Barnett v Medway NHS Foundation Trust

In this case the Claimant appeals the decision of HHJ Brian Forster QC, sitting as a Deputy High Court Judge, in a judgment handed down on 23 February 2015, in which the claim for damages for pain, injury, loss and damage was dismissed. The Respondent cross-appeals.

The Claimant was born in 1958. He suffers from a rare congenital condition, hypophosphatasia. As a result he has deficient bone mineralization and is prone to stress fractures. Over the years before the index event he has suffered inflammatory reactive arthritis, multiple stress fractures with a repeated pattern of bone destruction, and a condition known as “pseudo gout”. He has had long-term prescription of steroids. Despite his difficulties, he remained in employment, latterly as a manager of an equipment hire depot. On a number of occasions, stress fractures arising from his underlying condition have required surgical fixation. This sequence of intervention culminated with a surgical revision, or nailing, of his left thigh in March 2009.

On 6 October 2009, the Appellant was admitted to the Medway Maritime Hospital, Gillingham following a period of acute pain. On admission to the hospital it was noted that the complaint of pain was at the back of the Appellant’s left thigh. The Appellant’s account was that he complained of pain in both buttocks, but the judge found that this was an error: he had not made that complaint in the hospital. The Appellant was given a seven day course of antibiotics shortly after admission to hospital. The Appellant was detained in hospital until 19 October, when he was discharged home. On 22 November, he was readmitted to the hospital as an emergency. His condition was very serious and an MRI scan revealed that he had an abscess in his spine at the level of L5/S1.

Further investigation revealed that he had an infarction of the lower thoracic spinal cord which, despite surgery, left him with paraplegia at the level of T7.

The Appellant’s case at trial was that his paraplegia should, and would, have been avoided by a reasonable standard of treatment during his October admission to hospital and in the follow-up to that admission. In very short summary, the Appellant’s case was that blood samples for culture should have been taken before he was administered antibiotics on admission to hospital. If such cultures had been taken, it is likely that infection with staphylococcus aureus would have been found. If so, a longer course of antibiotics would have been administered and his infection effectively treated, preventing the development of the abscess and then infarction of the spine. He should not have been discharged when he was. Discharge would only have been acceptable if accompanied by a detailed plan for close, frequent, follow-up and monitoring. Such monitoring would have meant discovery of the ongoing infection in sufficient time.

In essence, the Respondent’s case was that even if blood cultures had been obtained on admission and before administration of antibiotics, it is not probable that any infection would have been detected. In such circumstances, it was proper to give antibiotics for seven days, and proper to discontinue them, given the signs and symptoms at the time. It was reasonable to ascribe the Appellant’s condition in October 2009 to the development of renewed stress fractures rather than to infection. It was reasonable to discharge the Appellant from hospital on 19 October. Given the course of events, if it had been the case that infection was found after that discharge, then such later intervention would probably not have prevented the consequences. At the least, the Appellant was not in a position to prove such causation to the standard of probability.

Both sides recognised that the case was particularly complex. The Appellant’s underlying condition is rare. He had a long history of complications arising from that condition. Stress fractures can produce a very marked inflammatory reaction. A key test for inflammation, denoting the presence of dead or damaged cells, is the test for “C-Reactive Protein” [“CRP”]. The Appellant had in the past been shown to have very high CRP levels, without infection being demonstrated. Normal concentration in healthy human serum is between 5-10mg/L. The Appellant had a CRP level of 361mg/L on 16 October 2008 whilst no infection was demonstrated. At the same time the Appellant was markedly vulnerable to infection. The Appellant was routinely administered a combination of drugs, including steroids, which will probably have diminished the effectiveness of his immune system. Damaged bone and tissue represent sites more vulnerable than normal to infection.

The Appellant does not pursue his original Ground 4. He does pursue the following Grounds:

The Respondent cross-appeals on three Grounds, which were developed in the skeleton argument. The First Ground complains that the Judge permitted the Appellant to amend the Particulars of Claim on the first day of trial. The amendments concerned an allegation that had a test for inflammatory markers been performed on 12 October 2009 they would have been abnormal, and have stimulated further investigation. The Respondent complains the amendments were unsupported by expert evidence. The Second Ground complains that the Appellant was permitted to adduce new evidence from his expert to support the amendments to the pleadings, and to rely on a body of research literature, in fact largely produced by the Respondent’s microbiology expert, Professor French. The Third Ground is that the Judge “could and should” have made findings favourable to the defence in relation to the causation issue arising from the blood cultures taken on 6/7 October: the mirror image of the Appellant’s Ground 1.

In his remarkably succinct judgment, the Judge summarised four key issues. Both parties accepted before us that he did so accurately. They were:

The Judge made findings of breach of duty not now challenged by the Respondent. He firstly found that:

“48.

The experts are agreed that the obtaining of blood cultures before the prescription of antibiotics is a basic and essential step. For an unknown reason that action was not taken.

49.

Mr Bishop QC on behalf of the Defendant does not concede that this failure constitutes a breach of duty. I find that there was a breach of duty. The need was basic. It was essential to attempt to identify any infection. The risk of not doing so was that an opportunity to identify infection was missed. The procedure was simple and could have been carried out.”

The second finding of breach must be analysed at greater length. The treating orthopaedic surgeon, Mr Ahmed, stopped the antibiotics on 12 October. The Appellant’s condition had improved; his temperature was at a more normal level. His presentation was consistent, with local deep pain in the thigh. Mr Ahmed saw no need for further close monitoring of inflammatory markers.

The Judge considered the evidence of the two orthopaedic experts, Mr Wilson-MacDonald for the Appellant and Mr Dyson for the Respondent. In their joint statement following pre-trial meeting, the orthopaedic experts stated:

“It was agreed that it was reasonable for Mr Ahmed to discontinue the antibiotics on 12/10/09. Mr Wilson-MacDonald considers that this should have been followed by a thorough assessment of the inflammatory markers, and that an attempt should have been made to find the cause of the raised inflammatory markers.”

In evidence, Mr Dyson considered that serial measurement of inflammatory markers was undertaken at reasonable intervals. However, Mr Wilson-MacDonald told the Judge that it was very rare to see such a high CRP result where there is no infection. He considered that a fracture or micro-fracture of the type being suffered by the Claimant would not in itself result in such a high CRP. When the antibiotics were stopped he emphasised it was essential to monitor the inflammatory markers. This should have been carried out on two occasions during the first week and thereafter on a weekly basis. In cross-examination he maintained that a failure to repeat testing of CRP at any stage was below the level of reasonable competence.

In evidence, Mr Dyson considered the case to be more complex than he had initially realised. He confirmed that the CRP of 371 was very high. At the time there was no clear explanation for this high result. Even now, he considered the reading to be in a sense unexplained. This was in part because, even if there had been an infection of the spine, it would not have given rise in itself to such a high result. In his opinion several things must have been going on, some of which we do not know. It was reasonable to stop the antibiotics, but he conceded close management of both the CRP and white cell count was required.

Following that evidence, the Judge concluded:

“62.

Taking all matters into account, I find that it was reasonable to stop the antibiotics, but that a doctor with full knowledge of the Claimant’s background, acting reasonably and prudently, would have arranged for close monitoring of the CRP and white cell findings.”

The judge’s conclusions in paragraph 62 represent a finding of breach. There was no close monitoring of the Appellant after his discharge, a point conceded in argument before us by Mr Bishop QC. However, the Respondent continues to argue that, if it had been the case that closer monitoring would have led to discovery of the infection after discharge from hospital, that would not have altered the outcome.

The critical questions on causation arising from the failure to culture blood samples were, firstly would the infection have been detected, and if so, would such detection have led to effective treatment. On this aspect of the case, the critical evidence came from consultant clinical microbiologists, Professor Wilson for the Appellant and Professor French for the Respondent.

The second issue arising from this breach was agreed between the experts and conceded before us: if infection had been discovered from blood cultures on admission, long term intravenous antibiotic therapy would have been given. The infection would have been eradicated. No spinal abscess would have developed and no infection. The Appellant would have avoided paraplegia.

The question whether blood cultures would have disclosed infection was much more complex, and depended on the timing and progress of infection. The judge recorded his conclusions as follows:

“67.

There are two issues. Whether there was then an infection in the spine or blood to be identified and whether it would have been identified.

68.

The microbiologists expressed different opinions as to whether there was a spinal infection at that time and as to whether there would have been positive blood cultures. Professor French considered that there would have been a silent spinal infection. In view of the absence of typical symptoms he believes that the Claimant comes within the group who would not produce positive cultures. Professor Wilson believes that the Claimant comes within the majority group identified in the research literature where a positive result would be obtained.

69.

Professor Wilson also considers that a staphylococcal bacteremia present during October could have resulted in the spinal abscess developing in November.

70.

I note that in his own report at page 31 Professor Wilson stated “There was no unequivocal evidence of infection before mid November 2009. The fever, leg pain and raised WBC in October 2009 would have been consistent with infection or repeated stress fractures due to hypophosphatasia….a staphylococcal bacteremia during October could have resulted in the epidural abscess developing.”

71.

Both the microbiologists stated that the case is difficult and it is clear to me that this is an area of considerable uncertainty. I can find no evidential base established to the required standard. This is an unusual situation in which there is a lack of actual evidence to help with a determination of the conflicting opinions. In such circumstances whilst I found the evidence of Professor French the most likely I have not been able to make findings concerning the onset and progression of the infection to the relevant standard of proof.

72.

I have carefully assessed the evidence and find that the Claimant has not established on the balance of probabilities where the infection was present or that blood cultures would have been positive. It is also only a possibility that a bacteremia was present at the time the blood cultures should have been taken.”

It is this conclusion which is attacked by the Appellant in Grounds 1 and 2, and by the Respondent in Ground 3 of the cross-appeal.

The Judge went on to consider the second causation issue: what would have been the effect of close monitoring of the Appellant following discharge from hospital? He firstly concluded that regular CRP tests from discharge on 19 October through to the appointment on 9 November 2009 would have shown “a falling CRP level”. By 19 October, Professor Wilson had accepted that the CRP level would probably have been down to around 150. When tested at the outpatient department on 9 November it had reduced to 97. The Judge found that that predicted CRP level and the result of the white cell scan:

“75.

…would have been considered as showing an improving picture. I can find no reasonable basis to suggest that there would have been any significant change in patient management before the result of the white cell scan was known. The scan report stated that the findings were in keeping with an inflammatory process, suspicious for infection, in the left knee. There was no evidence of osteomyelitis within the distal left femur.”

The Judge considered the evidence of the various experts as to whether an MRI scan of the spine would have been carried out and concluded that:

“77.

It has not been established on the balance of probabilities that an MRI scan should have been performed or that monitoring would have led to such a scan of the spine.”

Finally, the Judge went on to consider the question “what would have been the effect of recommencing antibiotics on 9 November?” Here he stated his conclusions as follows:

“79.

I consider this in part to underline the difference in opinion between the microbiologists.

80.

Professor French states that, in order to have produced signs and symptoms of cord compression, the spinal infection that led to an abscess must have started before the 15 October. If antibiotics had been recommenced on 9 November, given the period without antibiotic treatment there would have been infarction regardless of the restarting of antibiotic treatment on 9 November.

81.

Professor Wilson considers that the abscess could have started as late as 10 to 21 days before 15 November. That would be from 25 October to 5 November. Accordingly he considers that the giving of intravenous antibiotics from 9 November with surgical intervention would on balance have been sufficient to avoid infarction.

82.

Each expert considered this to be a complex and difficult case. Professor Wilson accepted that it was impossible to have all of the answers and, as stated above, Professor French was of the opinion that even now it was impossible to understand everything that had happened.

83.

Taking all matters into account, it has not been established to the required standard that the infarction would have been avoided if antibiotics had been recommenced on the 9 November.”

In his final summary of his conclusions, the learned Judge repeated that:

“the microbiologists expressed very different opinions as to when the spinal infection must have commenced and accordingly as to the latest state accordingly as to the latest date at which further antibiotic treatment would have prevented infarction.”

Thus, it had not been established on the balance of probabilities that the failures which were breaches of duty caused the consequences alleged.

The judge re-emphasised his findings that “in many respects the treatment received and practices followed were poor” (paragraph 91), but that because of the difficulties of causation there was judgment for the Defendant.

The Appellant groups together his grounds 1 and 2 concerning the missing blood cultures. The Appellant submits that having concluded that there was a breach of duty “the judge should have determined whether [or not] the blood cultures would have been positive for infection” and that the Judge “erred in law and/or was wrong by making impermissible resort to the burden of proof”. The Judge “failed to analyse the evidence”.

The Appellant submits that there was no dispute between the parties: there was in fact infection present during the Claimant’s admission to hospital between 6 and 19 October. The Appellant submits that the dispute was confined to the following: “(1) whether at the time of admission the infection was already in the spine (Professor French for D), or (2) whether it was in the form of bacteria circulating in the blood (Professor Wilson for C)”. The microbiologists were agreed that if bacteria were circulating in the blood, blood cultures taken on 7 October would have proved positive for infection. If the infection was already “seeded in the spine”, the Appellant criticises the judgment for failing “to acknowledge … that the microbiologists were agreed as to the likelihood of the infection being identified if there were bacteraemia”.

The Appellant also criticises the Judge for summarising the position of Professor Wilson as a suggestion that bacteraemia “was only a possibility”, see the judgment paragraphs 69 and 70. The Appellant submits that Professor Wilson’s position in evidence was that it was likely there was bacteraemia present. If Professor Wilson’s conclusions on this point were to be rejected, the Judge had an obligation to analyse the evidence and spell out the evidential basis for such a conclusion.

Appended to an expert report from Professor French served in advance of the trial were a number of research papers bearing on the likelihood of detecting infection in blood cultures taken from patients with spinal abscess. Although served by Professor French, this body of research had not been the subject of debate between the microbiologists before the trial, and it is this material about which the Respondent cross-appeals directly in the Second Ground and by implication in the First. It is helpful to address those points now.

In my view there is no substance in these arguments. The amendments to the pleading were in large measure related to the literature produced by the defence expert Professor French. It is of course correct that the amendments should have come earlier and that this literature should have been the subject of discussion between the microbiology experts before trial. The progress of the infection, the prospects of discovery through blood cultures and the effectiveness of treatment after discharge were all along critical issues. However, this material came from the Respondent. Time was allowed for the defence to accommodate the evidence. It was not explored in detail in the trial, but it could have been the subject of detailed cross-examination. There is no basis for saying there was any real injustice to the Respondent. The judge’s decision represented sensible case management. I would dismiss these Grounds without hesitation.

Returning to the substance of the case, Professor Wilson for the Appellant sought to rely on this body of literature as demonstrating overall a 59-80% chance of detecting infection from blood cultures. He argued that there was one paper which was “out of kilter with the others”, showing a chance of detecting infection in the blood samples of only around 43%.

In the course of his response on this evidence, Professor French accepted that around 60% of the spinal abscess patients in the studies did have positive cultures, but argued that the position of the majority of such patients was distinguishable from that of the Appellant. The criteria for inclusion in the studies for the most part included signs of neurological abnormality. Hence, the cohorts giving rise to these studies consisted of patients with more advanced conditions than would have been the case for the Appellant, viewed at 7/8 October 2009. The Appellant was not comparable to the cohorts studied. For that reason, Professor French’s view remained that it was improbable that blood cultures in early October would have revealed infection.

The Appellant is critical of the absence of analysis of this causation issue by the Judge. The Appellant relies on the principles laid down in Stephens v Cannon [2005] EWCA Civ 222, where in the course of his judgment, Wilson J (as he then was) said:

“46.

…A court which resorts to the burden of proof must ensure that others can discern that it has striven to make a finding in relation to a disputed issue and can understand the reasons why it has concluded that it cannot do so. The parties must be able to discern the court’s endeavour and to understand its reasons in order to be able to perceive why they have won and lost. An appellate court must also be able to do so because otherwise it will not be able to accept that the court below was in the exceptional situation of being entitled to resort to the burden of proof.”

The Appellant further recites two passages from the judgment of Auld LJ in Verlander v Devon Waste Management [2007] EWCA Civ 835:

“19.

…First, a judge should only resort to the burden of proof where he is unable to resolve an issue of fact or facts after he has unsuccessfully attempted to do so by examination and evaluation of the evidence. Secondly, the Court of Appeal should only intervene where the nature of the case and/or the judge’s reasoning are such that he could reasonably have been able to make a finding one way or the other on the evidence without such resort.

…

24.

When this court in Stephens v Cannon used the word “exceptional” as a seeming qualification for resort by a tribunal to the burden of proof, it meant no more than that such resort is only necessary where on the available evidence, conflicting and/or uncertain and/or falling short of proof, there is nothing left but to conclude that the claimant has not proved his case. The burden of proof remains part of our law and practice -- and a respectable and useful part at that -- where a tribunal cannot on the state of the evidence before it rationally decide one way or the other. In this case the Recorder has shown, in my view, in his general observations on the unsatisfactory nature of the important parts of the evidence on each side going to the central issue, particularly that of Mr Verlander, that he had considered carefully whether there was evidence on which he could rationally decide one way or the other.”

The Respondent re-emphasises the complexity of the evidence facing the Judge. Not only was the question inherently difficult, but the Respondent submits that the evidence advanced by Professor Wilson on behalf of the Appellant was much more complex than the Appellant submits, and suffered from inconsistency and internal contradictions. In his oral submissions to us, Mr Bishop QC for the Respondent drew out from the transcript of evidence what he submitted were evident inconsistencies, changes of position, and in one instance at least, acknowledged guesswork on the part of Professor Wilson.

One of the features of this case is that the judgment is so compressed that a review of the judge’s conclusions on causation cannot easily be conducted from the judgment itself. The matter can only properly be addressed from the transcript. I return to this below.

Before following Mr Bishop in looking at the transcript it seems to me helpful to give some context which, although not set out in the judgment, appears not to be controversial. The critically damaging event for the Appellant, the proximate cause of his paraplegia, was the compromise of perfusion of his spinal cord: the infarction. The infarction was caused by physical pressure on blood vessels supplying the spinal cord, which itself was a consequence of the formation and extent of the abscess. The abscess was derived from infection, localised to the spinal area already identified. Although timing is in issue, it is agreed the infection had been present at the time of the infarction for some considerable period. When the surgeons operated on the Appellant, they found localised damage to the facet joints and surrounding tissues. This damage was an indication of the presence of infection over a period. It was the product of the infection, not of the compromise of blood supply to the cord. Further, it was agreed that transmission of infection to the spine had been through the bloodstream. Therefore at some point or points the Appellant must have had bacteraemia, that is to say live bacteria being carried around the bloodstream, as opposed to localised infection in the spine or indeed elsewhere in the body. It also appears to be agreed that a spinal abscess can develop and persist without bacterial infection being detectable in the bloodstream. This generalisation is consistent with the scientific literature mentioned above. In the majority of the studies produced, a significant minority of patients with spinal abscesses cultured no bacterial infection from blood samples.

Both microbiologists agreed the complexity and difficulty of the sequence and timing of events. Although bacteraemia led to the “seeding” of the infection in the Appellant’s spine, there were no easy conclusions as to whether that occurred once, or more than once; whether there were multiple episodes of bacteraemia; where the infection began in the body; whether the abscess, once formed, sealed itself off and released no infection into the bloodstream, or whether bacteria were present in the bloodstream throughout. Finally, it was implicit from the way both experts addressed the matter, that the Judge was considering a connected cascade of events, beginning with the onset of pain before the Appellant’s admission to hospital in early October and culminating in the infarction.

In my judgment it would have been helpful had the Judge set the parameters for the complicated microbiological evidence in some way, so as to give a context for his consideration of the evidence of Professors Wilson and French.

In his principal report, Professor Wilson noted that “symptoms due to the epidural abscess started around 15 November 2009”. He noted that there was then altered bladder and bowel function with further abdominal back and leg pain from 18 November. He was admitted to hospital on 22 November with motor function and sensation still intact. “Appropriate antibiotic treatment” was started early on 23 November. Professor Wilson then said, giving his opinion on causation, that:

“It is unlikely that any alternative antibiotic management after 21.11.09 would have prevented the neurological sequelae. He would have had to be admitted and treatment would have had to be started earlier … If he had been treated with antibiotics intravenously before 21.11.09 he would still probably have needed early surgical intervention. Surgical decompression and drainage of pus on the afternoon of 23.11.09 would on balance have prevented spinal cord infection (sic) and resulted in full recovery. However intervention after 23.11 probably would not have altered the outcome.”

In his report of 11 November 2013, Professor French noted that the Appellant on 6 October “had pain and swelling in his thigh and signs and symptoms of possible infection … osteomyelitis of the femur was suspected”. Professor French said that the pain and abnormal inflammatory markers during the October admission:

“Were probably due to the spinal infection and/or infection at another site. …. I think the spinal infection began 4-8 weeks before 22 November.”

He went on to say, in relation to blood cultures during the October admission and/or outpatient visits on 9 and 19 November that he could not “confirm on the balance of probabilities that the cultures would have been positive”.

Remarkably, when they came to discuss the case before trial, the microbiologists were not asked to consider whether cultures taken on 6/7 October would have revealed infection. Professor French repeated his view that the spinal infection that led to the abscess must have started between 15 September and 15 October. Professor Wilson gave his view that the abscess could have started to develop as late as 10-21 days before 15 November; that is to say in the third or fourth weeks of October. Professor Wilson’s answer in the record of the meeting is not clear as to when he dated the onset of infection, as opposed to the development of the abscess.

Professor Wilson’s oral evidence was given in cross-examination by Mr Bishop and in re-examination. He observed that the CRP level shown by the Appellant on admission meant that “infection must come near the top of your diagnostic list”. He agreed with the rather obvious proposition that if the Appellant had “no infective process anywhere” the blood culture would have been negative. That question and answer beg the essential question: the CRP levels would have meant he would have still been concerned that an infection was present. He stated that:

“On the balance of probability the infection was there in early October. What might have happened, he may have had an infection in the bone at that point, or he may have had a staph aureus in the blood, which then seeded into the damaged area of bone in time. This is something we see not uncommonly in (inaudible) in the blood, and staph in the blood may have come from a number of other soft tissue sources of which (inaudible) other areas. So it could have been an early osteomyelitis then, or it could have been an area of bone that had just about been seeded with these organisms at that point.”

Professor Wilson went on to say that if there was an infection in early October:

“Then it is going to take several weeks before the organisms grow and pus accumulates to an extent at which we start to see obvious signs.”

Soon thereafter followed this exchange:

“Q: So if he had a disc infection in October you would expect that that infection would start to show, both signs and symptoms, by the end of October, is that right?

A: I would have expected that. Clearly, as we have seen, it is a complicated problem, because we just didn’t get the level of pain that we would have expected with that presentation.

Q: No.

A: I mean sometimes we do see that kind of thing, and we also see patients who don’t develop a fever when you really would expect them to. And in that situation we tend to rely more on the blood tests where double the white cell count means a CRP. So I’d be looking at both. I’d be looking at the patient and I’d be looking at the process.

Q: Yes. So if that statement is correct, then another reason why this is a very unusual case is that after --- if he had an infection of the spine when he went in in October you would have expected that by the end of October, even after his six days on antibiotics, he would start to become ill again?

A: Yeah, I would expect that.

Q: And that did not happen?

A: He, he is an unusual case.

Q: Yes. Might the statement not also mean that actually, it is more likely than not, this was a late developing infection and he did not have it in October?

A: I think given his presentation that we know occurred later, in retrospect I think he did have it then. I think that Professor French agreed with the, on the balance of probabilities he did have it. But I fully accept it is an unusual case.

Q: Yes.

A: And it does take a long time for that degree of pus to accumulate, and for the damage to (inaudible) up.”

And a little later came the following exchange:

“Mr Bishop: I just want to pick you up on something you told us a moment ago, that he had fever when he came in and then the antibiotics would have brought that fever down?

A: Well it could; yes, you, you’re right. It could have been an anti-inflammatory bringing the fever down. It is not necessarily the antibiotics that brought the fever down. I just think that is the most likely.

Q: Yes. Well again concentrating on whether there was infection in the spine in the October admission, the fact that he did not run a pyrexia, pretty much at all, after the first time he was assessed by the (inaudible) in the early hours of the 7^th October is against there being an infection, is it not?

Q: Well let me take you back to the answer that I gave earlier, was that he actually had staph aureus in his blood on admission.

Q: Yes.

A: Now that could have been from the soft tissue source. It could have been quite minor and, at that point, the organisms settled into the spine. Now if that’s the case then you wouldn’t see any visible signs. There wouldn’t be any obvious signs or symptoms of infection in the spine. But some weeks after that seeding had taken place the spine is infected, but there is no outward sign of it being infected, that, that would seem a possible explanation to me.

Q: Yes.

A: So you would expect then him to be getting an osteomyelitis in a very, very small area while he’s in, which might not be obvious amongst all the other degenerative disease.

Q: Well I think this is probably a case where we can make the facts fit pretty much into any theory that we put forward.

A: Unfortunately, yes. Unfortunately I (inaudible).”

As this evidence developed, the Judge intervened following a further question from Mr Bishop:

“Q: All right. Well what I suggest to you, sir, is that when one looks at the whole picture of that admission, even with hindsight when you look at all of the evidence, it is more likely than not that there was not an infection in the spine in October 2009 during that admission.

A: I, I think it’s more likely not that there was an infection, and not necessarily in the spine, in strict scientific terms --- I suspect there was an infection in the spine but not one that anybody would have been able to pick up.

Judge Forster: So really your considered view is that there is more likely than not to have been an infection, but that would have been an infection in the blood leading to the seeding in the spine?

A: Yes, and whatever soft tissue, or possibly bone elsewhere, had seeded the blood in the first place.

Judge Forster: Yes.

A: So there, there would have been ---

Judge Forster: But that in itself, you thought, could have been, in the scheme of things, a more minor situation?

A: It, it’s possible. The only, the only thing that worries me though, is that you have such a high CRP and that usually does mean deep seated infection, not just a superficial (inaudible). So it means that there has been, there’s something in there that, if it’s due to infection (inaudible), it’s something that’s been there for a little while and something that is going to get serious, usually in the bone or the deep tissue. So the source, with that level of CRP, would be more likely to deep than superficial, but I just can’t say it’s from the spine. The spine may have been a result rather than the cause.

Mr Bishop: So that rather goes contrary to the idea does it not, that actually what was going on was some sort of bacteremia from some minor source of infection that was, or has just seeded in his spine?

A: It is not a clear case.”

In the course of re-examination, the following exchange took place:

“Q: You have told the learned judge that you believe that this abscess took a long time to develop ---

A: Yes.

Q: --and it would have been there in October, yes?

A: Yes.

Judge Forster: Have you actually told me that?

A: Well I, I’ve said that that, I thought there were organisms inside him which were developing.

Judge Forster: Yes.

A: Now what the size of the abscess was during October I think is a matter for speculation. But to get to the size that it was in November, it must have been there a couple of weeks, or probably more than that, because it would be slowly growing in that time because it wouldn’t just suddenly appear. So I think there was an abscess in October but it may have been very small.

Mr Wilson-Smith: This was addressed in the Joint Statement. Do you understand there is any difference of view between you and Professor French, other than him thinking that it had taken rather longer to develop?

A: That’s correct, yes. I do, yes.

Q: Yes.

Judge Forster: I was just following the progression from being in the blood to seeding, if it happened in that way.

A: Yes.

Judge Forster: To then being described as an abscess.

A: Sir, sir, there’s two possibilities. Either there was a defined abscess there even in late October, which gradually grew over the two months thereafter. Or that was that the point at which the bone became seeded and then the abscess developed in the two weeks after that. I, I don’t know which of those two.

Mr Wilson-smith: You were asked to assume that there was no bacteremia, yes?

A: Yeah.

Q: And that you should disregard, when making that assumption, the high CRP, and you have pointed out other factors too, the high white cell count and so on. Are you able to help as to the likelihood of there being a bacteremia in this case?

A: I think you can say almost for certain that there was a bacteremia at some point. I don’t necessarily know exactly when it took place because that is how the psoas muscle and the bone would have been seeded in the first place. So there must have been a bacteremia, but I don’t know whether it occurred on the 7^th October or at some point earlier. My guess is that it occurred on the 7^th October.”

When Professor French came to give evidence, he began by altering one aspect of his view in evidence in chief. He qualified his view of the signs on admission to hospital and he stated that it had become his view the Appellant did not have infection in his leg on admission in October. He then said this:

“A: Well I don’t think he did have infection in his leg, although it’s possible, and we’ve heard several comments on this case saying that people don’t really quite know what was going on with him. It’s possible he did have some infection somewhere in his bones, his legs. I think he did have infection of the spine when he presented, but it does appear now that other people, radiologists, and indeed what Professor Wilson just said, that they may not feel that he did have an abscess in his spine at presentation.

The reason why I think he, he, well what I thought when I heard the additional information, what I (inaudible), was because of the degree of infection that he had when he finally went to surgery with the psoas abscess, osteomyelitis and the destruction of the spine at that point. The problem with this man is that he was on steroids and therefore that would have modified the time at which the destruction occurred. If we go back to his admission, in my opinion he didn’t have signs and symptoms of a spinal abscess, and indeed from what I’ve heard from the other experts in this trial, that is widely agreed. Professor Wilson, I think, has just been saying, if that happens to be (inaudible), that he now considers it is possible that the admission in October 2009 was associated with a staph aureus bacteremia and that the spine was seeded, that is infected, at that time.

In my report I think we both agree that the abscess which was eventually found in the spine, must have been infected by the bloodstream at some previous point. I had thought, and I think Professor Wilson had some (inaudible) this view, that the bacteremia, that is bacteria in the blood, that infected the spine and it would have infected, I agree, an area of the spine that had been previously damaged by degenerative disease. I (inaudible) see that it occurred some while before presentation of the (inaudible) 2009. Thus having the infection in the spine, but which did not produce the classical signs and symptoms of an epidural abscess. They occurred later. From what I have just heard, Professor Wilson seemed to be suggesting that another possibility is that he had bacteremia at presentation, and that the spine actually was infected then, in October 2009. If that was the case then he would of course have had no signs and symptoms at all of a spinal infection at that stage, because it would have been too early.”

Professor French then repeated that he did not think the blood culture would have been positive soon after admission, although he did concede that the extremely high CRP might indicate infection and that result:

“has never really been properly explained even with his spinal infection. CRPs at that level are still unusual even with infected cases. We have also had the information that he previous raised CRP … in other words he was a very difficult case to analysis (sic) … the possibility of infection remained but I must say … I would have still considered the most likely issue was an infection in the bones … I think I would have said he probably has got an infection somewhere. I think the infection is most likely in the leg … he probably has other rheumatologic problems as well.”

In a further passage, Professor French went on to say that on admission to hospital on 6 October:

“So if we now go back to the balance of probabilities assessment, I concluded on the balance of probabilities he would not have had a positive blood culture.

In addition I have to say I think there may be difference of opinion on this, but in my opinion he was not a patient who came in with severe sepsis. He had a fever but it was not very great and it soon settled. His main complaint was of pain in his leg and that might have contributed to his raised pulse rate and respiratory rate, because the pain was very severe. He wasn’t somebody, in my mind, to my mind, who presented seriously ill with sepsis. Such patients are obviously again less likely to be (inaudible) positive in the blood, remembering of course that most blood cultures are in fact negative (inaudible). So he is already a difficult patient and very difficult to analyse, and we’ve heard that repeatedly from the expert clinicians over the last few days. I think when he presented he was not presenting with sepsis, he was presenting with pain in his leg. I think he might have had an infection in his leg but the evidence is all on the side. I don’t think he had an abscess in his spine that was producing any kind of symptoms, and I think if a blood culture had been taken at that time it would have been negative on the balance of probabilities.”

In cross-examination, Professor French maintained his position, as indicated in his initial report, that there was already a spinal infection on 7 October, although he conceded this was uncertain. He confirmed that there would have to be a bacteraemia before infection could lodge in the spine. He agreed that a bacteraemia but no spinal abscess on presentation was “certainly a possibility”. He agreed that a bacteraemia “only” would be unlikely to explain the CRP of 371 and there must have been other factors at work. He agreed that accepting the presentation of the Appellant in hospital “the instant consideration is infection”. He agreed that the patient needed extremely close monitoring from his admission onwards and in particular if he was to be discharged from hospital.

The Appellant submits, in support of these Grounds, not only that the judge reached “impermissible” or erroneous conclusions, but that the judgment contains no sufficient analysis of the case, and thus the Appellant is left in doubt as to the judge’s reasoning. With respect to the Judge, there appears to me to be force in the latter complaint. There is great virtue in writing judgments concisely. However, the parties do need to know sufficiently what led to the conclusions reached. In this instance, the judgment gave only the briefest explanation. The obligation is all the clearer in a case of such complexity, and in a case where a key issue is decided on the basis that a claimant has failed to discharge the burden of proof, as the passage from paragraph 46 of Stephens v Cannon quoted above makes clear. The learned judge is to be commended for his brevity, but on this aspect of the case at least, it went too far.

However, having now conducted the exercise of examining and evaluating the evidence given before the judge, I for my part understand why he reached the conclusion he did on this issue. Not only was the medicine particularly difficult, but the evidence of the two microbiology experts was expressed in difficult and shifting terms. The excerpts from the transcript set out above are relatively long for an appellate judgment, but represent a small part of the relevant evidence. Both experts shifted position. The evidence of both experts was somewhat rebarbative. I consider the judge was justified in saying Professor Wilson considered bacteraemia in the blood was a “possibility”. Taken as a whole his evidence fell short of establishing probability. It is clear from an exchange with counsel after the close of evidence that the judge was seeking help to find firm ground in all this. I am not surprised he was unable to do so. In my view this was indeed one of those rare cases where the judge was justified in his inability to resolve an issue of fact consistent with the approach laid down in Stephens v Cannon and Verlander v Devon Waste Management. I therefore would reject the appeal on Grounds 1 and 2 and the cross-appeal Ground 3.

The Appellant’s complaint here is that the judge was in error in finding it was acceptable to stop the administration of antibiotics, and subsequently to discharge the patient, because the condition which made those steps acceptable, continuing close monitoring, was not intended to be carried out, and was not carried out. I reject this Ground. It appears to me a misunderstanding of the judge’s conclusion. His conclusion was that the failure was in the monitoring, not the cessation of drugs or the discharge. He summarized it as follows:

“The decision to stop antibiotics was reasonable, but having regard to the high CRP there should have been close monitoring of the Claimant.”

I see no basis in the evidence why that conclusion was to be faulted. The Appellant’s Ground 3 is based on a misunderstanding, indeed an inversion, of the judge’s justified conclusion. I would reject this Ground.

I have already indicated a good deal about the complex, shifting and conflicting views of the microbiologists as to the timing and sequence of infection, bacteraemia and the development of the spinal abscess. The platform for the allegation that antibiotics should have been resumed in November 2009 depends on more active monitoring after 12 October, and on whether monitoring would have led to earlier diagnosis of infection. Even if antibiotics were resumed earlier because infection was found in the bloodstream, the outcome would not alter unless antibiotics would have prevented the infarction.

Towards the end of cross-examination Professor French gave answers as to the effectiveness of antibiotics delivered in November, which encapsulates the difference between the experts on this issue:

“Q: …. so. There is a difference of view between you as to how effective antibiotics would have been later on in November, the 9^th and 19^th November, yes? You heard the explanation given yesterday by Professor Wilson as to his reasoning for that, yes?

A: Yes.

Q: He told us that cytokines and white cells respond to infection, yeah?

A: Yes. That, that’s his opinion, yes.

Q: Is he wrong? Let me take it, if I can, fairly shortly. Do you dissent from the view he expressed there?

A: I dissent to a certain extent. Can I, my Lord, may I explain why I don’t quite agree with him? It’s not (inaudible).

Judge Forster: But you are not saying that he is wrong?

A: I’m not saying he’s wrong but I disagree with the time, with the dynamics if you like. This is because, as he rightly says and I agree with him, or agree with most of (inaudible), he rightly said that antibiotics don’t immediately stop the inflammation. He then talked about the half life of cytokines, which I accepted in a test tube, but I’m not sure it’s entirely right within the body. The problem here is that he, my understanding is that he thinks the infection began later than I did. And because I think the infection began earlier, and that is partly related because of the fact that he had extensive bone marrow destruction, osteomyelitis deep in the bone, and it’s a large psoas abscess as well, in my view giving antibiotics --- and the answer, the question was asked, perhaps wrongly, “What would be the effect of IV antibiotics?” IV antibiotics (inaudible) they would have found it difficult to get into (inaudible) properly, and therefore with that massive pus inflammation and tissue destruction, I don’t think that antibiotics would have quickly reversed the problems. Although the cytokines would have been reduced there would still have been dead and living bacteria there stimulating these cytokines nevertheless. So this is why I think we have a difference of opinion because I think the infection began earlier, and I think at that time on the, certainly on the 19^th, on the 9^th as well I suspect, I think there’s more destruction than Professor Wilson, and I think he can speak for himself, and therefore I don’t think the antibiotics alone would have been effective. I think antibiotics together with drainage on the 19^th, it was too late. Antibiotics plus drainage on the 9^th, having heard his opinion, it may well have been effective at that stage, but it would have required surgery and drainage and I think indeed that is what he said.”

It follows that the debate as to the effectiveness of the resumption of antibiotics in November, consequent upon more active monitoring, threw the experts back into their earlier debate as to the sequence and timing of the infection, and the development of abscess.

I see no basis for criticising the judge’s conclusions here. Since the microbiologists are the prime experts on this issue, and since they provide no clear picture as to the course of the infection, I see no basis on which the judge can be faulted. For the reasons already developed, he was fully entitled to conclude as he did in paragraphs 79 to 83 as set out above.

For those reasons, I would dismiss the appeal and cross-appeal.

I agree.

I am indebted to Irwin LJ for his careful analysis and I too agree.