Jones v Conwy and Denbighshire NHS Trust

In this action Claire Louise Jones (“Claire”), claims damages against Conwy and Denbighshire NHS Trust (“the Trust”) in respect of her treatment at the Glan Clwyd Hospital on and after 13^th March 1995. At that time Claire was aged 12. She had a sinus infection which developed into orbital cellulitis with a subperiosteal abscess. This spread so that there was an intra-cranial collection of pus which necessitated a craniotomy and caused her to be epileptic. In essence she claims that the staff at the hospital failed to appreciate the seriousness of her condition and failed to carry out appropriate investigation in the form of a CT scan on the day of her admission. Had they done so the scan would, it is said, have indicated the need for immediate surgery which, if carried out that night would have avoided the spread of the infection to the brain, the need for a craniotomy, and the epilepsy.

Orbital infections, i.e. infections of the area of the eye, vary in form and seriousness. They occur most often in children. They are usually, as in this case, the consequence of sinus infection. In 1970 orbital complications of acute sinus infections were reclassified by Chandler et al as falling into five categories (Footnote: ¹). These categories are used by some, but by no means all, authors and doctors.

Category 1 is described by Chandler as Inflammatory Edema [i.e. swelling] of the eyelids. Its defining characteristic is that the swelling is preseptal i.e. located in front of the orbital septum. The septum is a fibrous layer within the eyelid which helps support the eye within the orbital cavity and provides a defence against infection extending behind it. The inflammation results from a sinus infection originating behind the septum. It is not itself an infection and probably results from impedence of the drainage of blood from the veins into the ethmoid vessels which are obstructed by pressure. Chandler records that there may be slight proptosis of the globe as a result of oedema of the orbital contents. Category 2 is Orbital Cellulitis. This is an inflammation of the orbital tissues, which have been infiltrated with inflammatory cells and bacteria, but without discreet abscess formation. It is usually caused by infection of the adjacent sinuses, often the ethmoid sinus. The ethmoid sinus lies between the superior half of the nasal cavity and the orbit. Category 3 is a Subperiosteal Abscess. This is a collection of pus between the periosteum and the bony wall of the orbit. The periosteum of the orbit is a fibrous tissue that lies along the surface of the bone of the orbit. Category 4 is an Orbital Abscess which is a discrete collection of pus within the orbital tissues. Category 5 is a Cavernous Sinus Thrombosis, a rare but very dangerous condition.

Three points are to be noted. Firstly, although the categories set out above represent a worsening of the patient’s condition, they should not be regarded as stages which necessarily follow one another in numerical progression (although orbital cellulitis is likely to follow inflammatory oedema). Secondly, inflammatory oedema resulting from sinus infection in Category 1 may be distinguished from preseptal inflammation or infection arising from sources in front of the septum such as a bee sting, trauma, or a stye. Thirdly, some authors reverse the order of seriousness as between categories 2 and 3 (or at least list them with 3 preceding 2).

Doctors also use the expression “periorbital cellulitis”, which may mean different things. Looked at literally the expression means no more than cellulitis, i.e. an acute diffuse spreading of infection of the subcutaneous tissue, which affects the regions around the orbital cavity. In this broad sense it can mean any infection which affects the area around the orbit and thus include preseptal or orbital (i.e. post septal) cellulitis. More restrictively it can be confined to preseptal cellulitis.

Inflammatory oedema or preseptal cellulitis is much less serious in its implications than the various forms of postseptal cellulitis. It is usually treated successfully with antibiotics. By contrast Chandler categories 2 – 5 have progressively more serious implications both for vision and for brain damage. Infection which spreads to the intracranial cavity, i.e. to the brain, is likely to produce brain damage and can cause death (Footnote: ²). Infection can spread to the intracranial cavity from an infection in the sinus and not via the orbit.

Evidence of the development of Claire’s illness is derived from two sources (a) the evidence of Claire’s mother, which I record in paragraphs 8 - 11; and (b) the medical notes and evidence of the examining and treating doctors and of the nurses.

On Saturday 11^th March 1995 Claire came to see her mother at work to get some money to go swimming that afternoon. She said that she had an ache behind her eyes, which her mother thought might be sinusitis. That evening, when she got home from swimming, she complained of pain behind her eyes. She described how, when she had dived into the pool and swum to the bottom, it had really hurt her; and that if she bent her head down it hurt. With hindsight she seemed to her mother to be a little off colour generally.

In the middle of the night Claire was sick several times. She complained of headache and pain behind the eyes, and, particularly, the right eye. She came into her mother’s bed for the rest of the night. In the morning her eye was badly swollen. Her mother’s description was that it seemed as though the whole eyeball was protruding out from the socket and looked as though she had a golf ball stuck in behind the eye.

On Sunday 13^th March Claire’s mother phoned her GP’s surgery. Dr Kakati, who was on call, came to see her and arrived at about midday. He prescribed the antibiotic flucloxacillin, which Claire’s mother then obtained from the nearest open chemist. Claire had her first tablet somewhere between 1230 and 1.00 p.m. She had no appetite and had eaten no breakfast. She was continuing to be sick (mostly of bile) and her mother encouraged her to drink fluids. The effect of her vomiting was that little of the antibiotic would have been ingested.

On Monday 13^th March the swelling was worse. On her mother’s description what had been golf ball size was now tennis ball size; it was impossible to distinguish her brow bone from her cheek bone; and the eye was totally closed and protruding still. Her mother telephoned the surgery again. Dr Kakati gave her a prescription for amoxycillin to be taken concurrently with what he had already prescribed. Claire took her first tablet from the new prescription at about 11.30. She did not improve. By late afternoon she was in such pain that she was crying and rolling around on the couch in the living room in agony. Her mother went to the GP and was given a prescription for pain killers which she obtained from the chemist. When she got home at about 5.20 pm she thought that Claire’s eyes had got still worse and decided to take her to Glan Clwyd Hospital.

Claire and her mother arrived at the hospital at about 7.40 p.m. The triage nurse recorded that she was complaining of photophobia. She recorded “PERL” i.e. pupils equal and reacting to light.

At 8.55 pm she was seen by a senior houseman. He recorded that she had been unwell for two days, having started with a swelling right eyelid, and was feeling generally unwell. He recorded the antibiotics prescribed by the GP. On examination her temperature was 37.7º C, i.e. slightly raised. He recorded her vision as “R: 6/9 L: 6/6”. This is a reference to the standard Snellen eye chart. “6/6” is treated as normal. “6/9” indicates that the person tested can see at six metres what a person with “normal” vision could see at nine. He also recorded in sequence “PEARL”, “Eye not red”; and “Extensive periorbital cellulitis ® eye”.

The reference to “Eye not red” was, as I infer, a reference to the eyeball and not the eyelids or surrounding tissue. I take that view firstly because in the light of the reference to “Extensive periorbital cellulitis” the eyelids and surrounding tissue would have been likely to be red. Secondly, the position of the reference in the notes indicates that the houseman noted the indicia from the eyeball before going on to record the periorbital cellulitis i.e. the condition of the lids and surrounding tissue. The houseman also wrote “? Refer paediatrics. Needs iv [i.e. intravenous] antibiotics”.

At 9.10 p.m. Claire was seen by the paediatric registrar – Dr Kelly. He recorded that she was complaining of a two day history of pain in the right eye, leading to swelling on Sunday, and that she was seen by the GP on Sunday, who prescribed floxapen, and again on Monday when amoxycillin was added. Since Sunday there had been an increase in the swelling and tenderness. She was vomiting 5 times a day and not eating. She had vomited bile on Sunday and black liquid on Monday. She had had a temperature since Sunday. She had “poor activity energy …Unable to open [eye] ® side since yesterday Headache – Bitemporal intermittent”. This description of poor activity/energy tallies with her mother’s description of her as lethargic. The registrar recorded “Photophobia º Drowsy º Alert + Weakness º Numbness hallucination º Behaviour º ”. (Footnote: ³)

The reference by the registrar to the absence of photophobia on examination contrasts with the complaint of photophobia recorded by the triage nurse. Claire would not have used the word “photophobia” to the triage nurse. She must have said something to the effect that the light hurt her eyes. The houseman makes no reference to photophobia on examination. The registrar states that it was absent. Both the triage nurse and the houseman must have shone a light in her eyes in order to note “PERL/PEARL”. It is likely that the registrar did so as well.

Mr Limb for the claimant submitted that, if there had been a complaint of photophobia which on examination turned out to be ill founded, the triage nurse, who would have been relatively senior, would be likely to have recorded that. Further, since the Trust has not called either the nurse, the SHO, or the paediatric registrar, any ambiguities or contradictory statements in the notes should not lead the Court to draw any inference inconsistent with Claire’s reported complaints or her mother’s evidence, unless they were obviously wrong.

I am not persuaded that it can be inferred from the fact that the nurse did not record anything more about photophobia than that Claire had complained of it that the nurse found that Claire had it. The function of the triage nurse is to determine the priority with which patients should be seen by others in the A & E department. Having noted the complaint and checked the reaction of Claire’s pupils it would not be surprising if he or she left further examination, and the recording of it, to the houseman. If the nurse had found significant photophobia he or she might be expected to have recorded it. If the houseman had found photophobia I would expect him to have included it in the notes of his examination. If the registrar had detected photophobia I would not expect him to record its absence. The absence of any confirmation of photophobia on examination in the notes of the nurse and the SHO and the registrar’s notation “Photophobia º” lead me to conclude that the likelihood is that, on examination, Claire did not show signs of significant photophobia.

The registrar recorded that Claire was bright and alert. Her hydration and nutrition were satisfactory and she had no meningism (stiffness of the neck). His notes then contain a notation “ENT” against which are an L and an R surrounded by a circle followed by the reference “TM” (i.e. tympanic membrane) indicated to be normal. This shows that he checked her ears. He then recorded a diagnosis of “® periorbital oedema”. His notes then contain a further notation of an L and an R, each surrounded by a circle, which may, and I think probably does, indicate that he checked the eye movements. I take that view because her eye movements probably were full (since they were found to be so the next morning) and because there is little else that the note is likely to refer to. The notes then contain details of her haemoglobin, white cell count and platelet readings. The white cell count (18.1: 86% Neutrophils) was raised. There are then details of her sodium potassium, chloride, urea and creatine levels which were within normal limits.

Dr Kelly consulted Dr Cameron, the paediatric consultant by telephone. He summarised her condition and indicated that the issue was whether there was anything to suggest intracranial infection. Dr Cameron asked whether there was photophobia or neck stiffness (which there was not). Dr Kelly said that she was bright and alert and that he felt she had periorbital oedema due to infection. He felt that her visual acuity was satisfactory.

Claire is recorded as having been admitted to the ward at 9.30 p.m. and her condition on arrival is recorded as:

“R eye – red and oedematous swollen

In some pain”

Her temperature was 38.3º C; her blood pressure and pulse were normal.

I do not regard the reference to “R eye – red and oedematous swollen” as inconsistent with the houseman’s “Eye not red”, but as referring to the eyelids and tissue around the eye. It is unlikely that the houseman, whose notes indicate a careful approach, would have noted that the eye itself (i.e. the globe) was not red when it was. The nurse would have seen the surroundings of the orbit as red and swollen. The globe cannot itself swell.

Intravenous Flucloxacillin and Ampicillin were prescribed and first administered at 2200.

It is now agreed between the experts that on admission Claire was at least at Chandler Stage 2, and that without CT scanning further classification could not be decided on clinical grounds. The manner in which Claire was treated i.e. by intravenous antibiotics, assumed that she had orbital cellulitis. Dr Cameron, together with Dr Sunderland, the expert paediatrician, and Mr Swift the expert ENT surgeon called for the trust, were at pains to point out that, in a case such as this, you would assume a post septal infection, as being worse than a pre-septal one, and treat accordingly.

The night nurse recorded in the morning that Claire had been given paracetamol for headache and pyrexia and that she was also complaining of her eye being painful. Claire was noted to be feeling nauseated and to have vomited after the paracetamol for headache that morning. Her eye was recorded as less red and swollen, which, again I think is unlikely to be a reference to the globe itself. The note also records that she was a very quiet girl and unwell. She continued to complain of headache.

At the ward round, attended by Dr Cameron, Claire was recorded as suffering from orbital cellulitis. Her temperature was 37.8 º C. Her eye movements were full. So were her neck and throat movements. She had blurred vision in her right eye, which is not uncommon if the tissue surrounding the eye is puffy. Dr Cameron advised continuation of the intravenous antibiotics and obtaining an ophthalmic opinion. Save for continuing antibiotics no further treatment or investigation took place that day.

The nursing notes for the day record that Claire was “vomiting +++” with any food. She was put on iced water until the vomiting stopped. She continued to complain of headache throughout the afternoon. Calpol was repeated during the evening. She had vomited all fluids and was described as very lethargic and irritable.

The first medical note records that the right periorbital fullness and inflammation (erythema) were not resolving. She had head aches and vomiting but no neck stiffness. The registrar suggested an urgent CT cranial scan to rule out cavernous sinus thrombosis and that she needed urgent ENT involvement and urgent decompression. The antibiotics were changed to Ceftazidine and Flucloxacillin. Blood cultures were requested. An urgent cranial CT scan was fixed for that morning. The ophthalmologist rang and said he was in his clinic in the morning and would ring back.

At 1015 Dr Cameron recorded that Claire had continued to vomit and had frontal headache and persistent pyrexia. She remained orientated and lucid with no neck stiffness. Her hearing was OK and she was able to see with her right eye - “somewhat “blurry””. The discs were not swollen. She was able to move her eyes up and down and side to side but had limited movement up and down. Her pupils were reactive. Dr Cameron approved an urgent CT scan; urgent ophthalmic examination on the ward; the change of antibiotics and repeat blood cultures.

The CT scan was carried out at about 1.00 pm and as reported and entered in the notes:

““Extensive orbital cellulitis is demonstrated, this appears to extend into the conus bilaterally and there is a small abscess on the medial margin of the orbit. No intracranial complications are identified.”

This is not an entirely accurate summary of the scan. Dr Forbes, a consultant radiologist who gave evidence as an expert, explained that the CT scan showed in the right eye, inter alia:

Item (a) is a description of the swelling around the eye that would have been visible to anyone examining Claire. Item (c) is evidence of the underlying sinus infection.

Item (b) is evidence of a degree of proptosis. It is possible to see from the CT scan that there is an asymmetry in the positioning of the globes. The right eye is displaced slightly forward and more significantly to the right side. The displacement is also apparent from the fact that the lens of the right eye is not visible nor is the slight beak formation of the anterior of the eye, both of which are visible on the CT scan in the left eye. In his report to the Court Dr Forbes says that the CT brain scans show “a well defined collection of pus in the supero-medial aspect of the right orbit displacing the eyeball anteriorly (proptosis)”. As is apparent from para 30 above the scan was reported contemporaneously as showing “a small abscess on the medial margin of the orbit.”

Most significantly the post contrast images through the intracranial compartment show a small low density collection (of pus) in the subdural space over the right frontal pole extending into the anterior interhemispheric fissure. In effect the infection has spread through the veins to the intracranial cavity. This complication was not noted on the history sheet.

The first ophthalmological examination took place in the afternoon of the 15th. The ophthalmologist recorded, inter alia “Right proptosis with extraocular movement [i.e. the movement of the muscle of the eye] limited”. This is the first note of proptosis in the records. Pupillary reaction was normal. There was marked tenderness of the right frontal sinus. A diagnosis of orbital cellulitis secondary to sinusitis was made. An ENT opinion was urgently required. He also recommended adding Metronidazole to her medication.

The first ENT examination then took place. Among matters noted were severe swelling over the right upper eyelid, fullness of the frontonasal angles, severe chemosis (swelling of the conjunctiva), limited movement of the right eye upwards both medially and laterally and blurred vision. The notes indicate “?? subperiosteal abscess”. After discussion with Mr Osborne, the consultant ENT surgeon, it was recorded that the abscess needed drainage under general anaesthetic.

The operation took place that evening. Mr Osborne was the surgeon. The operation notes record:

“® fronto-ethmoidectomy incision.

Pus ++ located between the orbital roof and the orbital periosteum [i.e. a subperiosteal abscess]. It had dissected posteriorly for some distance and had a distinctive smell - ? anaerobic”.

There was no pus in the frontal sinus, which was opened, but there was pus within the right maxillary sinus. A drain was left in the right maxillary sinus, which was to be irrigated regularly with saline solution.

The nursing notes record that Claire was not fully conscious and was sleeping a lot (this note appears to relate to the early morning after 0500). A later note records that she became vague and disoriented at 8.00 a.m. By 9.05 she was fully conscious. At an ophthalmic review she showed no signs of optic nerve compression. A further CT scan was recommended and a review in the eye clinic when she was more ambulant. At an ENT review she was recorded as still drowsy but responsive to vocal commands. She was apyrexial. She was to continue on intravenous antibiotics.

Claire was drowsy but responsive to verbal commands. Her temperature was 39.5 º C. Antibiotics were continued. She was then reviewed by Dr Cameron who noted that she was still very poorly and had a spiking temperature that morning. The swelling around the eye had increased a little that morning. She was unable to open her eye at all for examination. There was no neck stiffness. She was orientated and lucid but tending to lie quietly. The antibiotic regime (Ceftazidime, Flucloxacillin and Metronizadole) was to continue. She had pus oozing from the eyelid and from the orbital tube. Dr Cameron planned a repeat CT scan because he presumed that a collection of pus was still present. In discussion with the microbiologist it was agreed that the antibiotics were appropriate at the time but might change on the basis of the blood cultures.

At 1130 the eye was reviewed by Mr Osborne. There was an obvious collection of pus. After discussion with Dr Cameron it was agreed that she should return to theatre for right orbital decompression and that there would be a post operative CT scan.

At 1230 she returned to theatre where the old wound was reopened. “Pus ++” was recorded which was curetted out. A post operative CT scan showed a significant increase in the size of the subdural collection with evidence of focal and generalised brain swelling. After discussion with Mr Osborne Dr Cameron arranged for the CT images to be transmitted to the Walton Centre, a neurological centre of excellence. He explained the position to the neurosurgical registrar at Walton, who rang back to suggest continuing with the antibiotics and rescanning the next day. The pus culture had by now shown the infection to be Group C streptococcus and bacteroides, a gram negative anaerobic bacillus which is a rare germ generally associated with dental abscesses. The antibiotics were changed to Benzyl penicillin and Metronizadole. The latter is a specific treatment for bacteroides.

Claire had her first convulsion that evening. This was diagnosed as a fit secondary to extradural abscess. She was given a loading dose of phenobarbitone, an anti-epileptic.

A CT scan carried out in the evening showed that the collection was double the size of that shown the previous day.

On 19^th March Claire was transferred to the Walton Centre. After a further CT scan showed a further increase in size and extent of the subdural collection she underwent a right frontal craniotomy and evacuation of the subdural empyema.

Claire was transferred back to Glan Clwyd. She was allowed home on 3^rd April 1995; although she was readmitted on 16^th April 2005 and discharged again on 19^th April.

It is the joint view of the expert neurologists for the parties that the critical point determining whether epilepsy would complicate Claire’s illness would have been the point at which abscess formation (the accumulation of pus) followed on from preceding meningeal infection. Their view on the balance of probability is that meningeal irritation would have been present on first admission; that it is unlikely that pus would have formed intracranially by midnight on 13^th March; but that by 10.00 a.m. there was likely to have been the beginnings an accumulation of pus intracranially. They believe that epilepsy would not have followed surgical drainage at midnight on 13^th March but would have followed drainage at 1000 on 14^th March. They agree that if drainage had occurred at 1000 on the 14^th March Claire’s epilepsy would have been less severe than is currently the case but she would still have required life long anti-epileptic drug treatment and that, because of her psychological vulnerability, any reduction of seizure frequency would not have had any impact on her quality of life.

In a subsequent letter Professor Chadwick expressed the view that it would be “entirely wrong” to take a cut-off of midnight as an absolute cut off for the avoidance of epilepsy. He thought that he and Dr Schady would agree that the risk for epilepsy was increasing in a fairly linear way with the passage of time and that by 10 a.m. on 14^th the balance of probability would have been that epilepsy would not have been avoided. But he did not think that either he or Dr Schady would likely to be tied down to a position of saying that it was at 10 a.m. precisely that the balance of probability rested at 51% or that at midnight on 13^th it was only at 49%. I infer that Dr Schady is of the same opinion.

That report gives rise to the following questions: (i) what was there to be drained between midnight on 13^th March and 1000 on 14^th March?; and (ii) what would the position have been if drainage had taken place at, say, 1.00 am on March 14^th ? I consider these questions below.

The classic test for the standard of care required of a doctor is that:

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

Per McNair J in Bolam v Friern Hospital Management Committee [1957] 1 WLR 583, 587.

That test has been refined by the House of Lords in Bolitho v City and Hackney HA [1998] AC 232, where Lord Browne-Wilkinson observed:

“ In my view , the court is not bound to hold that a defendant doctor escapes liability for negligent treatment or diagnosis just because he leads evidence from a number of medical experts who are genuinely of the opinion that the defendant’s treatment or diagnosis accorded with sound medical practice. In the Bolam case itself, McNair J. …stated that the defendant had to have acted in accordance with the practice accepted as proper by a ‘responsible body of men.’ Later… he referred to ‘a standard of practice recognised as proper by a competent reasonable body of opinion.’ Again, in the passage which I have cited from Maynard’s case [1984] 1 WLR 634 …Lord Scarman refers to a ‘respectable’ body of professional opinion. The use of these adjectives – responsible, reasonable and respectable – all show that the court has to be satisfied that the exponents of the body of opinion relied upon can demonstrate that such opinion has a logical basis.

‘In particular in cases involving, as they so often do, the weighing of risks against benefits, the judge before accepting a body of opinion as being responsible, reasonable or respectable, will need to be satisfied that, in forming their views, the experts have directed their minds to the question of comparative risks and benefits and have reached a defensible conclusion on the matter….”

“…In particular, where there are questions of assessment of the relative risks and benefits of adopting a particular medical practice, a reasonable view necessarily presupposes that the relative risks and benefits have been weighted by the experts in forming their opinion. But if, 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.

I emphasise that in my view it will very seldom be right for a judge to reach the conclusion that views genuinely held by a competent medical expert are unreasonable. The assessment of medical risks and benefits is a matter of clinical judgment which a judge would not normally be able to make without expert evidence. As the quotation from Lord Scarman makes clear, 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. It is only where a judge can be satisfied that the body of expert opinion cannot be logically supported at all that such opinion will not provide the bench mark by reference to which the defendant's conduct falls to be assessed.”

I am, of course, concerned with the position as it was in 1995.

The claim originally included the contention that Claire was not given the right antibiotics. That is no longer pursued. The critical issues now are:

I have had the benefit of the reports of Professor D Chadwick and Dr W Schady, consultant neurologists and the minutes of their joint meeting which produced the agreement that I have summarised at paragraph 46 above. I have also read the report and heard the evidence of Dr W St C Forbes, a Consultant Neuroradiologist and had the benefit of his oral explanation of what the CT scan shows.

I have also had evidence from the following experts, all of consultant status, in the following disciplines:

The claimant’s experts (Dr Conway, Mr Clearkin and Mr Hughes) are of the view that a scan on admission on 13^th March was necessary either on the basis that in a case of suspected orbital cellulitis such a scan is mandatory, or, if not always mandatory, was required in respect of Claire, having regard to the symptoms with which she presented.

The proponents of this view in respect of someone in Claire’s position, including but not limited to Dr Conway, Mr Clearkin and Mr Hughes, rely upon a number of factors. Firstly it is important to distinguish between pre-septal and post septal cellulitis (in its various forms), which is a medical emergency. But it may be difficult to make that distinction on medical grounds alone, especially in the case of a child, who may be unwilling, or unable, to cooperate or may be unresponsive to instruction. Because the consequences of orbital cellulitis in its various forms are potentially very severe (loss of vision, brain damage, even death), in all cases other than those where it is clear that there is a preseptal source of inflammation, it should be assumed that orbital cellulitis is or may be present, which may develop into Chandler Stages 3 -5. It is not possible, by clinical examination alone, to reach a reliable view as to the stage of the disease, where it is, and whether surgery is necessary. A normal set of eye readings does not automatically mean that no abscess is present since, as Mr Clearkin explained, the abscess will have to have grown to a certain size in order to limit eye movement. By contrast a CT scan is an efficient diagnostic technique which will give valuable information as to the site of the disease and its stage and provide a baseline from which to determine treatment and, if necessary, to plan surgery; and from which to observe the progress of the disease. For that reason patients presenting with symptoms or signs of orbital inflammation should be given a CT scan without delay. Delay cannot be justified on the basis that many patients improve on antibiotics alone. This was particularly the case for Claire. She did not present with nothing more than a nasty inflamed eyelid but had a history of being unwell, vomiting, with notable swelling around the eye, pain and an increased white cell count.

To the proponents of this view it is not acceptable, and illogical, only to use CT scanning as a guide to surgery if a decision to initiate or contemplate surgery has already been reached as a result of clinical examination. To do so is to deprive the treating doctors of the information that they need in order to make a proper decision on surgery until such time as, on inadequate material, they have decided on, or at least contemplated, surgery; and thereby to run the risk that by then it will be too late. It is true that CT scanning, which involves radiation of the eye of a child, is not wholly without risk (of future cataracts in the long term or, conceivably of cancer – a risk of the order of 1 in 2000 ); and that surgery, if decided upon, itself involves risk and scarring. But the benefit of scanning is overwhelmingly greater than the risk of not scanning since, in its absence, serious and avoidable damage may occur. The risk from CT scanning is, in any event, low. Dr Forbes’ evidence was that all X rays carry radiation risks; and that the radiation from CT scans is now about 100 times greater than that of an ordinary chest ray; and the radiation doses were quite significantly greater in 1995 than today. But clinical urgency outweighs risk. He would regard the urgency as being dependent on what the referring clinician finds and, as a radiologist, he would accept the latter’s judgment on urgency at face value.

There is much literature which to a greater or lesser extent supports the propositions set out in paragraph 56. Many papers speak of the value of CT scanning which may be regarded as established. Not all of the recommendations are in mandatory terms, although some are, e.g.:

“[CT] is mandatory when signs of orbital infection are present” (Footnote: ⁴)

and

“If any degree of displacement of the globe, opthalmoplegia or visual impairment is present an urgent CT scan is mandatory”.

Some of the literature speaks, for instance, of a CT scan being:

“preferable .. in all cases of orbital cellulitis” (Footnote: ⁵) or as

“indicated in children with proptosis (Footnote: ⁶)”.

Some of the literature is unspecific as to immediacy, although the fact that CT scanning is usually dealt with in consideration of the investigation stage indicates that it should be done at an early stage. There do not appear to be any references to radiation risks or to the risk, to which Dr Sunderland referred, of too early a scan giving a false reassurance.

The Trust’s experts take a different view. In their view CT scanning on admission is not indicated for all cases of orbital cellulitis nor was it indicated in this particular case.

The basis upon which they take that view is this. In most cases of orbital cellulitis a CT scan will not be needed because most such cases resolve themselves by the use of intravenous antibiotics. A reasonable course is to treat the patient with such antibiotics and see whether or not that occurs. One should only responsibly subject a child to large doses of radiation if that is likely to affect the management of the disease. In this case the immediate treatment was going to be the same whatever the findings of the CT scan, namely to continue with the antibiotics to see if they worked. No CT scan was called for unless it was apparent that antibiotics would not work, either because there was no improvement or because there was a deterioration in the patient’s position. If surgery was contemplated a CT scan would have been taken as an adjunct to surgery. The experts also relied on their own personal practice and experience, in the case of Dr Sunderland at Birmingham Children’s Hospital, where he has been Consultant Paediatrician since 1984, and in the case of Mr Swift at University Hospital Aintree. In each case they indicated that they would have taken similar steps to those taken at Glan Clwyd.

There are, and were in 1995, a number of proponents of the view put forward by the Trust’s experts, as Mr Clearkin acknowledged. Examples may be found in the following articles (the last two of which post-date 1995):

The proper approach to the investigation and treatment of a patient cannot be determined without reference to the particular combination of characteristics with which the patient presents. The significance of those features cannot necessarily be determined by seeing how many features are or are not those shown on a textbook check list. With that caveat it is of some interest to note a table describing the distinctive features of Preseptal and orbital cellulitis in a 1992 Textbook of Paediatric Infectious Disease by Feigin and Cherry.

That table (Table 86.1.), which is in the clinical presentation section, is as follows

In one respect at least this table should be looked at with circumspection. It indicates that in the case of preseptal cellulitis proptosis will be absent or mild. However the body of the text records that “Proptosis, decreased motility or pain on eye movement, does not occur in preseptal cellulitis (see Table 86-1)”. Whilst I cannot be sure, I suspect that in that passage the authors had in mind infection arising from a pre septal source (e.g. a sting), whereas in the Table they may have been contemplating Chandler Stage 1, which, as he himself indicates, may involve a slight degree of proptosis as a result of oedema of the orbital contents (Footnote: ¹²).

One of the most significant features on presentation was that Claire’s ophthalmologic parameters were normal. The importance of this in a disease which can lead to visual impairment or blindness is obvious. The SHO checked her visual acuity. The difference between the normal reading in the left, and the slightly below normal reading in the right, was not a cause for concern. Her pupils were equal and reacting to light, which signified an absence of impairment of the optic nerve. Her eye was not red. This is an indication of the absence of chemosis or significant proptosis. On 14^th March her eye movements were recorded as full and they are likely to have been in that condition the day before, and, as I infer, they were probably specifically considered by the registrar. These assessments were not carried out by ophthalmologists but I do not regard them as devalued on that account. Mr Swift, whose evidence I found convincing, regarded the presence of full eye movement as a very important clinical sign, suggesting (I accept not conclusively) an absence of formation of pus.

I turn therefore to consider the features which were relied on by Mr Hughes, Mr Clearkin and Dr Conway between them (not all of them relied on all the features) as indicating that Claire had a particularly bad case of orbital cellulitis. Those features were: (i) fever; (ii) high neurophil count; (iii) bad swelling of the eye; (iv) proptosis; (v) drowsiness; (vi) vomiting and (vii) the duration of her illness. She was summed up, in general terms, as a sick or very sick child.

So far as fever is concerned Claire’s temperature was recorded on 13^th as 37.7º C and later 38.2º C, and on 14^th as 37.8º C. These are raised levels but only mildly so. Her pulse and respiratory rates were normal at 80 and 20.

She had a high neutrophil count, a feature of both preseptal and post septal conditions.

As to swelling of the eye, Claire had a severely swollen eye. There is no reason to doubt her mother’s evidence as to the degree of swelling, which tallies with the clinical notes, and as to the impossibility of distinguishing the brow from the cheek bone. Periorbital cellulitis, whether pre- or post-septal, gives rise to such swelling, not least because the loose tissue of the eyelid and surrounding the eye can swell to a considerable extent. This may appear alarming to someone who has not experienced this relatively rare condition.

As to proptosis, Claire’s mother’s description is that originally it looked as if there was a golf ball behind Claire’s eye and then a tennis ball. The first mention of proptosis in the medical notes is on 15^th March. It is not likely that, on 13^th March, Claire had proptosis consistent with her eye being pushed outward by something of the size of a tennis ball or a golf ball behind it. It is anatomically impossible for a tennis ball to have been within the orbital cavity; and inconceivable that something of the size of a golf ball was pushing the eye outward from behind without it being obvious to any examiner. I regard it as most unlikely that if there was proptosis of anything approaching that size, which would have involved dryness of the eye and made it red, it would have been missed by the GP, the SHO, and the Registrar on 13^th and by Dr Cameron, and others in the ward round, on 14^th March; or that, if it had been seen, it would not have been reported (Footnote: ¹³).

The scan carried out on the 15^th shows a slight degree of proptosis anteriorily, and a significant degree laterally; but, even then, it is not very great. By 15^th March there was a perceived deterioration of Claire’s condition as evidenced by the reduction in eye movement due to the developing subperiosteal abscess. In my judgment, although there was probably some proptosis on 13^th March, as the experts agreed to be the case (Footnote: ¹⁴), it was not to any significant degree. If the patient is suffering from orbital cellulitis there is likely to be some proptosis. Absence of significant proptosis is consistent with the eye not being red, although the fact that the eye was not red does not inevitably mean that there was no proptosis. Claire’s mother’s evidence is not wrong if it is understood as being a lay expression of a very heavily swollen eyelid and surrounding tissue.

As to drowsiness, the registrar recorded that Claire was, on 13^th March, “bright” and “alert”. Whilst I do not suppose that a little girl who was feverish and had been vomiting for 2 days was vivacious and lively, it seems to me unlikely that she was suffering from such drowsiness as would be a significant warning sign.

As to vomiting, Claire had been vomiting for two days. That is a common presentation of sick children. It does not appear to me to be a sign of significant systemic disruption. Her hydration was normal as were the electrolytes, especially the urea and creatinine readings.

As to the duration of her illness, Claire had been unwell for two days – a not atypical presentation in cases of this kind. She had been started on oral antibiotics; but, in the light of the vomiting, their potential effect must have been severely limited. In those circumstances her illness would be likely to get worse.

Taking all these parameters together, they do not appear to me materially to alter the clinical picture, which is of a sick child with features on the borderline between preseptal and orbital cellulitis.

Mr Clearkin suggested that the passage in the 1998 paper referred to in paragraph 60 (d) above

“Consensus in obtaining computerised tomography scan examinations, as an important ancillary test in the determination of orbital involvement exists in the following situations: suspicion of orbital involvement that cannot be determined solely by physical examination; or progression of the disease with no response to the proper antimicrobial treatment”

represented an acceptance of his position that in all or most of the cases where the cause of the periorbital swelling is not obviously preseptal, clinical examination would not determine whether there was orbital involvement, or its extent, and hence an immediate CT scan would be necessary.

I do not agree. Firstly, it is not clear to me that the reference to “suspicion of orbital involvement that cannot be determined solely by physical examination” was directed to a situation where a clinical diagnosis (or an assumption) of orbital cellulitis had been made, Secondly I do not accept that Mr Clearkin’s position is the message which the authors intended to convey, as is apparent from the appendix and algorithm to which I referred in that paragraph.

Further since one can rarely be certain that there is no orbital involvement, the application of such a principle would mandate scanning in most stage 1 cases, or cases of pre-septal cellulitis, which is certainly not standard practice in the UK even today. An audit was carried out of all patients admitted in 2006 with a diagnosis of periorbital cellulitis. (The admission statistics do not identify which of them had orbital cellulitis). 111 patients were identified, of which 17 (15%) were the subject of a CT scan. Mr Swift confirmed that that figure accorded with his clinical experience. All 111 were treated with antibiotics. 3 were treated surgically: 2 for drainage of an SPA; 1 for drainage of a dental abscess.

I also note that the literature does not contain a consensus bright line rule, that, if the clinician cannot be sure of a preseptal source, there must be an immediate CT scan. If I am to hold that failure to follow such a course is negligent, I would be condemning as negligent the practice of a number of hospitals and practitioners. I am not persuaded that this is a case where the evidence justifies such a course.

I do not regard the body of opinion which would not require an immediate CT scan in the Claire’s case as irresponsible or unreasonable. In circumstances in which (looking at the matter prospectively) a CT scan might detect something which would call for immediate surgery, it is arguable that a failure or refusal to do so is unjustifiable. Those who take a different view argue that a child should not be subjected to the risk of radiation, which although small is not illusory, unless it is essential to do so and that a CT scan should not be ordered unless the result is likely to alter the immediate treatment plan. Both Dr Cameron and Mr Swift indicated that the results of the CT scan would not have altered their immediate treatment plan.

That observation cannot properly be taken to extremes, and was not, I think, intended to be. If a CT scan showed cavernous sinus thrombosis, I have little doubt that surgery would have been effected. But it was, in my view, neither unreasonable nor illogical not to have an immediate CT scan in respect of a child who showed no significant reduction in visual acuity or ocular motility, or of intracranial involvement, and in respect of whom the result of the CT scan (whether it showed preseptal infection, orbital cellulitis or a small subperiosteal abscess) was unlikely to alter the immediate clinical management. I consider in paragraphs 97ff below whether, even on the assumption that there was a subperiosteal abscess on 13^th March, immediate surgery was mandatory.

I note also that, whereas Mr Hughes has always claimed that there should have been immediate scanning in Claire’s case, Mr Clearkin’s opinion’s original report was considerably less definite. In it he wrote:

“It is preferable to obtain a CT scan in all cases of orbital cellulitis as CT scan has been shown to detect subperiosteal and orbital abscesses which are not apparent clinically on plain films (Goldberg 1978; Schramm 1978) ….”

It is agreed that Claire was at least at Chandler Stage 2. But that stage does not involve a discrete abscess formation or the collection of pus. It is not easy to assess what a CT scan on the night of 13^th March would have shown. Mr Hughes did not feel able to commit himself to a view on that. Mr Clearkin thought that it would have shown a sub-periosteal abscess but a small one. Mr Hakin and Mr Swift thought it would not have demonstrated an abscess.

It is clear from the joint report from the neurologists that until about 1000 on Tuesday 14^th there was no intracranial pus, since their view is that it was only by 1000 that there were the beginnings of an accumulation of such pus. It does not, however, necessarily follow from their report that on 13^th March there was drainable pus either in the sinuses or in the orbit. The subdural collection on 15^th March was probably pus but all that that tells us for certain is that there was an infection, of which the sinus could be the source (Footnote: ¹⁵), which migrated into the intracranial space and produced pus there; not that on 13^th March there was drainable pus in the sinus or the orbit.

Mr Foster for the defendants submits that all that the neurologists’ agreement shows is that the focus of infection was present behind the orbit probably in the sinuses (Footnote: ¹⁶). The likelihood, he submits, is that there was no accumulation of pus as opposed to the existence of inflammatory cells and bacteria characteristic of orbital cellulitis, as described by Chandler: see paragraph 3 above. The report of the scan of 15^th March refers to “mucosal thickening of the anterior ethmoid air cells bilaterally, more particularly on the right and there are fluid levels in the right and left maxillary antra and in the right and left compartments of the frontal sinus”. The nature of the fluid levels is not specified.

He further calls attention to the fact that Dr Forbes in his oral evidence confirmed that, whilst the scan on 15^th March revealed abnormal tissue on the medial side of the top of the orbit, the CT scan cannot tell the difference between inflammation and infection. As, however, I noted in para 33 above Dr Forbes’ report refers to the existence of pus and pus was reported contemporaneously on 15^th: see para 30 above. Further pus was found when the first operation took place in the evening of 15^th.

I do not regard it as established that a CT scan carried out on the night of 13^th/14^th March would have revealed a subperiosteal abscess (or drainable pus elsewhere). I reach that conclusion for a number of reasons.

Firstly, on 13^th/14^th March there was no restriction of eye movements and even by 15^th March the restriction was only in one direction. That seems to me to be an indication that SPA was unlikely to have been present on the night of 13^th/14^th. I note that the European Position Paper on Rhinusinusitis and Nasal Polyps 2006 describes, at page 48, para 8-3-4, the clinical features of an SPA as:

“oedema, erythema, chemosis and proptosis of the eyelid with limitation of ocular motility and as a consequence of extra-ocular muscle paralysis, the globe becomes fixed (opthalmoplegia) and visual acuity diminishes”.

On the night of 13^th/14^th there was no significant proptosis, no significant loss of visual acuity, and no reduction in eye movement.

Secondly, I found the evidence of Mr Hakin and Mr Swift on this point convincing.

Thirdly I note that Mr Clearkin’s original report did not contain a claim that there was a subperiosteal abscess on the 13^th/14^th. What he wrote was:

“xiii)

Had Claire been appropriately assessed and managed on the evening of 13.3.95 it is perfectly possible that her condition would still have responded to medical treatment with intravenous antibiotics and nasal decongestants alone. Equally it is possible that surgery to drain her sinuses and possibly a subperiosteal abscess would have been required”.

Fourthly, even if there was by 13^th an abscess forming, it is likely to have been too small and might well have been impossible to pick up on the scan. Fifthly, there is insufficient material to signify that on the might of 13^th/14^th there was drainable pus in the sinus.

If that conclusion is right, it gives some support to the contention of Dr Sunderland that a premature scan can give a false degree of reassurance. Mr Clearkin disagreed with this objection to immediate CT scanning, which he regarded as necessary to establish a baseline for future diagnosis and treatment.

If, contrary to that conclusion, there was in fact a subperiosteal abscess to be found on a CT scan on the night of 13^th March, it would have been a small one. The next question is whether Mr Osborne would, in that event have drained it then. An allied question is whether, if there was no subperiosteal abscess but there was drainable pus in the sinus, he would have drained the sinus then. It is not suggested that there was some other surgeon whose likely course of action should be considered; if someone else is the relevant person, there is no evidence as to what he or she would have done.

As to that I do not think it likely that Mr Osborne would, in either event have intervened surgically on the night of 13^th/14^th. Although he indicated in his evidence that his indicator for surgery was the presence of pus he also indicated that he would not normally intervene surgically until a patient had been on antibiotics for 24 hours in view of the risk of osteomyelitis (Footnote: ¹⁷) and expressed the view that there was no urgency about intervention in the absence of any compromise to vision.

Mr Limb submitted that a problem with Mr Osborne’s evidence was that it was given against the background of the facts as he understood them to be; and that he was unwilling to consider what the position would be if the facts were otherwise. In the scenario presently under consideration the presence of pus would be established by the CT scan and, in that context, Mr Osborne’s indicator for surgery would have been shown to exist and the likelihood is that he would have operated.

I do not regard this likelihood to have been established. I do not think that the presence of a small SPA on 13^th, in the absence of any sinister ophthalmological signs, or of drainable pus in the sinus, would have caused Mr Osborne to depart from his norm. I reach that view in the light of Mr Osborne’s evidence and the fact, to which I refer below, that a substantial body of opinion would counsel against immediate surgery, even in the presence of a small SPA.

Many papers refer to surgical drainage as the appropriate treatment if an abscess is identified, and some to the unsafety of non surgical management: e.g. “Pediatric sinusitis and subperiosteal abscess formation: diagnosis and treatment” (Footnote: ¹⁸), which, after citing several authors who recommend that patients with abscesses uncomplicated by visual loss be treated with medical therapy only, states;

“We do not believe that the safety of exclusively non-surgical management of these patients has been established. We are unwilling to observe patients with physical and radiographic evidence of an SOA (Footnote: ¹⁹) while they are taking antibiotics alone, for the following reasons : children with SOA are difficult to evaluate for changes in visual acuity; SOA may develop and progress very rapidly; and serious intracranial complications of sinusitis (abscess or cerebritis) may develop without any change in visual acuity. We therefore perform abscess drainage and a definitive sinus drainage procedure for all children with SOA. An ethmoidectomy is performed in all cases; drainage of other sinuses depends on the distribution of disease as seen on the CT scan”.

Reference should also be made to “Complications of Sinusitis: Orbital Complications” in Disease of the Head and Neck, Nose and Throat (Footnote: ²⁰):

“It has been suggested that patients with a subperiosteal abscess can be treated effectively by intravenous antibiotics alone without surgical intervention providing that vision is normal. The safety of this approach is not established and has been considered hazardous in children in whom visual acuity is difficult to assess”.

There is, however, as Mr Clearkin acknowledged, a body of surgical opinion which would opt to treat even an SPA initially by way of intravenous antibiotics. There are a number of publications which favour this approach or which, whilst not doing so, refer to the controversy. They include:

As is apparent from the above citations, there are strong voices asserting that a policy which does not involve immediate surgery in the case of SPA is unsafe, or, at the lowest, one whose safety has not been established. They contend, with some force, that the fact that most patients recover if treated conservatively ignores, or, at least, does not adequately cater for, the position of those who would not do so. They make the case that, given the potentially severe consequences, the course of management should not be dictated by considering those who will, if conservatively treated, recover anyway; but by those who may not.

Those who have a different view, amongst whom the Trust’s experts are included, are not, however, in my judgement to be regarded as an unreasonable or irresponsible body; nor am I prepared to say that their logic is palpably defective. Operations are to be avoided if possible, particularly if they may, as the operation in this case did, involve scarring of a young girl. An operation carried out on 13^th March or early on 14^th March before intravenous antibiotics had any chance of working, and of sterilising any pus, ran the risk of seeding or spreading an infection. Mr Swift pointed out that surgical intervention into an area of acute infection ran the risk of disseminating the infection, including intracranially, and that the use of antibiotics before operating made the risks of dissemination much less. Mr Hakin thought that an operation might be irresponsible because it could lead to visual compromise.

Given a body of respectable opinion which would treat the infection with antibiotics initially (a course which also seems consistent with Mr Clearkin’s paragraph (xiii) – see para 90 above), it seems to me justifiable in this case for those concerned to wait and see if the antibiotics did their job – a course which would have the benefit that any subsequent operation would or might be able to take place in more propitious circumstances. Such an expectant period would have to end if there were signs of significant visual impairment (loss of visual acuity or restriction of eye movements) or there was no improvement after 24-48 hours. That is, in effect, what happened in this case where the deterioration in the patient’s eye movements and her failure to improve were the trigger for a CT scan and surgical intervention.

I have not ignored the fact that at the joint meeting in August 2006 Mr Swift agreed with Mr Hughes that surgery would be indicated that night. This was in answer to Question 13: “Please assume that the CT scan did show an abnormality that would have indicated a need for surgical drainage of the Sinus or some other surgical procedure. In those circumstances and having regard to your answers to the preceding question 9 please state by what time you would have expected such surgery to be performed.” By August 2007, when there was a further telephonic conference between experts, Mr Swift had changed his mind and recorded that he would not necessarily have operated that night.

I do not regard this change as invalidating Mr Swift’s evidence at trial. In changing his mind and in giving his evidence, Mr Swift was not, in my judgment, attempting to escape from an inconvenient truth to which he had assented, but expressing his true opinion. In this respect I note that both Mr Swift and Dr Sunderland expressed considerable discomfort at the course of the joint meeting of experts. I note also that some of the questions posed assumed what was to be proved. Thus question 13 assumed that the scan showed an abnormality that indicated a need for surgical drainage of the sinus or some other surgical procedure, without inviting a view as to what it was that was to be assumed to have indicated such a need. Against this background, I felt able to rely on Mr Swift’s evidence at trial.

Nor have I ignored the fact (a) that Mr Hakin indicated that he would expect an ophthalmological examination of Claire to include investigation of the following (i) visual acuity; (ii) pupillary response; (iii) eye movement; (iv) redness of the eye; and (v) degree of proptosis, and (b) that the notes of the registrar for 13^th March do not reveal that such an examination was done. However, items (i), (ii), and (iv) had been addressed by the SHO and item (ii) by the triage nurse; (iii) was probably addressed by the registrar, and was normal the next day; and (v) was insignificant and probably very difficult to detect. Events after 13^th March indicated that there were no visual problems.

These conclusions make it unnecessary to determine whether, if an immediate CT scan was mandatory, and would have indicated the need for immediate drainage, such drainage would or should have been carried out by such a time as would have meant that there was no epilepsy.

As to that, Mr Foster submitted that the evidence from the expert meeting as to how soon a CT scan, if required, would have been done was too vague to be of assistance. In answer to question 11 Mr Clearkin and Mr Hughes had said that a CT scan was required “that night”. Mr Clearkin said in his evidence that there should have been a scan by 9.00 p.m. and certainly by midnight. Mr Foster submits that on that evidence it would not have been negligent not to have had a scan until midnight by which time, on the neurologists’ evidence, drainage would be too late.

Had I concluded that, in the light of the history of Claire’s illness and her condition on examination by the registrar, an urgent scan was required to be carried out on 13^th, and, also, that such a scan would have shown a subperiosteal abscess of a kind that required drainage, I would have held that the scan could and should have been carried out in such a time as would have meant that drainage would have avoided the subsequent epilepsy.

On these hypotheses I would have been persuaded that an urgent CT scan was mandated once Claire had been examined by the paediatric registrar. That scan could and should have been carried out by 11 p.m. at the latest. I would also have been persuaded that if such a scan showed a subperiosteal abscess for which immediate surgery was required the likelihood is that it would have been carried out in time to avoid epilepsy. I do not regard midnight as a time by which, had the drainage operation not been complete, epilepsy would necessarily have followed. On the contrary in my view the likelihood is that an operation commenced, as on this hypothesis it should have been, as soon as possible after the results were known, i.e. before midnight, would be likely to have caught the problem in time.

It was for these reasons that, at the conclusion of the evidence and submissions on liability, I gave judgment for the defendants.