 1995. This was partly based on his retrospective analysis. That was his recollection and he regarded the use of Nifedipine at that point as being entirely reasonable.

The 1997 Defendant’s Protocol, written by Mr Fox and Mrs O’Sullivan said Nifedipine was as good as any other agent and was probably the safest drug for mother and baby. The reading list included the 1994 RCOG Guidelines and Kupferminc et al. (Item 5) (Footnote: ⁵⁶)

When Professor Thornton moved from Cambridge to Coventry in 1998, Coventry was using Nifedipine as a tocolytic. By 2005 53% of hospitals were using Nifedipine, 40% used Atosiban and Ritodrine was being phased out. (Footnote: ⁵⁷)

An important preliminary point is that the RCOG 1994 Guidelines (Item 10) and the 1997 Guideline (Item 13) both contain an endorsement in essentially the same terms. The endorsement is:

“These guidelines were produced under the direction of the Scientific Advisory Committee of the Royal College of Obstetricians and Gynaecologists as an educational aid to obstetricians and gynaecologists. These guidelines do not define a standard of care, nor is it intended to dictate an exclusive course of management. It presents recognised methods and techniques of clinical practice for consideration by obstetricians/gynaecologists for incorporation in to their practices. Variations of practice taking into account the needs of the individual patient, resources and limitations unique to the institution or type of practice may be appropriate”

This endorsement may properly be taken as some support for Professor Thornton’s view that the 1994 and 1997 Guidelines were limited in their intended remit. Their purpose was not to evaluate different tocolytics, much less to recommend a course of treatment i.e Ritodrine over Nifedipine.

Mr Hare accepted that it was reasonable to offer Nifedipine after the publication of Item 14 in 2002, though in the joint statement he added: “Many consider that the recommendation for the use of Nifedipine remains on inadequate evidence, especially with regard to safety.” He said that if the doctor is satisfied of the advantage of

Mr Hare agreed that the safety profile of Nifedipine had not changed since 1995 but said that doctors pay attention to authoritative Guidelines. In 2015 the NICE Guidelines recommended the use of Nifedipine as the first choice for tocolysis. Mr Hare said that NICE was written taking account of all considerations, including the cost of Nifedipine. His opinion was that, if the economics of Nifedipine were not taken into account, he did not think that NICE should recommend Nifedipine.

As to prescribing an unlicensed drug (Footnote: ⁵⁸), Mr Hare said that if it is given by a reasonable body of practitioners, it would need a consultant’s authority and would depend on whether there is a reasonable alternative that is licensed.

Mr Hare did not think that, even now, a doctor writing a Protocol on tocolysis should be suggesting Nifedipine. He said that no doctor, having informed him/herself of the research should write Nifedipine into such a Protocol. That is consistent with his view that the underlying research has not changed over the last 20+ years.

What therefore are the conclusions which the Court draws from this expert evidence? They are these:

I was taken to 5 particular papers between 1988 and 1994 (Items 1-3, 5-6). There were a number of other pre-1995 papers in the bundles to which I was not taken.

The 5 papers show:

Items 5 and 6 in 1993 and 1994 respectively (Footnote: ⁶¹) represent a much more positive basis for prescribing Nifedipine compared to Ritodrine. They were in highly respected journals. Certainly, although Mr Hare criticised Item 5 in particular, his evidence was that he was not aware of anything which has been published which demonstrates that they have been seriously undermined (Footnote: ⁶²). Item 5 is a research paper. It of course has a limited number of participants in the study, but it is an important part of a continuing and progressing picture. Item 6 is a review paper. It collates and assesses 90 references, the substantial body of literature on calcium channel blockers at the time.

I have already dealt above with Item 10 and why that cannot be regarded as evaluating different tocolytics or recommending Ritodrine over Nifedipine. As to Item 16, Professor Thornton’s 1995 Yearbook says, for example:

I take into account what Professor Thornton said in evidence about Item 16. I accept what he says. In my judgment he was, as a then Senior Registrar, setting out the conservative viewpoint rather than the avant garde or ‘cutting edge’ viewpoint. He does not mention the recent literature on Nifedipine. His task in writing was not to advocate change. At that time he was at Cambridge where the consultants supervising him still used Ritodrine. He described Cambridge as ‘very conservative’. Item 16 does not negate a contention that a responsible body of obstetricians was using Nifedipine in 1995.

In my judgment, the literature as at April 1994 was such that it would be wrong to say that practitioners who sought to use tocolysis would be in breach of their duty of care.

What evidence is there of practitioners using Nifedipine in 1995? The answer is that there is relatively little. It does not follow from this that only a tiny minority was doing so. Over 20 years have passed, there are no extant surveys and no wide-ranging documentation from other hospitals. The 2002 RCOG Guidelines (Item 14), before recommending the use of Nifedipine over Ritodrine, says: “There is little reliable information about current clinical practice but it is likely that Ritiodrine hydrochloride..remains most widely used.” There was a sort of survey which led to the letter in the BJOG that by 2005 53% of hospitals were using Nifedipine, but that was after the 2002 RCOG Guidance, so tells us nothing about 1995. It is simply not known whether very few or a substantial minority of hospitals were then doing so. The only evidence we have about 1995 is:

Thus, after this period of time there are some pieces of evidence from recollection and from the literature that Nifedipine was being used in the UK, the USA and the Netherlands before and during 1995.

In addition there was Professor Thornton’s recollection. It is difficult to give great weight to that, but it is in line with the fact that a number of hospitals were then administering Nifedipine as a tocolytic.

Taking the evidence as a whole, and weighing the risks and benefits of using Nifedipine as they were known in 1995, Professor Thornton’s opinion is eminently defensible on this point. It has also been borne out by subsequent events such that Nifedipine is now the drug of choice for tocolysis, despite the lack of any further hard primary research. It was also borne out by the 1997 paper (Item 7) which reference 8 papers which were from 1995 or earlier for the proposition that:

“Studies comparing ritodrine with nifedipine in the management of preterm labour suggest a similar tocolytic efficacy but fewer maternal side effects and no adverse fetal side effects with nifedipine.”

This large-scale study then examined this finding, effectively corroborated it and found other advantages of Nifedipine over Ritodrine.

For those reasons it was not a breach of the duty of care to prescribe Nifedipine as a tocolytic in 1995.

There had been an allegation that it was negligent for Mrs Harris to have been given Nifedipine sublingually, rather than orally. However, there were a number of references in the literature of this practice. Mr Hare accepted that, if it was reasonable to prescribe Nifedipine, then it was not a breach of duty to have administered it sublingually.

In the 1997 Protocol, the Nifedipine Regimen requires: “Site a venflon and start infusion of Hartmann’s solution.” It is not known what any 1995 Protocol required. Doctor Bett said she would have followed the requirements of a Protocol.

In his 1^st report, Mr Hare acknowledged that, though in some of the studies setting up an i/v infusion was thought to be a necessary prelude, it is not mentioned in others. His

In evidence Mr Hare said that the incidence of hypotension is reported as high as 40%. Therefore, not to have the precaution of i/v infusion should hypotension occur is unacceptable. He accepted that the 2015 NICE Guidelines do not require an i/v infusion to be set up. He did not know why this was the case.

It was clarified that there was no separate allegation that, if i/v infusion had not been set up prior to administration of Nifedipine (Footnote: ⁶⁵), then it should have been set up later. Mr Hare said that, though he thought that should be done, the beneficial effect would have been minimal.

Professor Thornton said that he disagreed with the 1997 Protocol in suggesting that a cannula should be put in advance into a woman with threatened preterm labour. A cannula should not be put into a vein unless there is a good reason to do so. He said that if hypotension followed a dose of Nifedipine, the first response should be to lift up the woman’s legs. If there was real concern, then it would not be unreasonable to put up an i/v line and start fluids. In Mrs Harris’ case he would not have been concerned if the blood pressure fall was Nifedipine related. The fall had been only for a minute or so and was not much of a fall.

There is therefore conflicting evidence from Mr Hare and Professor Thornton on this issue. On the basis of Professor Thornton’s evidence, which is logical and consistent with surrounding evidence – e.g. the 2015 NICE Guidance and a number of respectable studies – I do not find that there was any breach of duty on this basis. I find that a reasonable body of clinicians would not have set up an i/v line prior to the administration of Nifedipine.

Mr Hare’s evidence as to whether a 2^nd dose of a tocolytic drug was indicated depended entirely on the evidence on contractions. If, contrary to his opinion, the midwife’s examination at 0120 demonstrated continuing contractions, then it was reasonable to give a 2^nd dose. Professor Thornton said that once the regime for administration had started, a doctor would follow the Protocol. Protocols differed. The key question was whether it was justified to start tocolysis. In order to start it, he would expect there to have to be contractions. If contractions continued it would be reasonable to give a 2^nd dose. If they did not continue, since the initial dose was a relatively small dose of 10mg, there would be no hard line that a 2^nd dose should not be given. In any event, although he accepted that at 0120, it does not suggest that Mrs Harris’ abdomen was palpated again, Professor Thornton’s understanding of the note which refers to her abdomen being painful “in waves”, was that Mrs Harris was still complaining of contractions before the 2^nd dose was given at 0135 (Footnote: ⁶⁶).

Doctor Bett said she would have written in the drug sheet “If contractions cease then no further Nifedipine”, as that is what the Protocol must have said (Footnote: ⁶⁷). I accept that.

I find on both grounds for the Defendant on this issue. That is to say:

(1) It was reasonable for Doctor Bett to diagnose, as she did, that Mrs Harris was in threatened preterm labour. It was therefore reasonable to prescribe tocolysis

In summary, the Claimant has not proved any breach of duty on the part of the Defendant and the claim must therefore fail.

I have previously stated in this judgment that I shall try the issue of the prescription of Nifedipine as a tocolytic drug by the standards of November 1995, not subsequently. This accords with the traditionally understanding of the authorities. On that basis I have found for the Defendant. However, there were brief submissions by Mr Moon QC that there is nothing in the Bolitho test that requiresme to do this. If a doctor who would have been held liable in 1995 for breach of duty in prescribing a drug whose use was not accepted as appropriate by a responsible body of practitioners is subsequently vindicated, such that a doctor prescribing the same drug in 2002 would not be in breach of duty because of changes of opinion in the profession, should the 1995 doctor be held to be negligent in a trial taking place after 2002? The point has not seemingly arisen before. Mr Sweeting QC submitted that a Claimant is entitled to be treated by reference to the standards at the time of treatment.

It is not necessary for me to decide the point and I do not do so, leaving it for consideration of a higher court if and when it arises.

Nifedipine versus Ritodrine for suppressing preterm labor. Meyer et al. 1988 The Journal of Reproductive Medicine.

“Fifty eight women in preterm labor were selected randomly to receive either oral nifedipine or intravenous ritodrine hydrochloride. In comparison to ritodrine, nifedipine had similar tocolyltic efficacy with fewer adverse maternal and fetal side effects … Preliminary data suggest that nifedipine is a safe, effective and well-tolerated tocolytic agent. It may prove to be a suitable alternative to ritodrine hydrochloride, especially for women in whom beta-sympathomimetics are contraindicated.

Introduction

…

The efficacy of sympathomimetics, including Ritodrine hydrochloride, is generally well accepted, yet the potential side effects have lessened enthusiasm for its use. Currently, interest centres on the tocolytic use of calcium channel blockers. The aim of the present investigation was to compare the tocolytic efficacy and safety of nifedipine and ritodrine hydrochloride.

…

Results

Between August 1986 and February 1987, 34 women were randomly selected to receive nifedipine, while 24 received ritodrine hydrochloride …

Conclusion

Our study lent support to the clinical use of nifedipine for preterm labor. In comparison to ritodrine hydrochloride, nifedipine demonstrated similar tocolytic efficacy with less severe maternal side effects and metabolic alterations. Fetal homeostasis as measured by external fetal monitoring … appeared unaltered. Prematurity, with its inherent dangers, seems a much greater risk to the fetus than does nifedipine tocolysis.”

The safety and efficacy of Tocolytic Agents for the treatments of preterm labor: Besinger and Niebyl. Obstetrical and Gynaecological survey 1990, by Williams and Wilkins (USA) Vol 45 No. 7, page 415.

“… a diverse variety of tocolytic medications have been proposed for clinical use, with betamimetics and magnesium sulphate being the common therapeutic agents of choice in the United States today. The clinician using these agents should be aware of the significant maternal and fetal side effects associated with these particular medications. New classes of pharmacological agents, including … calcium channel blockers … have been proposed as tocolytic agents and are currently undergoing critical evaluation. The purpose of this review is to provide a compilation of the available clinical studies that document the safety and efficacy of these various tocolytic agents.

…

In 1980 the United States Food and Drug Administration approved Ritodrine hydrochloride … for inhibition of preterm labor. Among all the tocolytic agents to be discussed in this article, this is the only one so approved and the other agents should be considered experimental. However, the use of approved drugs for nonlabelled indications may be entirely appropriate based on medical advances extensively reported in the medical literature … the purpose of this article is to provide a compilation of the available clinical studies that document the efficacy and safety of these various tocolytic agents

…

Beta-Adrenergic Agonists (Footnote: ⁶⁸)

…

The intravenous administration of betamimetic agents can stimulate beta-receptors in multiple organ systems and is responsible for the various clinically significant side effects associated with these medications (33). Maternal cardiovascular side effects are most frequently seen with beta-adrenergic agonists, including hypotension, tachycardia, and arrhythmia.

…

As the clinical use of beta-adrenergic agonists has become more widespread, more than 80 cases of pulmonary edema have been reported in the literature (39, 51-70). This life-threatening complication has been reported in up to 5% of patients receiving intravenous betamimetic therapy (35, 39).

…

Several maternal deaths have been associated with administration of betamimetic therapy (57, 77, 78). While most of these maternal debts were not directly attributable to the drug, a majority of these patients had a history of cardiac disease or pulmonary hypertension, pointing out the importance of pretreatment screening for cardiopulmonary disease.

…

A more recent multicentre European study of 99 patients randomised to receive either intramuscular Ritodrine or placebo followed by oral therapy did not show long term efficacy … these more recent studies raise serious questions regarding the efficacy of Ritodrine in the treatment of preterm labor.

Comparison of Nifedipine and Ritodrine for the Treatment of Preterm Labor. Bracerio et al. American Journal of Perinatology/Volume 8, number 6, November 1991.

“ABSTRACT

Treatment of preterm labor with beta-sympathomimetics has been questioned because of the many maternal and fetal complications associated with its use. Nifedipine, a calcium antagonist, has been shown to suppress uterine activity in vitro and in vivo. A randomised prospective study was performed to compare the efficacy of Nifedipine to Ritodrine in the suppression of preterm labor. Data obtained from 42 women, of which 19 were randomised to the Ritodrine group and 23 to the Nifedipine group, were analysed. Ritodrine and Nifedipine were proved to be equally effective in the suppression of perterm labor. However, the Nifedipine group had fewer maternal and fetal complications.

…

There is a need to find a tocolytic agent that is efficacious and has fewer side effects than the agents currently in use. The most widely used agents are beta-sympathomimetics. These agents are associated with several side effects in both mother and fetus …

DISCUSSION

… therefore, calcium antagonists seem ideally suited to suppress premature uterine smooth muscle contractions. On the other hand, the wisdom of using beta-sympathomimetics to treat preterm labor is being questioned. A review article on the treatment of preterm labor includes that the use of Ritodrine should be discouraged because of its questionable efficacy and life-threatening maternal adverse side effects. …

Harake and associates report that on pregnant sheep Nifedipine decreases uterine blood flow and fetal arterial oxygen content. In published human studies there has been no indirect confirmation of Harake’s finding in sheep. In our study antepartum fetal heart rate monitoring did not reveal any significant abnormalities…

…

Most encouraging of all, neonatal RDS (Footnote: ⁷⁰) and other morbidities were lower in the Nifedipine group, resulting in a significantly lower hospital stay for these infants. Additionally, women who receive Nifedipine experienced fewer subjective and objective drug side effects.

SUMMARY

This is a small randomised trial in which both Nifedipine and Ritodrine prove to be effective in suppressing preterm labor. Nifedipine, however, had fewer maternal side effects and less neonatal morbidity.”

Best Practice in Labour and Delivery, 2^nd edition (2016). Chapter 21 The Management of Preterm Labour.

“Historical perspective

…

Nifedipine versus Ritodrine for Suppression of Preterm Labor. Kupferminc et al. British Journal of Obstetrics and Gynaecology December 1993, vol 100 pp 1090-1094.

“ABSTRACT

Objective To compare the efficacy of tocolysis with specific regimens of Nifedipine and Ritodrine. Maternal side effects and neonatal outcome also were evaluated.

Design A prospective, randomised trial.

Subjects 71 women, including 11 with twin pregnancies, who had uterine contractions and observed cervical changes.

…

Conclusions Nifedipine is as effective as Ritodrine in suppressing preterm labour. Its uses are associated with less frequent side effects.

Preterm delivery is a common obstetric problem. The incidence of preterm delivery is about 7 to 9% of pregnancies, and it accounts for as much as three quarters of the mortality and morbidity among newborns without congenital abnormalities … Therefore major emphasis has been placed on perinatal and neonatal research directed at preventing and lessening the consequences of preterm birth …

The pharmacological choices limited by the number of drugs available and by their safety and side effects, thus necessitating a continuous search for effective drugs with minimal side effects. Currently the most commonly used tocolytic agents are betaadrenergic drugs, particularly Ritodrine. However, the incidence of troublesome, and occasionally fatal, side effects associated with this drug are of serious concern.

…

There is a growing body of evidence that Nifedipine is effective in suppressing preterm labour with minimal maternal and fetal side effects. In this paper we present the results of a prospective randomised study which was designed to compare the efficacy of oral Nifedipine with our existing regimen for administration of Ritodrine.

…

Discussion

This study shows that Nifedipine is an effective tocolytic agent, comparable to Ritodrine, but it causes fewer side effects and less haemodynamic compromise.

…

Since calcium channel blockers are known to have both a vasodilatory effect and a negative inotropic effect on the myocardium …, haemodynamic side effects are of concern … Ferguson et al … reported a statistically significant increase in maternal heart rate and decrease in both diastolic blood pressure and MAP after sublingual and oral administration of Nifedipine, but they considered these changes unlikely to be of physiological importance. Our findings are consistent with theirs. The decrease in blood pressure which we observed after oral administration of Nifedipine, although statistically significant, was unlikely to be of clinical importance and was significantly less than the decrease associated with Ritodrine.

Nifedipine caused an increase in maternal heart rate following each dose, but this was transient and much less pronounced compared with women treated with Ritodrine. Similar observations were reported by Ferguson et al … No significant changes were noted in the fetal heart rate.

Consistent with previous reports (Ulmsten et al 1980, 1984), we found that other side effects associated with oral Nifedipine were trivial, and less common than with Ritodrine even in women who received a 2^nd dose of Nifedipine. Treatment with Ritodrine, however, often requires discontinuation due to severe side effects and complications …

We conclude that Nifedipine is a useful tocolytic agent comparable in efficacy to Ritodrine, but with a lower frequency of side effects.”

Holmes Childress and Katz Obstetrics and Gynaecology, Volume 83, number 4, April 1994, page 616.

“Objective: to review studies and investigations regarding the safety and efficacy of Nifedipine.

Data sources and methods: we reviewed the published literature on calcium channel blockers and their pharmacology and therapeutic applications in obstetrics and gynaecology. We paid particular attention to methods of animal research and recent clinical evaluations.

Conclusions: the dihydropyridine group of calcium channel blockers … and, specifically, Nifedipine are safe for use in pregnancy. They have little teratogenic (Footnote: ⁷¹) or fetotoxic potential. Nifedipine’s mechanism of action is through smooth-muscle relaxation secondary to blockage of the slow calcium channels into the cells. In vivo, there is minimal effect on the cardiac conducting system. Multiple studies in women have demonstrated the effectiveness and safety of Nifedipine as an antihypertensive. … Nifedipine is as effective as beta-mimetics in decreasing uterine activity. As a tocolytic agent, it is more effective as there are fewer patients who have to discontinue with Nifedipine because of side effects. …

Fetal effects

…

Numerous investigators have reported positively on the fetal and neonatal effects of Nifedipine over prolonged periods during pregnancy. Studies comparing maternal treatment with Nifedipine and other agents have found that fetuses treated with Nifedipine fared at least as well and sometimes better (because of fewer maternal side effects) than fetuses exposed to other medications. Short-term fetal effects from Nifedipine have also been examined. The fetal heart rate during labour, fetal heart rate shortly after maternal ingestion, and umbilical artery Doppler flow studies have not been adversely affected by Nifedipine.

Animal studies have shown varying effects from the dihydropyridine agents. Harake et al, Parisi et al, and Ducsay et al found deleterious effects from high doses of Nifedipine, including declines in fetal arterial oxygen content, acidosis, and fetal deaths in lambs after maternal Nifedipine and Nicardipine treatment. These early reports caused widespread concern among physicians. In contrast, other investigators did not find adverse fetal effects. The contrasting results of animal and human study may be related to maternal haemodynamic effects, changes in the distribution in placental blood (an effect which is unlikely to occur in humans), and the dosages used. Taking all the studies together the large number of women treated with Nifedipine without adverse fetal effects suggests that the animal studies may not be applicable to pregnant women. Nifedipine has not been shown to affect birth weights, even with prolonged fetal exposure. There are fewer studies of Nitrendipine and Nicardipine, but abnormal effects have not been found with these agents, either.

The use of Nifedipine for tocolysis

Nifedipine and Ritodrine in the management of pre-term labor: a randomised multicenter trial. Papatsonis et al., Obstetrics and Gynaecology Volume 90(2) August 1997

“Objective: to compare the efficacy of Nifedipine with Ritodrine in the management of preterm labor.

Methods: 185 singleton pregnancies with preterm labour were assigned randomly to either Ritodrine intravenously (n = 90) or Nifedipine orally (n = 95)…

……..

Conclusion: Nifedipine in comparison with Ritodrine in the management of preterm labour is significantly associated with a longer postponement of delivery, fewer maternal side effects, and fewer admissions to the NICU…

Chapter 22

“The early onset of labour

…

Treatment with Ritodrine by intravenous fusion for 24-48h, followed by oral administration for 5-7 days, has been described by Wesselius de Casparis et al. (1971). The drug is now used quite frequently in clinical practice, though its efficacy appears to remain unverified by controlled clinical trials (Hemminki and Starfield 1978; O’Connor et al, 1979). Furthermore, there have been disturbing reports of pulmonary oedema in mothers following its use in conjunction with betamethasone to supress premature labor (Elliot et al, 1978; Tinga & Aarnoudse 1979), and one of Ritodrine-induced acidosis in pregnancy (Desir et al 1978). (Footnote: ⁷³)

…

Calcium agonists such as Nifedipine (probably magnesium sulphate also) are being investigated currently, but only limited data are so far available (Read and Wellby 1986; Odum & Pipkin, 1988). …

…

Probably the beta-adrenoagic tocolytic drugs are the agents most widely used, but they are not without significant risk to the mother in some circumstances, and in most women are associated with unpleasant side effects, including flushing, tremor, headache, sweating and tachycardia. Alternative agents such as prostaglandin inhibitors, calcium antagonists and oxytocin inhibitors, have as yet been insufficiently tested to be introduced into routine use.”

Turnbull’s Obstetrics 2^nd edition 1995, Chapter 33. Preterm labour and delivery of the preterm infant

“3a. Therapies to improve outcome: delaying delivery

Several drugs are now available to delay delivery in spontaneous preterm labour and where possible these should be utilised to allow other therapies which may improve outcome to be given.

There are few complete contraindications to the inhibition of

preterm labour … Current drugs used are betamimetics, prostaglandin synthetase inhibitors, magnesium sulphate, Calcium Channel Blockers and antibiotics.

Beta-Adrenergic Agonists

…

Safety. Maternal side effects of betamimetics are well documented …

Most of the concern over the use of beta-agonists centres on the severe cardiovascular side effects…

The greatest controversy has been reserved for the risk of pulmonary oedema which occurs in 0.3% of cases (Canadian Preterm Labour Investigators Group 1992) … Whatever the cause, the injudicious use of beta-agonists can be lethal and appropriate monitoring needs to be in place…

…

Administration

………..

Beta-agonists should only be used where there is good evidence of preterm labour …

…

Calcium Antagonists

…

Safety. … Severe side effects are extremely rare. Initial concerns over fetal welfare have been largely dispelled (Hanretty et al 1989).

…

Efficacy. Clinical experience is limited with Nifedipine. Trials to date have not been of high quality, and have not demonstrated any benefit over Ritodrine in the prolongation of pregnancy (Bracero et al, 1990; Ferguson et al., 1990; Meyer et al., 1990). It appears well tolerated, and no adverse fetal or neonatal effects have been reported. Its use should be confined to appropriate trials at present but its apparent safety justifies continuing investigation.

RCOG April 1994 Guidelines“For The Use of Ritodrine”

“1.

Introduction

The risk-benefit ratio of Ritodrine hydrochloride …, a betaagonist licensed to inhibit preterm uterine activity, has been reviewed recently following a series of events. These include the publication in July 1992 of a paper in the New England Journal of Medicine by the Canadian Pre-term Labor Investigators Group that indicated that the drug had no beneficial effects on perinatal mortality but was associated with increased maternal morbidity – particularly the potential to cause pulmonary oedema in the mother. Shortly after this publication a further two fatal cases of complications relating to pulmonary oedema were reported to the Committee on Safety of Medicines (CSM); the pharmaceutical company circulated a letter to doctors reminding them of this side effect and emphasising measures to minimise the risks; and the data sheet has been reviewed and revised.

2.

Background

Ritodrine hydrochloride is a beta-agonist … The drug was introduced in the UK in 1974 and has been widely used for the inhibition of preterm labour in the UK, rest of Europe and the USA. …

3.

Effectiveness of Ritodrine

The Canadian Pre-term Labour Investigators Group represents the largest placebo-controlled investigation conducted on any tocolytic agent. Seven hundred and eight women were randomised to intravenous Ritodrine … The results indicate that Ritodrine significantly reduced the proportion of women who delivered within 24 and 48 hours after treatment, but there was no significant difference in birth weight or neonatal morbidity overall. …

Drugs and Therapeutics Bulletin, Volume 30, number 25, 7^th December 1992: Prescribing unlicensed drugs or using drugs for unlicensed indications

“Unlicensed use –

…

….even when the prescriber is fully aware of the contents of the data sheet there are occasions when non-adherence seems justified, for example where:

•

the licensed indications do not reflect current knowledge. The data sheet for prednisolone … recommends that the drug be given in ‘divided doses’ through the day but clinical evidence strongly favours once daily or even alternate-day dosing.

•

the indications do not include well proven uses of a drug. Mountain sickness, a recognised use for acetazolamide, is not mentioned in the data sheet, nor is treatment of dystonia with benzhexol or more recently the use of magnesium sulphate for acute myocardial infarction.

•

the licensed indications are over restrictive. Junifen and Brufen syrup both contain 100 milligrams Ibuprofen in 5ml liquid yet for Brufen the indications include use as an anti-inflammatory, and for Junifen as an antipyretic. Brufen costs much less than Junifen, and on cost grounds doctors would be justified in prescribing Brufen syrup for both indications.

Such anomalies may arise if the manufacturer does not wish to alter the product’s market niche and/or is unwilling to sponsor the trials necessary to support an application for a licence change.”

A Guide to Effective Care in Pregnancy and Childbirth,2^nd edition, (1995). Enkin et al.

“3 Treatment of active preterm labour.

……..

3.7

Other drug treatments

……..

Apart from trials in which calcium antagonists were used mainly to supplement labour-inhibiting treatment with betamimetic drugs, there have been few attempts to evaluate these agents in preterm labour. There are not enough data on any of these agents to justify their use outside the context of well-designed and carefully monitored randomised trials.”

RCOG Guideline January 1997. Beta-agonists for the care of women in preterm labour.

“1.

Introduction

…

A wide variety of agents have been advocated as suppressing uterine contractions. Currently the most widely used is Ritodrine hydrochloride, a beta-agonist………. The aim of this guideline is to summarise the evidence about the effectiveness of betaagonists for prevention and treatment of preterm labour and to provide guidance as to how to incorporate this evidence into clinical practice.

2.

Evidence from systematic reviews

… Taken together, these studies show that beta-agonists reduce the proportion of deliveries occurring within the first 48 hours after beginning treatment. This is not reflected in any decrease in perinatal mortality or serious morbidity…

3.Risks and side effects

The most common symptoms associated with beta-agonist use are palpitations, tremor, nausea, vomiting, headache and restlessness … Beta-agonists are also associated with hypotension, although this is less of a problem with Ritodrine than some of the earlier agents …

Rare but serious side effects have also been reported following beta-agonist use. These are potentially life threatening and there have been a small number of maternal deaths associated with the use of beta-agonists. Pulmonary oedema is a well-documented complication and most cases are associated with aggressive intravenous hydration. Myocardial ischaemia is another uncommon but serious side effect and is a consequence of the increased cardiac output associated with beta-agonist administration. Women with known cardiac disease should not be given a beta-agonist. Other drugs, such as calcium antagonists and beta blockers, have been tried as adjuncts to beta-agonists in an attempt to reduce the cardiovascular side effects. None have been shown to have the desired effects and the available data do not justify their use…”

Tocolytic drugs for women in preterm labour

“1.Purpose and scope

…

A wide variety of agents have been advocated as supressing uterine contractions. Those in current use include beta-agonists, calcium channel blockers, prostaglandin synthetase inhibitors, nitrous oxide donors and oxytocin receptor antagonists. There is little reliable information about current clinical practice but it is likely that Ritodrine hydrochloride, a beta-agonist, remains the most widely used. …

… The aim of this guideline is to summarise the evidence about the effectiveness of tocolytic drugs for preterm labour and to provide guidance as to how to incorporate this evidence into clinical practice.

…

4.Choice of tocolytic drug

If a tocolytic drug is used, Ritodrine no longer seems the best choice. Atosiban or Nifedipine appear preferable as they have fewer adverse effects and seem to have comparable effectiveness. Atosiban is licensed for this usage in the UK but Nifedipine is not.

…

6.

Summary

There is still no clear evidence that tocolytic drugs improve outcome following preterm labour and so it is reasonable not to use them. … There is insufficient evidence for reliable conclusions about more substantive effects on perinatal or infant mortality or on serious neonatal morbidity. It remains plausible that, for selected women such as those who require transfer for neonatal care or time to complete a course of corticosteroids, there may be benefit associated with tocolysis. However, this benefit has not been formally evaluated in randomised trials.

If a tocolytic agent is used, Ritodrine no longer seems the best choice. Alternatives such as Atosiban or Nifedipine appear to have comparable effectiveness in terms of delaying delivery for up to 7 days and are associated with fewer maternal adverse effects. Atosiban is licensed for use as a tocolytic but the purchase price is relatively expensive. Nifedipine is not licensed for use as a tocolytic and the ideal dosage and formulation are unclear. For both these agents, further evidence is required about their relative effects on substantive outcomes such as neonatal mortality and morbidity, and on safety and long-term outcome for the child.

In view of the current lack of evidence for any substantive benefit for the baby from tocolysis, and the possibility of hazard for the mother, the available evidence should be discussed with the woman and her partner and their preferences taken into account in determining her care.”

General Medical Council (2013)

“Prescribing unlicensed medicines.

………..

68.

You should usually prescribe licensed medicines in accordance with the terms of their licence. However, you may prescribe unlicensed medicines where, on the basis of an assessment of the individual patient, you conclude, for medical reasons, that it is necessary to do so to meet the specific needs of the patient.

69.

Prescribing unlicensed medicines may be necessary where:

a)

there is no suitably licensed medicine that will meet the patient’s need …

70.

When prescribing an unlicensed medicine you must:

a)

be satisfied that there is sufficient evidence or experience of using the medicine to demonstrate its safety and efficacy

b)

take responsibility for prescribing the medicine and for overseeing the patient’s care…

c)

…

Information for patients about the licence for their medicines

71.

You must give patients … sufficient information about the medicines you propose to prescribe to allow them to make an informed decision.”