Birch v University College London Hospital NHS Foundation Trust

This is a clinical negligence action in which liability is the issue. Quantum has been agreed at the sum of £621,000 as at the date of judgment. The claimant is Mrs Janet Birch, who suffered a stroke caused by a cerebral catheter angiogram at the defendant’s National Hospital for Neurology and Neurosurgery, Queen Square, on 21 June 2003. The angiogram involved inserting a catheter at the groin and then working it up through the vascular system to the area of the brain to be scanned. It was undertaken to exclude the possibility that Mrs Birch had a posterior communicating artery aneurysm. This is potentially life threatening. Given her history and presentation, Mrs Birch contends that Queen Square was negligent in deciding to undertake an angiogram and that the investigation of her condition should have been by the non-invasive method of magnetic resonance imaging (MRI). My conclusion is that on established legal principles the neurosurgeons and neuroradiologists at Queen Square were not negligent. However, the defendant is liable to Mrs Birch because it did not properly obtain her consent to the angiogram. Despite a paucity of authority my view is that in the special circumstances of Mrs Birch’s case the hospital should have discussed with her the different imaging methods – catheter angiography and MRI – and the comparative risks. Had it done so Mrs Birch would have declined catheter angiography and thus avoided her stroke.

The claimant in this action, Mrs Janet Birch (“Mrs Birch”) was born in September 1947 and so was 55 at the time of the events on 20 June 2003 giving rise to this action. She has suffered from Type 1 diabetes since the age of 35 and is insulin dependent. The defendant is responsible for management and control of the National Hospital for Neurology and Neurosurgery, Queen Square, London (“Queen Square”) and for the care and treatment provided there to patients. Queen Square is a tertiary referral centre for the specialised investigation and treatment of neurological abnormalities and diseases. It receives many referrals for suspected cerebral vascular lesions and is a sub-specialist centre of excellence for post graduate training.

On 18 June 2003 Mrs Birch woke up with a bad headache, blurred vision and her left eye in towards her nose. She had not been feeling well for a few days. She had had vision problems in the past. In 2001 she had had focal laser treatment because of diabetic retinopathy in her left eye. The following year her consultant ophthalmologist had examined her after she had been complaining of diplopia for several weeks. She had right lateral rectus palsy, suggestive of a micro-angiopathic lesion in the sixth cranial nerve. The symptoms resolved themselves naturally. On that occasion an MRI scan of her brain was undertaken but no abnormality was detected. Then Mrs Birch experienced similar double vision a few months before the events which concern me. She had seen her ophthalmologist, had a scan but again the symptoms passed. It seems it was ischaemic sixth cranial nerve palsy. On 16 June she saw her GP, who thought she had sinusitis and recommended antibiotics.

When Mrs Birch presented herself to her GP on 18 June, with the symptoms she described, he arranged an emergency admission to Watford General Hospital (“Watford”). At Watford she was kept in for observation. During her time there the notes recorded her persistent headache, its gradual onset, her nausea and vomiting and right periorbital swelling and ptosis. At one point her temperature was noted as 37.3˚c, at another 37.6˚C, but she was also noted as afebrile by the consultant physician when examined on the evening of the 18^th, and in the transfer notes from Watford to Queen Square. A test found that her diabetes was under very poor control. The specialist registrar at Watford who saw her noted that there was no objective clinical evidence to support a diagnosis of a sub-arachnoid haemorrhage or meningism but recommended that a neurologist see her and that a CT scan be performed to rule out sub-arachnoid haemorrhage.

On the morning of 20 June, a Friday, Mrs Birch was seen by Dr Gavin Giovannoni. At the time he had honorary consultant contracts with both Queen Square and Watford. Since late 2006 he has been Professor of Neurology at the Royal London Hospital, Barts and The London NHS Trust. Professor Giovannoni saw Mrs Birch as a ward referral on his round at Watford. He had the hospital clerking notes. His view was that she was likely to have a painful pupil-sparing right third nerve palsy. In his experience the most likely diagnosis of a painful pupil-sparing third nerve palsy in a patient with long-standing diabetes was due to occlusion of a small blood vessel supplying the third nerve, usually referred to as a “vascular third nerve palsy”. He had seen many such cases, including those during the four years he conducted a clinic at the Moorfield Eye Hospital. It is a benign condition and treatment in such cases is conservative, with care by opthalmagists, physicians or neurologists, the condition typically disappearing in six weeks. However, Mrs Birch was atypical in some respects with, for example, the swelling around her eye, the gradual onset of her symptoms, the nausea and the possible infection which the GP had treated with antibiotics. “Wait and see” was not an option. Professor Giovannoni therefore recommended an urgent MRI. His notes concluded:

“PAINFUL PUPIL-SPARING RIGHT III PALSY. Plan:

(i)

URGENT MRI → exclude:

1.

POST-com [artery] aneurysm.

2.

CAVERNOUS SINUS PATHOLOGY.

(ii)

Please let me know outcome of scan.”

P-Comm aneurysm, or posterior communicating artery aneurysm, in this context refers to a bulge in the wall of a blood vessel behind the right eye which may have been affecting the third (III) nerve behind the right eye, leading to the condition of painful pupil-sparing right third nerve palsy. The risk of such a condition is that the aneurysm, or bulge in the wall of the blood vessel, may rupture, leading to bleeding in the skull, and in all probability, death. Cavernous sinus pathology affects one of the two cavernous sinuses on either side of the sphenoid bone in the skull, through which the nerve passes. It can be in the form of a tumour, thrombosis or infection and, if untreated, can produce a dangerous neurological deterioration, with risk of brain damage or death. Professor Giovannoni was not intending to indicate an order of priority in terms of the investigation. Indeed he thought cavernous sinus pathology more likely, the swelling being a pointer away from an aneurysm. He wanted MRI for Mrs Birch. This is non-invasive in character but gives sectional imaging of the skull, including the sinuses. MR Angiography supplies mapping of the blood vessels by means of a dye injected into them. (For convenience both MRI and MR Angiography are referred to as MRI in this judgment). In Professor Giovannoni’s opinion MRI would have satisfactorily excluded both a posterior communicating artery aneurysm and cavernous sinus pathology. He did not want cerebral catheter angiography (described in the judgment as an “angiogram” or “catheter angiography”) since this would only have excluded a posterior communicating artery aneurysm but not cavernous sinus pathology. Moreover, an angiogram was invasive and had increased risks for someone like Mrs Birch, who was a diabetic.

Following Professor Giovannoni’s recommendation, an MRI Request Form was completed, indicating that Mrs Birch had painful third nerve palsy and that the possible causes to be investigated were posterior communicating aneurysm and cavernous sinus pathology. However, Professor Giovannoni’s plan for an urgent MRI scan was thwarted first, by the view of the consultant radiologist at Watford that given the normal CT scan it was not urgent, and secondly, because in any event there were no slots available at the Hemel Hempstead General Hospital (part of the same NHS Trust) or at the Mount Vernon Hospital, where Watford at the time had MRI scans performed. That afternoon, the Friday, the medical team at Watford saw Professor Giovannoni in his outpatients clinic to discuss the CT scan result, which was now available. The result of the CT scan, which was clear, did not change his opinion although it reduced the possibility of there being cavernous sinus pathology. Professor Giovannoni asked the medical teams to contact the National Hospital for Neurology and Neurosurgery at Queen Square, London (“Queen Square”) so that Mrs Birch could be transferred there for investigation. They informed him that they were contacting the Neurosurgeons at Queen Square, because Professor Giovannoni’s specialist registrar at Queen Square, Dr McCabe, had informed them that there were no neurology beds available. However, he was expecting Mrs Birch to be transferred when a neurology bed became available.

The next that Professor Giovannoni heard of Mrs Birch was the following evening, the Saturday, when he was contacted by the neurology specialist registrar on call, informing him that “my patient”, Professor Giovannoni’s patient, Mrs Birch, had indeed been transferred to Queen Square under the care of the neurosurgeons the night before, had had catheter angiography that morning, but that there had been complications by the development of a dense left hemiplegia, presumably due to a cerebrovascular accident. Professor Giovannoni was still of the view that a catheter angiography had not been appropriate for Mrs Birch given that an aneurysm was very unlikely and the risks of the procedure outweighed the benefits in her case.

The referral letter for Mrs Birch from Watford to Queen Square read:

“Thank you for accepting this 55 year old lady who presented with a gradual onset headache behind the right eye. She has a painful third nerve palsy and was seen today by Dr Giovannoni who felt that a posterior communicating aneurysm or cavernous sinus pathology need to be ruled out. CT – NAD [no abnormalities detected].”

It went on to indicate that she had been a diabetic of twenty years. There was nothing in the letter about Professor Giovannoni’s recommendation for an urgent MRI. However, some sixteen pages of the Watford notes accompanied Mrs Birch, including Professor Giovannoni’s plan, when she arrived at Queen Square on late Friday night, the 20^th. She was admitted to a Neurosurgical ward. Mrs Birch was first seen by the senior house officer that evening. He noted a history of “right-sided eye headache” with associated vomiting and nausea. There was a decrease in vision in the right eye, which was painful. She had never had this before. Her diabetes was noted. He thought the right side of her face was puffed up. He noted that she had had a CT scan, which was normal and had a painful third nerve palsy. Professor Giovannoni’s differential diagnosis was noted. The plan was review by the specialist registrar to consider an MRI scan or an angiogram in the morning. (The notes read “For R/V by SPR? MRI? ANGIO MANE [in the morning]: NBM [nil by mouth] from midnight”).

On the Saturday, around lunch-time, a catheter angiogram was performed on Mrs Birch. The key persons involved were the following:

Apart from Dr Al-Jeroudi, none of these doctors have any recollection of Mrs Birch’s case. No criticism can be levelled at them because of this, given the length of time which has elapsed but more importantly the many, often serious, cases which they have been involved with since. Unfortunately the Queen Square records leave something to be desired both before and subsequent to Mrs Birch’s treatment. The first example of this relates to references to a fifth person, Professor David Thomas. In the Queen Square admission notes he is referred to as the consultant to whom Mrs Birch was referred. In 2003 Professor Thomas was an honorary consultant at Queen Square and Professor of Neurosurgery in the Institute of Neurology, University College, London. Professor Thomas is also named as the consultant in charge of Mrs Birch’s case on the report of the angiogram and later when she is referred for an MRI scan on 21 June. However, as was correctly noted later that day, 21 June, Mr Kitchen was the consultant in charge of Mrs Birch’s case. Professor Thomas had no involvement but it appears he was incorrectly named as the consultant in charge because he was on the rota that weekend as the general consultant neurosurgeon on call.

Given the lack of recollection by the doctors and the state of the hospital records, it is necessary to reconstruct how the decision came to be made to perform the catheter angiogram. At the outset it needs to be said that the Queen Square neurosurgical team were entitled to take the decision themselves. When Professor Giovannoni was contacted on the Saturday evening, after Mrs Birch had suffered her stroke, the neurology specialist registrar at Queen Square, Dr Hu, could not tell him why she had an angiogram rather than an MRI. Dr Hu herself had not been asked by the neurosurgeons to assess Mrs Birch prior to it being performed. However, Professor Giovannoni said he would not have expected that, since Mrs Birch’s condition straddled the boundary of neurology and neurosurgery. Although not essential for him to have been contacted by the neurosurgeons he thought it would have been good etiquette if they had informed him that they had his patient. No one, even Mrs Birch’s own experts, thought that there was a need for the neurosurgeons at Queen Square to consult anyone else, given current medical practice in Britain.

To return to the hospital records there are the Friday night admission notes of the senior house officer already mentioned. The defendant’s expert, Mr Kirkpatrick, said that they are not especially impressive. In particular they describe Mrs Birch’s diabetes as “controlled” when the very opposite was the case, and there is the omission of the earlier episodes of double vision. Dr Hardie, one of the claimant’s experts, gave a clear account of how important it is to take an accurate history of a patient: this was not an accurate history. At that point both an MRI and catheter angiography were possibilities. Dr McEvoy and Mr Kitchen opine that it was likely that the senior house officer would have discussed this management plan, and the fact that it could wait until the Saturday morning, with the on-call neurosurgical specialist registrar that evening, although the specialist registrar would not have arranged a catheter angiogram investigation without discussion with a consultant. However, there is no record of any such discussion in the notes.

Prior to the catheter angiography, the only notes about Mrs Birch, subsequent to those of the senior house officer on the Friday night, were made by another senior house officer, Dr Young, who accompanied Dr McEvoy on his ward round on the Saturday morning. It is worth quoting them in full:

“WRSPR (McEvoy)

Admitted with 1/52 history headache

Right orbital headache

Ptosis since 4/7

GCS 15/15

BP 180/80

Pulse 70 regular

O/E pupil sparing right third nerve palsy

Plan for angiogram today ↓ discuss with radiographer and on-call radiologist for 11.30am today. Thank you. Dr Young SHO.

HB 14.3, WBC 7.5. pH 257, INR 0.95

APPT 28, NA 135, K 4.1. U 3.4 CR70”

Dr McEvoy attributes the brevity of the notes to the fact that there were some 90 patients on the ward to be seen over the four hours so the notes for each had to be relatively brief. Maybe, he says, they were written retrospectively by the junior doctor accompanying him on his rounds. In any event there are important omissions, such as the pain on which Mr Kitchen placed some reliance as a feature demanding that an aneurysm be excluded. So by, or at this point, the possibility of an MRI had been discarded and the decision made to conduct the angiogram.

Dr McEvoy’s evidence was that an angiogram is a significant procedure and he would not have gone ahead without the agreement of both the on-call consultant neurosurgeon, Mr Kitchen, and the on-call consultant neuroradiologist, Dr Brew. If the decision had not been taken the previous evening, he would have discussed the patient’s presentation, history and management with them. He would have asked the senior house officer to contact the specialist neuroradiologist and the radiographer on his behalf. In turn the specialist neuroradiologist would have discussed the management plan with the consultant neuroradiologist on call. As Dr McEvoy put it, there would have been a circle of telephone calls. However, Dr McEvoy and Mr Kitchen both accepted that it was a mistake that there was a lack of any record, not least of Mr Kitchen’s involvement. Dr McEvoy said that maybe that was because the telephone calls to arrange the procedure were made from theatre when the notes were not available. From Dr McEvoy’s view point Mr Kitchen’s word was final. Mr Kitchen was an international expert in neurovascular surgery, he was not. He, Dr McEvoy, would have passed on all relevant information, for example, that Mrs Birch’s face was puffed up although he now thought it may simply have been an asymmetry of her face. If he had thought Mrs Birch may have had cavernous sinus pathology it would have been recorded. All Mrs Birch’s factors would have been taken into account. Although there was no note of Mr Kitchen’s involvement he, Dr McEvoy, would not have proceeded with an angiogram without Mr Kitchen’s authority. An angiogram was a one hundred percent impossibility without Mr Kitchen’s involvement. It was a judgment call. Angiography was a huge undertaking and major operation, and it had to be sanctioned at the highest level. Dr Brew would have absolutely had to agree with it. Mr Kitchen would have been contacted on the telephone or might have come into the hospital.

Despite the paucity of evidence I find that Mr Kitchen was contacted and Mr Kitchen did approve the catheter angiography. There is some support for this in Dr Al-Jeroudi’s evidence that he thought that when he suggested an MRI Dr McEvoy said he had spoken to his consultant who had said more was needed. Mrs Birch was being treated as a vascular case and Mr Kitchen was the appropriate consultant. When the stroke was discovered it was Mr Kitchen who referred Mrs Birch for further examination. In any event Mr Kitchen accepted that he must have had Mrs Birch under his care that weekend. His evidence was that Dr McEvoy would have been familiar with his views and policies for dealing with a decision like this. Dr McEvoy would certainly have called him if he had been at all uncertain what to do. He could recognise the rationale which informed the management of Mrs Birch. The decision for angiography would have been made by the on-call neurosurgeons in consultation with the on-call neuroradiologist.

As to the actual decision to perform angiography, Dr McEvoy said that Mrs Birch was being referred to eliminate a problem she may not have had. But an aneurysm was life threatening and the possibility had to be excluded. A judgment had to be made as to which was the most dangerous of the diagnoses, and that was an aneurysm, for which an angiogram was most appropriate, rather than cavernous sinus pathology. His consultant, Mr Kitchen, would have agreed. The angiogram would not eliminate cavernous sinus pathology but it was necessary to rule out an aneurysm with 100 percent certainty.

For Mr Kitchen, the backdrop as he saw it was that there is a marked degree of collegiate management at Queen Square and a pooling of experience and knowledge. There was no written protocol at Queen Square as to the management of third nerve palsies, with or without pupil sparing. However, Mr Kitchen took the view that a painful presentation of a third nerve palsy demanded that the diagnosis of a cerebral aneurysm be excluded. The only way to have one hundred percent certainty that the painful pupil sparing third nerve palsy was not caused by an aneurysm was catheter angiography.

“One cannot afford to treat on the balance of probabilities, because of the risk of it being an aneurysm and the danger that if it is are both greater than the risks associated with angiography.” ”

That the third nerve palsy was pupil sparing in Mrs Birch’s case would not have largely reassured him that this was a more benign situation than an aneurysm. She had atypical features which were consistent with an aneurysm. Diabetics could have aneurysms.

“In 2003, it would always have been my practice to proceed to an angiogram without an MRI scan or MR angiography first with a painful [third nerve palsy] and when I suspected an aneurysm … even after taking her presentation, diabetic history and age into account.”

Mr Kitchen accepted that there could be an ascertainment bias in his unit, so that those like Mrs Birch could potentially be exposed to inappropriate procedures. Exclusion of an aneurysm fell within the field of neurosurgery. That was the task that had been set and in 2003 use of catheter angiography was the way to do this, with a greater degree of comfort than if MRI were to be used. It was a judgment call taking into account the advantages and disadvantages of the procedure for Mrs Birch.

So catheter angiography was performed. The radiology request form, signed by a Dr Young, the surgical senior house officer, contained the request for the angiogram. The clinical abstract identified Mrs Birch’s headache and right, third nerve palsy and continued: “[N]eed to exclude P Com aneurysm. [No blood] on CT scan.” There is no mention of the differential diagnosis, cavernous sinus pathology. First, however, Mrs Birch’s consent needed to be obtained. Since the decision had been made to perform a catheter angiogram, the evidence of Dr McEvoy was that it was not appropriate to discuss the alternative, less invasive option of an MRI with Mrs Birch, since it was not under consideration. Initially Mr Kitchen took that view, because MRI would not have been diagnostic of the posterior communicating artery aneurysm which urgently needed excluding. However, he said in evidence that someone should have discussed with Mrs Birch the different imaging methods, catheter angiography and MRI, their modalities and their risks.

At some point Mrs Birch signed the standard consent form. It names the procedure, cerebral angiogram. The standard form then reads “I have explained the procedure to the patient/parent. In particular, I have explained: ….” Nothing has been entered there or under the heading “The intended benefits”, although the entry under “Serious or frequently occurring risks” is: “[B]leeding. renal damage. 1% stroke.” Mr Brew explained that the figure 1% stroke” entry is always used. Taking away the extremes the figures generally range from 0.5% to 2%. In fact at Queen Square the risk is lower than 1%. However, Mr Brew said that the figure is not tailored for every patient, since it would be meaningless, given that it could simply not be said what the risk was for a 55 year old diabetic like Mrs Birch. So the standard 1% risk figure is given. The standard consent form continues: “I have also discussed what the procedure is likely to involve, the benefits and risks of any available alternative treatments [including no treatment] and any particular concerns of those involved.” It is signed by Dr Al-Jeroudi and then, a little later on the form Mrs Birch’s signature appears as giving her consent.

There is a disagreement in the evidence as to how the consent form was completed. Mrs Birch says that she does not remember Dr Al-Jeroudi explaining the procedure, although she recalls a nurse explaining the 1% stroke risk. She is certain that it was when she was taken to the theatre she met Mr Brew and Dr Al-Jeroudi for the first time and that neither of them had visited her on the ward. She remembers being asked to sign the form when she was on the trolley in the ante-room waiting to be admitted to theatre by a nurse called Wendy, who did not discuss the procedure further. It was the other patients on the ward who explained what would happen. She was not seen by the neuroradiologist on the ward. By contrast Dr Al-Jeroudi recollects obtaining the signature in this case. His practice was to arrive at the hospital and an hour and a half beforehand to visit the patient on the ward. Both Dr Al-Jeroudi and Mr Brew told me it was logical to obtain a patient’s consent that length of time in advance. That avoided the trouble and expense of medical staff coming in on a weekend, if the patient was approached for their consent at the last moment and did not give it. The only exception, said Mr Brew, was if there was an acute case which had to be done urgently, but that wasn’t the case here. Dr Al-Jeroudi told me that he would explain the procedure, the bleeding and in particular the risk of a stroke, and then ask whether there were any questions. He remembers going to the ward to see Mrs Birch. Mrs Birch remembers seeing him only in “greens” but he said the explanation for that was that he would come in on a motorbike and change immediately into his greens.

At the end of the day I agree with the submission of Mr Aldous QC for the defendant that it is almost certainly immaterial which version of the evidence on this point is accepted. In my view, however, Dr Al-Jeroudi’s version is to be preferred. There is no reason that the standard procedure should not have been followed on this occasion. The nursing transfer note to the angiogram room also suggests that Mrs Birch had signed the consent form on the ward. There is also Mrs Birch’s evidence that the nurse looking after her in the morning before the angiogram was a different nurse from the nurse looking after her in the afternoon after the procedure. Mrs Birch’s evidence was that the nurse in the afternoon was called Wendy. Therefore if both Mr and Mrs Birch are correct Mrs Birch cannot have been asked to sign the consent form by Wendy in the morning. That Mrs Birch’s recollection is mistaken should not be surprising. She had suddenly been admitted to Watford. Taken late on the Friday she found herself being transferred to Queen Square in central London and on the Saturday morning was told she may have an aneurysm. I am not in the least bit surprised that she said in her evidence that she was a bit bewildered and, of course, she had a bilious headache.

The crucial point, to which I return, is that in Dr Al-Jeroudi’s evidence there was no explanation of alternatives and Dr Al-Jeroudi did not claim to have provided them. On his evidence it was for the neurosurgeons to have discussed with Mrs Birch her history, condition and most likely aetiology, and to have obtained her consent to the angiogram. His role was only to advise her of the risks of the catheter angiogram procedure. Mrs Birch’s evidence was that she did take the attitude that “doctor knows best” but that if she had been given the choice of an MRI without a risk of stroke, and a cerebral angiogram with that risk, she would have chosen the former. Mrs Birch said in cross-examination that if Dr Al-Jeroudi had explained the two procedures, the MRI and the catheter angiogram, she would not have understood it. If Dr Al-Jeroudi had said that the catheter angiogram was best to exclude an aneurysm she would have taken the view that doctor knows best.

The catheter angiogram was performed just after mid-day on the Saturday. It was conducted by Mr Brew, a consultant neuroradiologist at Queen Square, and Dr Al-Jeroudi, the on-call locum specialist registrar there. Dr Al-Jeroudi, as the on-call specialist registrar, would have been first approached although no one can remember what exactly happened in Mrs Birch’s case. In any event Dr Al-Jeroudi’s view was that the clinical decision to proceed with a catheter angiogram was for the neurosurgeon to make. He was told that the neurosurgeons thought more was needed than an MRI. Dr Al-Jeroudi said he would have contacted Mr Brew, to pass on details and to inquire when it was convenient for Mr Brew to come in, but the neurosurgeons would also have contacted Mr Brew direct.

The normal procedure over the weekend, in Mr Brew’s evidence, was that the specialist registrar would contact him, having already been in contact with the neurosurgical consultant. A team would need to be assembled, including nurses additional to the on-call surgical nurses. Mr Brew himself lived near to the hospital so could arrive quickly, but on a weekend, as with the others, he would prefer not to have to attend. When Mr Brew arrived at the angio-suite he would discuss the case with the specialist registrar and examine the available imaging, here the CT scan from Watford. He would introduce himself to the patient and since he remembers Mrs Birch he almost certainly did that on this occasion. He does not remember any facial swelling. He would have run over the procedure to explain to Mrs Birch what would happen. In Mr Brew’s evidence there was next to no chance, as he put it, that Mrs Birch had cavernous sinus pathology. He had reviewed the case on a number of occasions subsequently and could still not account for what happened. Mr Brew frankly acknowledged that he had little recollection of Mrs Birch’s case and anything he did remember had now merged with his subsequent review of the case.

The catheter angiography procedure is described by Mr Brew as follows:

“The patient lies on the table and the staff scrubs. The patient is draped. The patient is lying on their back. The neuroradiologist advises that he is to go into the right groin. He palpates the artery. He applies local anaesthetic. He goes into the groin with a needle into the artery. The wire is put up in the artery. A sheath is placed into the iliac artery. The catheter is then put in. The wire then goes up to the aortic arch and is then taken out. It is then necessary to get into one of the carotid arteries and then the vertebral arteries. Once the catheter is in the desired position, a small volume of contrast is injected to determine the catheter position and to ensure that the artery has not been damaged. A roadmap to show the course of the artery that is to be catheterised may be obtained to aid correct placement of the catheter. Images of that artery are then obtained. The catheter is repositioned as above into any other artery that needs to be examined. At the end of the procedure, if the neuroradiologists are happy with the images, then the catheter is taken out. The sheath is pulled out and the neuroradiologist presses on the groin to stop the bleeding and the patient is transferred back to the ward.”

There is some uncertainty as to who performed the angiography procedure on Mrs Birch. Dr Al-Jeroudi recalls that he only went so far before Mr Brew took over but Mr Brew thinks that Dr Al-Jeroudi may have had full conduct of the procedure. In this case Mr Brew cannot remember how much of the procedure he performed, but was content to accept Dr Al-Jeroudi’s account. Certainly there was no difficulty in the procedure in Mrs Birch’s case. Nothing turns on who performed the angiogram. Mrs Birch remembers being somewhat upset by the atmosphere in the theatre but again nothing turns on that either. The procedure appeared to have gone well. Crucially, there was no aneurysm. Mr Brew said he is fanatical about reporting and immediately completes the necessary report electronically. In this case the report (again, incorrectly with Professor Thomas named as the consultant) sets out the clinical details, the procedure and that result (“No evidence of an aneurysm. No immediate complications”).

Mrs Birch suffered a stroke that afternoon. All the experts are agreed that it was caused by the catheter angiography, which causes the risk of a stroke by formation and dislodgment of thrombosis, a clot, which may then travel to the brain. Mrs Birch would have avoided a stroke had she been examined by MRI, followed by watchful conservative treatment. After Mrs Birch had the stroke she was given an MRI headscan. While it was designed to address the stroke it uncovered no explanation for Mrs Birch’s third nerve palsy, in particular cavernous sinus pathology. Mr Brew says that on this basis they would still have performed a catheter angiography to eliminate the possibility of an aneurysm. It is now agreed that all along Mrs Birch was suffering from what Professor Giovannoni diagnosed – diabetic vasculopathic ischemia of the third cranial nerve, causing third nerve palsy.

The effect of the stroke on Mrs Birch’s life has been traumatic. As described in her own words, it has “changed my whole life”. She is now classified as disabled, has a weakness down the left side of her body, lacks function in her left arm and hand, cannot carry, has difficulty with her personal care and dressing, and cannot do housework. She was a passionate gardener before the stroke; she has had to abandon that hobby. She played sport and danced, but this is no longer possible. She could not return to her secretarial job although fortunately she can still sit as a magistrate.

Mr Birch was on the ward with his wife when she had the stroke. He is an architect. Not unnaturally he was extremely concerned at what had happened. Over the following weeks he asked for meetings with those who had performed the angiogram to understand what had happened and why. Not that anything turns on it but I can understand Mr Birch’s frustration with the apparently cavalier way in which the hospital handled his perfectly reasonable requests. Eventually he met with Mr Brew. Mr Birch had prepared a list of questions, with the assistance of a family friend who is a general surgeon. He made a note of Mr Brew’s answers. I accept the submission on behalf of Mrs Birch that as a professional man her husband would have appreciated the importance of carefully recording what was said and that his note of this, and the meeting with Dr McEvoy, are likely to be true.

Mr Brew told Mr Birch that he was not part of the decision making team which had decided to perform the angiogram but that in order to remove an aneurysm from the equation an angiogram would be automatic. He was not aware of Mrs Birch’s medical records, or that she was diabetic, but that nationally strokes occurred in about one half of a percent to two percent of patients undergoing the procedure. He could not be sure of the risks regarding diabetics but they would be greater. Mr Brew’s evidence of this meeting was that he just happened to be on the ward in July when he was asked by Eleanor White, a senior nurse, to meet with Mr Birch. He thinks that what Mr Birch had written in the note represented what he would usually have said about the risk of stroke, that it ranged from 0.5 percent to 2 percent. He was unlikely to have said that performing an angiogram was an automatic procedure. Subsequently Mr Birch had a meeting with Dr McEvoy who said that in his opinion an angiogram was the only way to detect whether his wife’s problem was caused by an aneurysm. Had an MRI been carried out first and revealed nothing, an angiogram would then have been performed anyway. Time was always of the utmost importance and the lucky people were those in the hospital rather than those lying on the pavement.

There were five expert witnesses and in the course of the trial I was referred to some thirty-four articles or extracts from the medical literature. The three experts called on Mrs Birch’s behalf were:

The experts met and were able to give a joint statement covering a considerable number of issues. However, they remained divided on some crucial aspects of the case. In this regard the experts’ evidence reflected, in part, their practices and experiences. For example Mr Kirkpatrick is a neurosurgeon in an acute vascular unit, which is similar to Queen Square. Thus he sees patients, as he explained, who have already passed through a number of filters and who therefore have a higher chance of harbouring an aneurysm. As a neurosurgeon he would be less likely to see the type of patients Mr Choksey and Dr Hardie see, who are suffering from a medical palsy such as pupil third nerve palsy. On the negligence issue the expert evidence can be divided, for convenience under a number of heads. The evidence on consent is left until later. The starting point is that it was universally accepted that the Queen Square doctors needed to undertake a risk-benefit analysis of any treatment undertaken in Mrs Birch’s case.

All experts agreed with a statement in Dr Hardie’s report, that patients with diabetes are notoriously prone to lesions of both cranial and other peripheral nerves secondary to diabetic microangiography which interferes with the nutritional and vascular supply to the individual nerves. The so-called “rule of the pupil” is stated in various sources before the court. In what is apparently a standard textbook, J Patten, Neurological Differential Diagnosis, 2^nd ed, 1996, 55, it is said:

“A painful third nerve palsy with pupil sparing is nearly always due to diabetes. The presumed pathology is infarction of the nerve trunk, and hence in third nerve lesions due to these causes the pupil is usually spared. … The prognosis in all instances is for complete recovery of function to occur over some 4-12 weeks. If recovery does not occur the diagnosis must be reconsidered.”

There was some disagreement among the experts about the “nearly always”. In a more recent publication, a commonly held clinical extension of the rule of the pupil is said to be that “a neurologically isolated complete external dysfunction and pupil spared [third nerve palsy] in a patient with vasculopathic risk factors (e.g. over the age of 50 years, diabetes, hypertension) is usually due to an ischaemic infarct of the third nerve” (Lee et al, “The Evaluation of Isolated Third Nerve Palsy …” (2002) 47 Survey of Ophthalmology 137, 141). Overall the experts agreed with this approach.

Thus the rule of the pupil suggests that given the pupil sparing, and Mrs Birch’s diabetes, she was unlikely to have had an aneurysm, her condition being attributable to her diabetes. This accords with the study by Blake and others that pupil sparing, third nerve palsies in those with a history of diabetes mellitus were most likely attributable to the infarction of the extraaxial segment of the nerve. (Blake, et al, “MR of Oculomotor Nerve Palsy” AJNR, 16, Sep 1995, 1665, 1670). The matter is summed up in a contribution by Kasner and his co-authors.

“In older patients, the major confounding source of isolated third nerve palsy is microvascular infarct of the nerve, which characteristically produces ptosis and opthalmoparesis of the four relevant extraocular muscles but spares the pupil. This ischaemic cranial neuropathy is usually associated with other systemic vascular disease and is most common in patients with hypertension, diabetes mellitus, or smoking. Typically, patients develop acute diplopia, retro-orbital pain, and headache. Pain is a common accompaniment in about 60% of cases. The pupil is not involved in the majority (70% to 80%) of patients.” (Kasner et al, “Neuro-opthalmologic aspects of aneurysms” in R W Hurst (ed), Neuroimaging Clinics of North America: Cerebral Aneurysms, Neurosurgery, 1997, Vol 7, No 4, 687-689 (footnotes omitted)).

In ordinary circumstances the treatment in such cases is conservative: the patient is observed over a period to see if the problem resolves spontaneously.

Just how likely Mrs Birch was to have had an aneurysm was a matter of disagreement: her experts placing it as low as one or two percent, the defendant’s experts placing it higher, although Mr Kirkpatrick accepted that the fifteen percent figure he gave at an earlier point may have been too high. Overall the experts agreed that the chances of an aneurysm in Mrs Birch were low, but not negligible. All the experts agreed, however, that an aneurysm can present with a pupil sparing nerve palsy. Significantly they also went on to agree that an aneurysm needed to be excluded, and within 24 hours.

In examining Mrs Birch Professor Giovannoni identified cavernous sinus pathology as part of his differential diagnosis. The literature helpfully identifies the frequency of symptoms and physical abnormalities in patients with septic thrombosis of the cavernous sinus: 80 – 100 percent have cranial nerve palsies and ptosis; 50-80 percent have lethargy, headaches and periorbital swelling; but less than 20 percent have diplopia (Ebright et al, “Septic Thrombosis of the Cavernous Sinus”, Arch. Intern. Med, v. 161, Dec 2001). Partly on the basis of this all the experts agreed that the chances of Mrs Birch having cavernous sinus pathology were low, but not negligible. Diabetics were more prone to infection than non-diabetics. The relevant experts went on to agree that “periorbital swelling”, “right side face puffed up”, and the concomitant “nausea, vomiting and general lethargy”, while non-specific, were more consistent with infection than with an aneurysm. They also agreed that the chances of Mrs Birch’s condition being due to cavernous sinus pathology, though very small, were on the available evidence greater than its attribution to an expanding aneurysm.

All experts agreed that in 2003 catheter angiography could not be applied to every presenting condition in which an intracranial aneurysm was suspected. In the view of all, performing catheter angiography would not at that time conform to reasonable standards of skill and care as the necessary first choice investigation for patients with the claimant’s history and presentation. Moreover, according to reasonable standards of skill and care all experts agreed that the doctors investigating the third nerve palsy in Mrs Birch should have made a considered choice between catheter angiography and MRI.

There was considerable debate among the experts about the relative merits of catheter angiography compared with MRI in detecting aneurysms. The issue does not arise in relation to cavernous sinus pathology since catheter angiography cannot detect it. There was general acceptance that catheter angiography has a superior sensitivity in detecting aneurysms although disagreement as to how much. On the other side of the coin are the risks attached to its use compared with the minimal risks with MRI, which is non-invasive. There was reference by some of the experts to catheter angiography as the “gold standard”, and this was reflected in the literature. This issue is to be set against the background that the experts all agreed that by the standards of care and skill in 2003 doctors investigating third nerve palsy had to make a considered choice between the two.

One strand of the expert debate was directed to whether it was necessary to use catheter angiography in the case of a patient like Mrs Birch. The literature was trawled extensively on the claimant’s side in an attempt to establish that catheter angiography was unnecessary. The abstract of the study by Cullom et al, “Relative Pupillary Sparing Third Nerve Palsies” (1995) 15(5) J. Neuro-Opthalmology 136 reads: “The prevalence of aneurysm in patients with relative pupil-sparing [third nerve palsies] may be low enough to preclude the use of routine angiography in this condition.” However, the study itself acknowledges that the sample on which it was based was small. There was also selection bias since all patients presented as out-patients, not to an acute service. Working from the different diameters of aneurysms, Jacobson and Trobe conclude that MRI should be the definitive screening only in patients with a low likelihood of an aneurysm and a high likelihood of suffering a complication from catheter angiography (Am J Opthalmol., 1999, no 128, 94).

Some emphasis was placed on two studies from the late 1990s by authors based at Queen Square. The first, by authors including Mr Kitchen, the consultant neurosurgeon in this case, opens with the statement that technical advances in MRI were such that aneurysms as small as 3mm could now be detected. “Dynamic MRA should be considered the sequence of choice” (Jäger et al, “Contrast-enhanced MR Angiography of Intracranial Giant Aneurysms”, AJNR 21, Nov/Dec 2000, 1900). The second suggests that digital subtraction angiography could no longer be assumed to represent the reference standard with subarachroid haemorrhage. If MRA was positive, a case could be made for proceeding to surgery without digital subtraction angiography (Jäger et al, “MRA versus Digital Subtraction Angiography in Acute Subarachnoid Haemorrhage”, Neuroradiology, v. 42, 2000, 313). However, the study involved patients with acute subarachnoid haemorrhage and there was a limited number where MRA had not detected aneurysms less than 5mm. Moreover the article concludes:

“It must be emphasised that these conclusions are valid only for centres which perform MRA to a high standard, when the images are interpreted by experienced neuroradiologists” (326).

As for the defendant, reliance was placed on a commentary earlier this year by two members of the Department of Radiology at the well-known Mayo Clinic in the United States: Kaufman and Kallmes, “Diagnostic Cerebral Angiography”, AJR, 2008, v 190, 1435. The authors recall that cerebral angiography has been of inestimable value in the diagnosis of diseases of the central nervous system, such as vascular lesions like aneurysms. However, it notes that catheter angiography is an invasive procedure, albeit minimally so, involving the placement of catheters in important and sometimes sensitive arteries. In the light of CT (computerised tomography), MRI and their further advances – non invasive techniques – was cerebral angiography worth its inherent risk of complications? The authors conclude affirmatively, given the spatial and temporal resolution possible. CT and MRI were still not sufficient for many patients with diseases such as intracranial aneurysms.

“Clearly a modern CTA [computerised tomography angiography] or MRA examination will be more than adequate for many patients needs: for example, in imaging some small asymptomatic aneurysms that wouldn’t be treated or in unequivocally mild carotid atherosclerotic disease. But there remain common situations, for example in subarachnoid haemorrhage, where a 90% negative predictive value for aneurysm with CTA or MRS is not good enough … [E]ven in the year 2008, for many patients with potentially life-threatening neurovascular conditions, we practitioners – and those patients – must still accept these risks to gain the potentially life-saving diagnostic information provided by the high-quality cerebral angiogram.”

To similar effect is an editorial from late last year in the American Journal of Neuroradiology penned by Kallmes (of the Mayo Clinic) and others, contending that catheter angiography was to be preferred to CT angiography (“CTA”). Patients with a relatively low pretest probability for aneurysm, such as those with minor trauma or with relative contraindications to angiography such as those on warfarin, might be reasonably studied with CTA but the instances in which it may be reasonable to use it “are not usually present in a tertiary case setting” (Editorial, “Death by Non Diagnosis” v. 28, Nov-Dec 2007, 1835). Both the commentary and editorial are, as these designations suggest, more advocacy than scientific in character, and both have their limits (for example, the very subtitle to the editorial indicates it is concerned with patients with subarachnoid haemorrhage, not Mrs Birch’s case). But given the eminence of the signatories to both pieces they cannot be neglected. It is fair to add that the editorial by Kallmes et al provoked a letter defending CT angiography and concluding that the editorial was the last spasms of a dying, out-classed antiquated way of practice which is the diagnostic catheter angiogram (Agid, et al, “Life at the End of the Tunnel: Why Emergent CT Angiography Should Be Done for Patients with Acute Subarachnoid Haemorrhage”, AJNR, 27 March 2008, 1).

Another strand of the evidence concerned whether MRI would have detected an aneurysm in Mrs Birch. It will be recalled that Professor Giovannoni had planned for an MRI to investigate her condition. Since he was a consultant at Queen Square her case was that clearly he thought the defendant’s MRI was capable of the task. Mrs Birch’s expert, Dr Molyneux, said in evidence that unenhanced MRI would clearly have been sufficiently sensitive to rule out an aneurysm to the necessary degree of confidence, having regard to the fact that she had a pupil-sparing third nerve palsy, and given her diabetes and age. The defendant argued that it reasonably concluded that it did not have MRI facilities at Queen Square of sufficient sensitivity to exclude an aneurysm.

Again neither party was sparing in citing sources to advance its case. Favouring MRI, for example, is the piece by Kupersmith, et al, “Magnetic Resonance Angiography and Clinical Evaluation of Third Nerve Palsies and Posterior Communicating Aneurysms”, J Neurosurg, (2006), 105(2), 228. It showed that the probability of failing to diagnose an aneurysm of sufficient size to cause a third nerve palsy was below two percent. This included patients with pupil involvement, which made an aneurysm a much more likely cause. Only an aneurysm of 2mm was missed by MR angiography, and all aneurysms in this series causing a third nerve palsy were 4 mm or greater in size. On the other hand a systematic review of the literature over the ten years from 1988 found that MRI had a 94 percent sensitivity rate in detecting aneurysms larger than 3mm (White, et al, “Can Non-Invasive Imaging Accurately Depict Intracranial Aneurysms? A Systematic Review” Radiology, 2000, v 217, at 367). Both pieces are simply illustrative and other sources were referred to during the hearing.

The issue was not resolved during the trial. Indeed it seemed to become more polarised. Dr Molyneux concluded that the likelihood of an aneurysm large enough to cause third nerve palsy being missed on enhanced MRI was one in one thousand or less. He also opined that in 2003 MR angiography was equally sensitive with catheter angiography in detecting aneurysms 4 mm or greater. But to Dr McConachie the sensitivity of MRI to detect aneurysms of 4 mm or greater was 95 percent, less than 50 percent for those 3 mm or smaller. Thus the size of an aneurysm, as might be expected, is of relevance.

There was considerable debate among the experts as to the size of aneurysms causing pupil sparing third nerve palsy and whether MRI would have ruled out, with a sufficient degree of confidence, an aneurysmal cause for palsy. One aspect of this issue was the incidence of small aneurysms. The smallest aneurysm in a study of aneurysms causing third nerve palsies was 2mm, with the smallest maximum diameter being 5mm. Thus in one study no aneurysm smaller than 5mm in at least one dimension resulted in third nerve palsy in the absence of subarachnoid haemorrhage (McFadzean, et al, “Composterial Tomography Angiography in Isolated Third Nerve Palsies”, J Neurosurg., v. 88, 1999, 679, at 680). There is a Japanese study of 16 patients with unruptured cerebral aneurysms presenting with oculomotor nerve palsy, one patient having a relatively small aneurysm, 4 x 3mm (Yanaka et al, “Small Unruptured Cerebral Aneurysms Presenting with Oculomotor Nerve Palsy”, Neurosurgery v. 52/3, 2003, 553). The study is so small that it is not possible to derive further benefit from it apart from its recording of a small aneurysm. I accept the evidence of both Mr Kirkpatrick and Dr McConachie that in their practices they had encountered aneurysms causing third nerve palsies smaller than 4 mm although in both cases it was very rare.

There was no agreement on this. Mrs Birch’s experts assessed the risk of rupture at no more than one – two percent a day if she had had an aneurysm and this had been undetected. The defendant’s case was that the risk of rupture was higher and in any event one could not derive a daily rate from longer term experience. The need to canvass the literature and evidence to determine a rate is obviated by my conclusion that the key issue is that there was a threat of rupture with potentially catastrophic consequences.

One study, based on a meta-analysis of three published prospective studies, found that the risk of permanent neurological complications associated with cerebral angiography in patients with a subarachnoid haemorrhage, cerebral aneurysm and arteriovenous malformation was low, at 0.7 percent (Cloft et al, “Risk of Cerebral Angiography in Patients with Subarachnoid Haemorrhage, Cerebral Aneurysm, and Arteriovenous Malformation” Stroke v. 30, 1999, 317). A more recent article, studying 39 neurological complaints in 2899 procedures, found that age was a factor in terms of permanent complications. While all those over 55 years, Mrs Birch’s age, had a significantly greater risk of complications, those around that age were average in terms of risk (Willinsky, “Neurologic Complications of Cerebral Angiograpghy” Radiology, v. 227/2, 2003, 522). Finally to be mentioned is a study of 19,826 consecutive patients at the Mayo Clinic from 1981 to 2003 who underwent cerebral catheter angiography. It found neurological complications in 522 patients (2.63 percent), strokes with permanent disability in 27 (0.14 percent) and death in 12 (0.06 percent). The mean patient age was just under 53 years. Advancing patient age was significantly associated with higher rates of neurologic complications, but the effect of this variable was quite small. Atherosclerotic cerebrovascular disease was an indication of increased risk of neurologic complications. (Kaufmann, “Complications of Diagnostic cerebral Angiography” Radiology, v 243/3, June 2007, 812). The task in assessing these studies is eased because at the end of the hearing most of the experts seemed to agree that the risk of stroke could be taken as a range to be around one percent. They also all agreed that age was associated with an increased risk of stroke but that the evidence for diabetes per se was limited.

At their joint meeting all experts agreed that by reasonable standards of care and skill the doctors investigating the third nerve palsy of Mrs Birch were obliged to make a considered choice between catheter angiography and MRI. That choice, they agreed, had to be tailored to (a) the particular facts of a patient’s case; (b) the relative likelihood of the differential diagnoses in that patient’s case; (c) the respective risks, absolutely and to the particular patient, of the different modalities; and (d) the balance of risk versus benefit to the given patient, having regard to those facts and risks and the most likely diagnosis.

One of the claimant’s experts, Dr Hardie, attempted to systematise the risk-benefit analysis involved in the decision-making for Mrs Birch’s treatment by drawing up a number of pathways. His conclusion was that even if the overall risk of stroke or death from catheter angiography was one percent, the total number of strokes or deaths was almost 40 times greater than the alternative approach with non-invasive MRI as the investigation of first choice. Put another way, the risk associated with MRI was nearly 40 times less than that associated with catheter angiography. Mrs Birch’s other expert, Mr Choksey, gave evidence along the same lines: if 10,000 patients were taken presenting with Mrs Birch’s condition, 200 might have an aneurysm. If subjected to MRI 95 percent would be diagnosed as having an aneurysm and 5 percent would not. The pre-test probability of missing an aneurysm was then 10 in 10,000 or 1:1,000. Even if ten percent of aneurysms went on to rupture the overall risk of morbidity or mortality would be 1:10,000. This compared to a 1:100 risk of stroke with catheter angiography. Dr Molyneux approached the matter in a somewhat similar manner, combining the chance of there being an aneurysm, the risk of missing it and the risk of the missed aneurysm rupturing over the weekend to come to an overall risk of harm of one in ten thousand.

The defendant mounted a fierce attack on the reasoning, in particular on Dr Hardie’s pathways. The pathways, it submitted, were proffered at the very last moment, not based on any peer review, consensus document or proper discussion. There was no proper literature analysis. It was clearly a purely forensic exercise produced very late in the day and, in Mr Kirkpatrick’s evidence, would not get past a review panel publication. It attempted to use statistics which were wrong, speculative and unjustifiable. It did not even attempt to factor in the benefit of intervention if an aneurysm was found and treated. It bore no relation to the exercise that could be expected to be carried out in practice and if it had been written into the notes as a justification for not performing a catheter angiogram it would have been torn to shreds if Mrs Birch had died of a ruptured aneurysm. In my view, however helpful Dr Hardie’s evidence was as to the background, in particular the risks and benefits highlighted in his pathways, it has little bearing on the legal issue I needed to determine.

The primary submission on Mrs Birch’s behalf is that the decision to perform the catheter angiogram which lead to her stroke was negligent. There was no proper assessment of her history and condition and no proper risk benefit analysis of how her condition should be investigated. No reasonable body of doctors would have proceeded to catheter angiography, rather than MRI, and if they would have done so the decision in Mrs Birch’s case would be neither reasonable nor responsible.

There was no disagreement as to the relevant legal principles in relation to this aspect of the case. The Bolam test is axiomatic. As McNair J put it (Bolam v Friern Hospital Management Committee [1957] 1 WLR 583, 587):

“[A doctor] is not guilty of negligence if he has acted in accordance with a practice as accepted as proper by a responsible body of medical men skilled in that particular art … Putting it the other way round, a doctor is not negligent, if he is acting in accordance with such a practice, merely because there is a body of opinion that takes a contrary view.”

The gloss on this is that a court may still find that doctors are negligent, even though they have acted in accordance with a practice accepted as proper by a reasonable body of medical opinion, if it is not convinced that the body of opinion is reasonable or responsible. As Lord Browne-Wilkinson put it in Bolitho v City of Hackney Health Authority [1998] AC 232, 243C:

“[I]f, in a rare case, it can be demonstrated that the professional opinion is not capable of withstanding logical analysis, the judge is entitled to hold that the body of opinion is not reasonable or responsible.”

By suggesting that the court can depart from professional opinion in rare cases, Lord Browne-Wilkinson was indicating that such an opinion is not to be lightly set aside. The body of medical opinion must be incapable of withstanding logical analysis, in other words, cannot be logically supported at all. If there are different practices sanctioned by two bodies of medical opinion, both withstanding logical analysis, there is no basis for a finding of negligence against the doctor choosing one rather than the other. The matter may simply boil down to a different weighing of benefits and risks. If there is no failure to weigh the risks and benefits of each practice the Bolitho approach cannot be used to trump Bolam, even though the adherence to one body of medical opinion has led to the adverse outcome in the particular case. As Lord Browne-Wilkinson put it in Bolitho:

“[It] would be wrong to allow such assessment to deteriorate into seeking to persuade the judge to prefer one of two views both of which are capable of being logically supported” (at 243D).

Not only am I bound by this view but I conceive it to be eminently sensible: it would be folly for a judge with no training in medicine to conclude that one body of medical opinion should be preferred over another, when both are professionally sanctioned and both withstand logical attack.

In masterly fashion, Mr Badenoch QC branded as negligent the Queen Square decision to proceed to catheter angiography in Mrs Birch’s case. No reasonable body of clinicians that would have done what the defendant did, on the particular facts of her presentation and history. Had the notional ‘reasonable body’ properly assessed Mrs Birch, which the defendant did not, taken a full history, which the defendant did not, and undertaken an appropriate ‘risk benefit analysis’, which the defendant did not, it would not have elected to use catheter angiography in preference to MRI. The defendant was therefore in breach of duty in accordance with the test set out in Bolam. Alternatively, if there existed a body of professional opinion that would have done what the defendant did, on the facts it would have been neither reasonable nor responsible. The approach could not withstand logical analysis, so in terms of Bolitho it would not provide the benchmark by reference to which the defendant’s conduct fell to be assessed. In particular such body of professional opinion would not have directed its mind to the comparative risks and benefits and reached a conclusion which took adequate account of the balance of risks and benefits.

As a matter of fact Mr Badenoch QC’s submission was that there was never any proper assessment of Mrs Birch’s case. In terms of the evidence there could be no confidence that Mrs Birch’s history, as contained in the Watford notes which accompanied her to Queen Square, was considered at all. Important in that regard were her diabetes and relevant history such as her sixth nerve lesions in previous years. To any clinician who had applied his mind with adequate care to Mrs Birch’s presenting condition and history, it should and would have been obvious that by far the most likely diagnosis was diabetic ischaemia and that aneurysm and cavernous sinus pathology were unlikely. No reasonable body of neurologists or neurosurgeons would have supported carrying out an angiogram on the basis of Mrs Birch’s presenting condition at the time, assuming it was considered.

As for the two possibilities Professor Giovannoni had identified, he had requested an MRA, a request contained in the Watford notes which accompanied Mrs Birch and in the transfer referral. Yet Dr McEvoy said in evidence that he did not know that Professor Giovannoni had requested MRI, and in any event the view of Mr Kitchen, his consultant neurosurgeon, was to be preferred. So there was no proper risk-benefit analysis on the question of whether she should have catheter angiography or MRI. The Queen Square investigation of her condition should have been tailored to the risk. Given that Mrs Birch was known to be a diabetic, and thus more likely to develop complications, and that the chances of an aneurysm were low, catheter angiography had risks outweighing the benefits. Coupled with the low possibility of her having an aneurysm, MRI would have been satisfactory to exclude it to an acceptable degree. Catheter angiography was never justified. There was no logical reason why Mrs Birch would not have been adequately investigated and protected by MRI. MRI would have been sufficient to rule out with sufficient confidence both an intra-cranial aneurysm and cavernous sinus pathology, with safety assured by subsequent monitoring.

Instead of a systematic analysis of the various factors there was, in Mr Badenoch QC’s submission, the adoption of the neurosurgeons’ favoured investigation in cases where an aneurysm was the principal suspect, namely, catheter angiography. That failed the test of logical analysis. Catheter angiography was the routine procedure in such cases, the automatic response as Mr Birch recorded of what Mr Brew had told him shortly after the procedure had had its tragic consequences. What ran through Dr McEvoy’s evidence was that he simply adopted the standard approach, which was that there was a need to exclude an aneurysm with one hundred percent certainty and that catheter angiography was the gold standard in achieving this. He had told Mr Birch that an angiogram was the only way to detect the problem. As for Mr Kitchen, he could not remember Mrs Birch’s case. Although he said he weighed up all the factors there must be real doubt, in Mr Badenoch QC’s submission, as to whether he was told them all, not least because of the inadequacies of the Queen Square notes. Moreover, he focused on the need to exclude an aneurysm, and not on the cavernous sinus pathology which Professor Giovannoni had also identified. He adopted the approach of proceeding with catheter angiography when he suspected an aneurysm, even in Mrs Birch’ case while taking into account, as he claimed, her presentation, diabetes, history and age. Moreover, Mr Kitchen conceded that there might be ascertainment bias among neurosurgeons at Queen Square. In either event his approach defied logical analysis.

As for the neuroradiologists, they simply followed the lead of the neurosurgeons. In as much as during the trial Mr Brew gave an impression, contrary to the passive role described in his statement, that he was instrumental in the decision-making in weighing up risks and benefits, that was doubtful on the evidence. Mr Brew now said that the CT scan from Watford showed no opacity of the sphenoid and ethmoid sinuses, which seems to be the primary source of infection predisposing a person to septic thrombosis of the cavernous sinus, and thus he could be reasonable confident that she did not have cavernous sinus pathology. In Mr Badenoch QC’s submission, not only was this not supported by the defendant’s own expert, Dr McConachie, but it was all an ex post rationalisation by Mr Brew to explain why he had not investigated for cavernous sinus pathology. In fact he had simply sanctioned catheter angiography, on the unjustified basis that there were strong and pressing reasons to suspect an aneurysm. Again that could not withstand logical analysis.

Overall, Mr Badenoch QC’s submission was that the decision to subject Mrs Birch to catheter angiography could not withstand logical analysis. That was the case whether the decision was put on the basis that the risk of a stroke occurring was outweighed by the risk that an aneurysm was present, that it was small (and thus may avoid detection through MRI) and that it would rupture over the weekend, or on the basis, as Mr Kirkpatrick put it, that because Mrs Birch was in a tertiary referral institution like Queen Square, that in itself was a factor which made it necessary to exclude an aneurysm by catheter angiography. The first basis failed logical analysis, since it did not properly match the risks and benefits set out in Dr Hardie’s pathways, which simply put in systematic form what all doctors must follow when they undertake the risk benefit analysis all experts agreed was necessary. So too did the second basis fail logical analysis, because the inevitable outcome of indiscriminate catheter angiography in every case of painful pupil-sparing third nerve palsy was to subject 95 percent of patients to a completely unnecessary and invasive investigation. In either event the result was that the defendant had breached its duty to Mrs Birch and that breach had caused her damage.

The forensic exercise, as at least one of the experts rightly commented, is in a number of ways artificial. No more so than with the medical literature, which the experts had trawled for any references relevant to the matters before the court. Passages were taken from these contributions, often without the context, to support particular arguments being advanced. In my judgment it is necessary to approach this and the matters before me with some appreciation of what are the obvious realities. As far as the medical literature is concerned, I take the view that individual contributions to the literature cannot be taken as the absolute truth. Clearly that does not mean that the literature is to be set aside but it means that one must always have regard to the age of the study, the standing of the journal, its author and the institutions where the work was undertaken, the size and selection of any sample of patients forming the basis of the study, and how the conclusions accord with those in other studies. The second point, and it was touched on by Mr Kirkpatrick, is that practicing doctors obtain a great deal of their knowledge from practical experience, from their teachers and colleagues and from attending seminars and conferences. They will certainly not have read the thirty-four medical articles presented to me, or even most of them. Thirdly, and it is based on a point agreed by all the experts in the joint statement: the neurosurgical and neuroradiological teams on call at Queen Square on that fateful day for Mrs Birch in 2003 would have been expected to have a working knowledge of the current literature and learning concerning pupil sparing, third nerve palsy; its occurrence, incidence and likely aetiology in diabetics; the detection capabilities of the different investigation modalities; and the most appropriate in respect of the safest choice of modality of investigation. But the phrase I underline is “working knowledge”, for it would be quite unrealistic for the law to require more.

With that as background, let me turn to the issues. They begin with Professor Giovannoni’s plan for Mrs Birch’s treatment. What Professor Giovannoni had done, in examining her, was to identify three possible explanations for the conditions she presented, an ischaemic lesion of the third cranial nerve; cavernous sinus pathology of various types; and a posterior communicating artery aneurysm. The first was most likely, and with hindsight the true explanation. It could be expected to heal itself spontaneously with time. The second, cavernous sinus pathology, was unlikely, but possible, and if present could be ultimately fatal. The same consequence could flow with an aneurysm, which if untreated might expand and rupture. Given the possibilities of cavernous sinus pathology or an aneurysm, albeit both low, we are no longer in the territory of “wait and see”, the treatment which is the corollary of the “rule of the pupil”. The “rule of the pupil” had failed to exclude Mrs Birch from further investigation. Her presenting conditions were such that Professor Giovannoni had referred her to Queen Square. The issue became one of how to investigate her condition.

At Queen Square there was the decision to perform catheter angiography to rule out the possibility of an aneurysm. In so far as it is relevant I find that decision was made by Dr McEvoy in consultation with Mr Kitchen. The decision was made prior to the involvement of the neuroradiologists, which accords with Dr Al-Jeroudi’s recollection and is what Mr Brew says in his statement. I also find that Mrs Birch’s history and presentation were considered at Queen Square. As I have already remarked there are deficiencies in the Queen Square notes, not least the failure to record Mrs Birch’s poor diabetic control and her neuropathic history. There were, however, the Watford notes. I accept Mr Kitchen’s evidence that he would have ascertained that Mrs Birch was diabetic and her relevant history such as the retinopathy and her past sixth nerve lesions. Mr Kitchen is clearly an outstanding neurosurgeon. I found him to be straightforward in his evidence. He readily conceded to me that the Queen Square notes were inadequate and should, for example, have contained a reference to his own involvement in the decision-making. He also made concessions on other matters such as the ascertainment bias of neurosurgeons at Queen Square. So although he cannot remember Mrs Birch’s case I accept his reconstruction of the events surrounding the decision in her case.

The decision for catheter angiography in Mrs Birch’s case raises three immediate issues. First, it was to investigate for an aneurysm, but what of the cavernous sinus pathology? There was disagreement as to how Professor Giovannoni’s notes should be read, with their order (1) aneurysm and (2) cavernous sinus pathology. Professor Giovannoni said in evidence that he had not intended to indicate any order of the pathology which he considered most likely, although the Queen Square witnesses said that this was how they would read it, with a cerebral aneurysm as most likely. If the neurosurgeons in Queen Square were entitled to make their own investigating plan, as the experts all said they were, the order in which Professor Giovannoni expressed his intentions is not that crucial. In any event it was common ground between the parties that the possibility of a post communicating aneurysm needed to be excluded as a matter of urgency. Although there were other possible explanations for Mrs Birch’s complaint, in particular cavernous sinus pathology, which was more likely, it was not so potentially life threatening. As Mr Choksey, Mrs Birch’s own expert put it, “A painful third nerve palsy should be held to be due to an expanding posterior communicating artery aneurysm until proved otherwise.”

The second issue which arises relates to the agreement at the joint meeting of experts, that given the condition which Mrs Birch presented an aneurysm needed to be excluded urgently, within 24 hours. Moreover, the experts also all agreed that patients referred for a neurosurgical evaluation with a painful third nerve palsy should be considered likely to have a posterior communicating artery aneurysm until proved otherwise. It is said that I should overlook the latter answer. It is said that the question, insisted on by the defendants’ solicitors, was not limited to pupil sparing third nerve palsy, as was Mrs Birch’s case. Moreover, it is submitted that the question was a misquotation from one of the defendant’s medical references, which reads that an isolated third nerve palsy with pupillary involvement should be considered as housing a posterior communicating artery aneurysm until proven otherwise (Kasner at al, “Neuro-Opthalmologic Aspects of Aneurysms”, in R Hurst, Neuroimagery Clinics of North America, v. 7 (4), 1997, 688. Whatever be the provenance of the question the fact is that it was asked and answered. To suggest that I should somehow discount the answer lacks reality.

The third immediate issue concerns the availability of MRI at Queen Square, or at least MRI of a sufficient quality. The MRI available at Queen Square in 2003 was a Signa 1.5 T. The two expert neuroradiologists agreed that it was able to detect cavernous sinus pathology, but they disagreed on whether it was able to detect, with similar reliability to catheter angiography, aneurysms of 4 mm diameter or above. However, Mrs Birch’s expert, Dr Molyneux assumed, perhaps not surprisingly, that since Jäger and his co-workers had used contrast enhanced MRI for the research reported in the paper referred to earlier, it could also have been used for clinical purposes. Dr Molyneux was not able to take the matter further in evidence, since he had not undertaken any testing of the MRI available for clinical purposes at Queen Square. I accept the evidence from Mr Brew that high resolution type MRI was not available for clinical use at Queen Square in 2003 and would have involved a different team for research purposes.

Let me turn, then, to the Bolam issue: in deciding on catheter angiography in Mrs Birch’s case, were the Queen Square neurosurgeons acting in accordance with a practice accepted as proper by a responsible body of neurosurgeons? Undoubtedly they were in terms of actual practice, given the evidence of Mr Kirkpatrick and Dr McConachie that there were other large, and responsible units, which would have done the same. There is no way I can disregard that evidence, especially the evidence of a leading consultant neurosurgeon such as Mr Kirkpatrick:

“Having read the Particulars of Claim, I remain of the opinion that should Mrs Birch had presented to any other Neurosurgical Unit within the untied Kingdom the vast majority would have proceeded directly to digital subtraction angiography if she had fallen into Neurosurgical hands for the purpose of excluding an aneurysmal cause.”

Of course it is obvious both from the literature and the expert evidence presented on Mrs Birch’s behalf that there was no consensus within the profession in 2003 as to the better imaging method for diagnosing aneurysms of the type Mrs Birch might have had, angiography or MRI. It may be that there has been a trend since 2003 in favour of MRI and other non-invasive techniques as the diagnostic tool of choice, as illustrated by some of the literature mentioned earlier. But the plain fact is that there was a responsible body of medical opinion in 2003 which would have taken the Queen Square decision.

The issue thus becomes one of whether the Queen Square approach, and that of other units adopting it, can withstand logical analysis. My unhesitating answer is that it can. I have already found that Mr Kitchen did undertake a risk benefit assessment, despite the unfortunate absence of any note of his having done so. He focused on the aneurysm, rightly in view of the agreement of all the experts that an aneurysm needed to be excluded urgently, within 24 hours. In deciding on catheter angiography Mr Kitchen would have weighed the likelihood of Mrs Birch having an aneurysm against the dramatic improvement that could be provided if she had one, the improvement for her being so huge that it justified doing the slightly more sensitive test. This is the balancing which Dr McConachie identified and is embedded in neurosurgical practice. As Mr Kirkpatrick put it, there was a bias among neurosurgeons in favour of excluding an aneurysm because of the dramatic consequences of missing such a diagnosis. In that culture the risks of carrying out a catheter angiogram were not a deterring factor and even in recent years catheter angiography has been regarded as the best first-time method of urgently excluding aneurysms large enough to cause third nerve palsy. In my view this clinical judgment, exercised on a daily basis in neurosurgical wards up and down the country, clearly withstands logical attack.

As a corollary so, too, does the decision not to use MRI. MRI was the recommendation of Professor Giovannoni and Mrs Birch’s case is that it was the logical choice for investigation of an aneurysm in her circumstances. At the hearing it was accepted on all sides that Queen Square doctors were not bound by Professor Giovannoni’s approach but could start afresh. That Dr McEvoy may have been ignorant of Professor Giovannoni’s request is thus beside the point. In starting afresh as to the modalities of investigation there had to be a logically valid choice between catheter angiography and MRI. There was. The fact is that there was no convincing evidence before me to question Mr Brew’s assertion that he would not have been satisfied with the images from MRI, even though he accepted that unenhanced MRI had a 90-95 percent sensitivity. Dr Molyneux’s evidence about the capability of MRI was based on the premise that the quality of the imaging had to be good. As will be recalled his 98 percent figure for MRI at Queen Square was based on the mistaken assumption that the contrast enhanced MR Angiography used for the research paper by Jäger and others was also available for clinical use. That high quality MRI was available is also a premise in the experts’ joint statement. The choice between catheter angiography and MRI has also to be seen against the backdrop of disagreement in the literature as to the relative merits of these two imaging methods. In my judgment the decision to choose catheter angiography over MRI because of the Queen Square view that the MRI available was not adequate to provide a definite diagnosis stands up to logical analysis.

There is no dispute that the defendant had a duty to disclose the medical risks associated with the cerebral angiogram procedure. There is no dispute that Mrs Birch was informed of a one percent risk of stroke. She signed a form to that effect and although the evidence was in conflict about the circumstances in which that was done it has no bearing on the fact that she did so. However, Mrs Birch was not informed of the comparative risks associated with the alternative procedures, MRI as recommended in Professor Giovanni’s plan, and the cerebral angiogram which the defendant in fact performed. The crucial issue for me is whether the defendant was in breach of duty in its failure to do this. In general terms, my task is to determine what was the content of the duty to disclose in Mrs Birch’s case. The defendant contends that the obtaining of consent was the neuroradiologist’s responsibility, and that once the decision had been made to perform a catheter angiography all the neuroradiologist had to do on obtaining consent was to describe the technique and make Mrs Birch aware of the risks of the procedure. It was not appropriate at that stage to make her aware of the different imaging modalities available or discuss their strengths and weaknesses in investigating her condition.

In the defendant’s additional closing submissions, Mr Aldous QC helpfully cited the judgment of Lord Woolf MR in Pearce v United Bristol Healthcare NHS Trust [1999] PIQR P53 as to what the law demands as regards informing patients of medical risks.

“In a case where it is being alleged that a plaintiff has been deprived of the opportunity to make a proper decision as to what course he or she should take in relation to treatment, it seems to me to be the law, as indicated in the cases to which I have just referred, that if there is a significant risk which would affect the judgment of a reasonable patient, then in the normal course it is the responsibility of a doctor to inform the patient of that significant risk, if the information is needed so that the patient can determine for him or herself as to what course he or she should adopt” (at p59).

That statement of principle, he pointed out, was approved by Lord Steyn as part of the majority view in Chester v Afshar [2004] UKHL 41, [2005] 1 AC 134 at para 15. Lord Steyn added, in his own words, that generally speaking, in modern law medical paternalism no longer rules and “a patient has a prima facie right to be informed by a surgeon of a small, but well established, risk of serious injury as a result of surgery” (at para 16). The obvious rationale is patient autonomy and respect for the reality that it is the patient who must bear any consequences if a risk transforms itself into a reality.

I had asked for additional closing submissions on the issue of consent partly because I was troubled on reading Sidaway v Board of Governors of the Bethlehem Royal Hospital [1985] 1 AC 871 which Mr Aldous QC had referred me to during the course of the hearing. That decision appears to contain differences as to how a court is to assess whether risks are to be conveyed to the patient, even putting to one side the dissent of Lord Scarman. Lord Diplock’s approach, based firmly on Bolam [1957] 1 WLR 582, asks what a reasonable body of medical opinion would regard as proper to be disclosed (at 893). At least three of their Lordships added the gloss that in some circumstances the proposed treatment may involve a substantial risk of serious consequences which, notwithstanding reasonable medical opinion, must be disclosed to the patient. That seems consistent with Bolitho v City and Hackney Health Authority [1998] AC 232, in that notwithstanding what a reasonable body of clinicians would have done, a failure to inform of certain risks may not be reasonable, responsible or logical. In any event, in Pearce v United Bristol Healthcare NHS Trust, Lord Woolf MR was fully aware of the differences in Sidaway in articulating what he stated to be the law. Admittedly Chester v Afshar was primarily concerned with causation but, as indicated, Lord Woolf MR’s statement of the law was endorsed in the House of Lords. Given that Lord Woolf’s approach is advanced by the defendant as a current statement of the law perhaps any discussion of those legal nuances may be regarded as unduly pedantic. I can only plead that for this judge the matter is not as straightforward as it could be.

If patients must be informed of significant risks it is necessary to spell out what, in practice, that encompasses. In this case the defendant informed the patient of the probabilities, the one percent, and the nature of the harm of this risk becoming manifest, the stroke. But these were the objectively significant risks associated with the procedure which was performed, the catheter angiogram. Was it necessary for the defendant to go further and to inform Mrs Birch of comparative risk, how this risk compared with that associated with other imaging procedures, in particular MRI? No authority was cited to this effect but in my judgment there will be circumstances where consistently with Lord Woolf MR’s statement of the law in Pearce v United Bristol Healthcare NHS Trust the duty to inform a patient of the significant risks will not be discharged unless she is made aware that fewer, or no risks, are associated with another procedure. In other words, unless the patient is informed of the comparative risks of different procedures she will not be in a position to give her fully informed consent to one procedure rather than another. The difficulty is in delineating, in general terms, the circumstances in which the duty arises to inform of comparative risks. In my judgment, in the special circumstances of Mrs Birch’s case, that duty arose.

The defendant’s expert, Dr McConachie, a consultant neuroradiologist, said that once the decision had been taken that Mrs Birch should undergo a catheter angiography, Dr Al-Jeroudi’s role in obtaining consent was simply to describe the technique and make her aware of the risks, but that it was not necessary to discuss the comparative strengths and weaknesses of the different imaging techniques. Notwithstanding Dr McConachie’s obvious standing and expertise I do not accept that this goes far enough in the circumstances of Mrs Birch’s case. Moreover, it focuses on the role of the neuroradiologists to the exclusion of that of the neurosurgeons when both are involved in the treatment of the patient. Mrs Birch’s own consultant neuroradiologist expert, Dr Molyneux, opined that there was a failure to inform Mrs Birch and her husband that the risk of her harbouring an aneurysm as the cause of her symptoms was exceedingly low and also a failure to discuss that risk in the light of the increased risk of stroke which she ran from a catheter angiogram, given that she was a longstanding diabetic patient. The consultant neurologist, Professor Giovannoni, had taken the view that in the circumstances an MRI should be performed. The possibility of a pupil sparing third nerve palsy in a diabetic patient being due to an aneurysm is extremely low. “The failure of the neurosurgical staff to properly inform [Mrs Birch] and her husband of these facts, … and to properly explain properly the risks, … clearly in my opinion depart from the reasonable standard of care and a failure of properly informed consent to the procedure.”

The Queen Square consent form indicated that alternative treatments might be explained. Crucially, in my view, what Mrs Birch should have been told is contained in the evidence of Mr Kitchen, the defendant’s own neurosurgeon. I have already mentioned his international standing as a neurosurgeon. I have also referred to my favourable impression of Mr Kitchen’s evidence: he conceded mistakes at Queen Square, when they existed, for example, the standard of the hospital notes and admitted to neurosurgical bias. In the course of his evidence, Mr Kitchen was asked by Mr Badenoch QC about this issue of informing of comparative risk. He said that someone should have discussed with Mrs Birch the different imaging methods and their associated risks. Given the heavy reliance I have already placed on Mr Kitchen’s evidence in the resolution of this case it is difficult to gloss over so clear a view.

As a matter of law it is difficult to state in general terms when the duty to inform about comparative risk arises. Suffice to say that in my judgment, in the special circumstances of Mrs Birch’s case, Mr Kitchen is correct and there was a duty to discuss the comparative risks of the catheter angiography alongside MRI. There is no dispute that that duty was breached since the defendant concedes that comparative risks were not raised. The smallness of the risk associated with catheter angiography is irrelevant. The fact is that there was no risk of stroke at all from MRI. What should have occurred is that at Queen Square Mrs Birch should have been given a full and fair explanation of this and of the preference for catheter angiography.

Mrs Birch’s circumstances were unusual. She had been referred to Queen Square by one of its neurologists, but because of a problem with beds she entered a neurosurgical, rather than a neurology, ward. That neurologist, Professor Giovannoni, had specifically recommended an MRI. The Friday night admission notes identified both an MRI and catheter angiography as possibilities for the next day. It was only later that the decision for a catheter angiogram was made. Even then Dr Al-Jeroudi’s evidence was that he was suggesting an MRI. Given this background, where two procedures were open for Mrs Birch, she needed to have explained to her the comparative risks. While I do not accept the claimant’s submission that she needed to be informed explicitly what Professor Giovannoni had proposed, she did need to know than an MRI would have provided answers to the concerns Professor Giovannoni had raised about both an aneurysm and cavernous sinus pathology, and to the degree of confidence he thought necessary. The fact that Professor Giovannoni had recommended MRI may well have been revealed during the process. Mrs Birch also needed to know there was no risk associated with this mode of imaging. She needed to know that catheter angiography was more sensitive in detecting an aneurysm but could not detect cavernous sinus pathology so she would still need an MRI. She needed to know that while the risk of stroke with catheter angiography was low, there was a slight increase in risk since she was a poorly controlled, long standing diabetic, with a vascular history. Perhaps she should also have been reminded that the most likely cause of her troubles was a benign ischaemic lesion of the type she had suffered previously, and which had spontaneously resolved itself.

There was disagreement between Dr McConachie and Dr Molyneux as to whether, on the facts of Mrs Birch’s case she should have been informed of these matters. In the light of Mr Kitchen’s important evidence, I am convinced that in Mrs Birch’s case no reasonable, prudent medical practitioner would have failed to discuss the respective modalities and risks with her along the lines outlined. In their absence she was denied the opportunity to make an informed choice. Even if I am wrong on this, the failure to discuss with Mrs Birch these matters could not be described in law as reasonable, responsible or logical. On either approach, therefore, the failure to provide her with this information was in breach of duty.

Mrs Birch was not informed of the comparative risk of the different imaging procedures, along the lines Mr Kitchen indicated in his evidence and which I have held the special circumstances of Mrs Birch’s case required. To establish liability on the defendant’s part, Mrs Birch needs also to demonstrate, on the balance of probabilities, that had she been so informed she would have declined catheter angiography. For the defendant, Mr Aldous QC contended that Mrs Birch would have agreed, in any event, to a catheter angiogram. He invited me to draw that conclusion on the basis of Mrs Birch’s evidence: she took the view that “doctors know best”, and she said in her evidence that she would not know the difference between an MRI and a catheter angiogram, and if Dr Al-Jeroudi had explained that the latter was necessary to eliminate an aneurysm she would have agreed to it.

In my judgment that is too ready a conclusion to draw from Mrs Birch’s evidence. The issue is not whether she would have understood the nature of the procedures or what she would have in the face of a recommendation as to catheter angiography. The crucial issue is what she would have decided when given a dispassionate account along the lines indicated, in other words of the two procedures for imaging, and the advantages and disadvantages of each for someone with her background. In her evidence Mrs Birch said explicitly that if the comparative risks had been explained to her she would have chosen an MRI. I accept that evidence. Mrs Birch struck me as an intelligent and sensible individual, well able to have made that decision. She had also had diabetes for many years, which was poorly controlled, and had previous ischaemic lesions of the same nature in the immediate past, which had resolved spontaneously without the need for risk-carrying tests. It is clear to me that had she been given a fair and balanced account in the way I have held was necessary she would have rejected catheter angiography in favour of MRI. In other words, properly informed she would have declined the procedure leading to her stroke.

What happened in this case is tragic. Mrs Birch was correctly diagnosed by Professor Giovannoni. His plan was sensible. Things started to go wrong when, for reasons which are unclear, the consultant radiologist at Watford blocked an MRI for Mrs Birch. If this imaging method had been available at Watford that would have been the end of the matter. Because it was not available Professor Giovannoni then sought Mrs Birch’s transfer to his other hospital, Queen Square. Unfortunately, there was no bed available in the neurology wards there so that, in a phrase used by the defendant’s own expert, Mr Kirkpatrick, Mrs Birch fell into the hands of the neurosurgeons. They saw that Professor Giovannoni wanted an aneurysm excluded. It was a matter of urgency and for understandable reasons they focused on this. The priority became to eliminate the possibility of an aneurysm because of the catastrophic consequences which could follow if there was an aneurysm and it burst. The preferred method for doing this at Queen Square was by catheter angiography, an alternative and more sensitive imaging method to MRI. With hindsight Mrs Birch was subject to an unnecessary procedure which caused a stroke. But there was no negligence in the decision at Queen Square to perform catheter angiography rather than the alternative imaging method, MRI. Where Queen Square fell down was in failing to discuss with Mrs Birch, as their own leading neurosurgeon, Mr Kitchen, put it, the imaging methods and their comparative risks. For that reason the defendant is liable to her for breach of duty.