Hannigan v Accident Compensation Corporation

IN THE DISTRICT COURT AT NELSON [2013] NZACC 327 ACR 472/12 UNDER THE ACCIDENT COMPENSATION ACT 2001 IN THE MATTER OF AN APPEAL UNDER SECTION 149 OF THE ACT BETWEEN BRENT HANNIGAN Appellant AND ACCIDENT COMPENSATION CORPORATION Respondent Hearing; 16 July 2013 Appearances: D Wadsworth (advocate) for Mr Hannigan J Castle for the Corporation Judgment: 11 October 2013 JUDGMENT OF JUDGE RODERICK JOYCE QC Reason for appeal [1] Mr Hannigan has cover for a back injury sustained on 22 July 2006 and a neck injury sustained much earlier in 1992. [2] On 13 January 2011 the Corporation advised Mr Hannigan that it had assessed his claim for a treatment injury but could not approve same because it did not meet the criteria for that. [3] Mr Hannigan applied for a review of this decision but his application was dismissed on the basis that the medical treatment in question had not resulted in any distinct injury. [4] The written submissions for the Corporation on appeal included this: 1.2 ... there has been some confusion as to whether this decision was made under the treatment injury provisions sections 32-33 of the Accident Compensation Act (the Act), or as a personal injury that is a consequence of treatment given to the person for another personal injury for which the person has cover under section 20(2)(d) of the Act 1.3 The decision letter did not refer to the specific legislative provisions, however the Treatment Injury Report attached to the decision letter states that the exclusionary criteria of necessary part and ordinary consequence does not apply in this case as the treatment was for a covered personal injury. 1.4 The position of the respondent is that no personal injury has been established, and secondly, causation with the identified treatment is not made out. [5] In any event Mr Hannigan has appealed the dismissal of his review application and the purpose of this judgment is to determine that appeal. Background [6] On 25 February 2008 Dr Xiong wrote that; Mr Hannigan was in quite severe constant pain. His basic activities were independent but walking was rather limited. He was severely depressed. He had "a background lumbar spine spondylosis as the primary pathology and then (had) a triggered event from an incident that had triggered a chronic pain syndrome rather than a specific injury that (had) caused a secondary chronic pain syndrome .. (there was) only weak correlation between the ongoing symptoms to the accident of 27.07.06 obviously the ongoing symptoms (were) part of a chronic pain syndrome which (had) been also negatively affected by his depression." No other health conditions were identified "apart from a pre-existing lumbar spine spondylosis which was an ageing process contributing to his back pain and obviously his depression (was) contributing to his limitations". [7] On 16 August 2008 Dr John Mac Vicar, musculoskeletal pain medicine specialist, referred Mr Hannigan to Mr Nicholas Finnis, neurosurgeon, for assessment. When first reporting on 17 September 2008 Mr Finnis recorded that Mr Hannigan was in considerable distress because of pain, He said that the origin of this, which was back pain, was difficult to ascertain given that MRI imaging had not disclosed any significant nerve root impingement. [8] He proposed a provocative discogram to see the extent of disc disruption and whether any provocation would mimic the pain. He wrote again on 31 December 2008, this time to the Corporation, and, having noted as he did before that the LS/S1 disc was not normal, he said: The problem at the 15/SI disc is not just an annular tear, however the problem represents quite marked internal disruption with nuclear displacement. This is not an ageing process but is the result of abnormal loading through the disc and an acute event can cause this. The MRI scan therefore shows only subtle changes anatomically and this has led to the interpretation that this disc looked reasonably healthy. The discogram has shown that this interpretation is incorrect and the disc is actually quite disrupted and it is this which is accounting for the problems that he has been experiencing. It is my opinion that since the symptoms started with the injury he described, it is likely that he had a disruption of the LS/SI disc ... and it is this that has been causing his pain. [9] On 23 March 2009 Mr Finnis, assisted by Mr Malcolm Gordon, operated on Mr Hannigan, the description being of an: 15/S1 anterior lumbar interbody fusion with BMP and Synfix cage. It was noted that the cage had been filled with BMP (INFUSE Medtronic). [10] By 23 April 2009 Mr Finnis had written that Mr Hannigan was getting what was described as a little more pain than would be expected at that stage, that flowing down his legs, and radiating into his testicles. But Mr Finnis said that he could not detect any obvious complication. [11] Come 13 October 2009, Mr Finnis now said that Mr Hannigan was still having problems with back pain some of which radiated down the backs of his legs and that: My concern is that Mr Hannigan has not fused across the 15/S1 level. This has subsequently led to loosening of the prosthesis and the comment about ucency around the screws. It is not clear whether this alone is related to his pain or if there are other factors involved. ... Mr Hannigan is probably heading for a revision operation for his fusion if clearly things do not settle for him. ... [12] Shortly afterwards on 5 November 2009 Mr Finnis completed an ARTP for a revision of the original surgery, in particular an L5/S1 posterolateral fusion which was obviously approved by the Corporation with surgery proceeding on 17 December 2009. The operation was described as: LS/S1 posterolateral fusion with pedicl screw instrumentation. [13] Following the revision surgery Mr Finnis reported that he was happy with the fusion and that there were no complications. But by 27 January 2010 he had written to Mr Hannigan's GP to say that Mr Hannigan was having a lot of back pain post surgery, that more than would have been expected, He now wrote that: The operation was able to be achieved with limited access given that he is thin and I was happy with the technical aspect. The only problem was the porosity of the cancellous bone on the right sacral alar which made purchase of that sacral screw questionable and this was supplemented with bone cement. [14] On 22 April 2010 he wrote again, this time to say that Mr Hannigan was still having pain problems. Clinically speaking, there had not been a good response to the revision but Mr Finnis hoped that as the tissues settled down further solid fusion would be accomplished. [15] By this time Dr Mac Vicar had written that the source of Mr Hannigan's cervical spine symptoms had not been identified although negative responses to local anaesthetic blocks of the right C2-3, C5-6 and C6-7 zygapophysial joints appeared to have excluded those structures as sources of pain. [16] Dr Mac Vicar recommended a CT scan. But, as he told the Corporation on 22 July 2010, that had not revealed any diagnostic findings, he adding: There are some minor changes at the CS-6 level but they are of doubtful significance. Brent's pain has consistently been perceived in a distribution that would suggest both a lower cervical and upper cervical origin and the lower part of his pain could well be of C5-6 segment origin but there is nothing on the CT or on previous radiological imaging to explain the upper neck pain and referred pain to the right side of his head. I have already excluded the C5-6 zygapophysial joint and the lower part of Brent's pain could possibly be arising from the C5-6 disc but there is insufficient evidence to suggest that surgical intervention would be worth pursuing, particularly given the unsatisfactory response that he has had to surgery on his lower back. [17] Mr Finnis saw Mr Hannigan again on 18 August 2010, he now saying that he would arrange for further radiography to ascertain the status of the bone in the posterolateral, as well as interbody, areas. [18] On 31 August 2010 he wrote again to say that a 24 August 2010 CT scan of the lumbar spine showed fusion across the interbody device, with the screws appearing to be intact, and that flexion and extension x-rays showed stability. [19] However he did remark that the pain now described by Mr Hannigan was over the operative site where there was a sensitivity to even light palpation, So he now proposed a second opinion from Mr John Rietveld, orthopaedic and spinal surgeon. [20] A treatment injury claim was lodged by Mr Hannigan's GP in late September 2010 following which the Corporation sought further information and advice from Mr Finnis, Responding on 15 October 2010 he wrote that: Mr Hannigan had had a delayed, rather than a failed or non-union, fusion. There was no injury during surgery. Both operative procedures were technically satisfactory and although the second operation (to achieve fixation in the sacrum on the right side) required bone cement this was achieved with a recognised method. * * ** *** * **** * ***** ** The underlying nature of Mr Hannigan's pain was not entirely clear, but might be multifactorial. A delayed fusion was recognised in a proportion of people and the ongoing problems with pain were similar to those Mr Hannigan had had prior to each operation. He particularly said that "Delayed fusion is recognised in patients undergoing any fusion procedure. With the fusion induced by bone morphogenic protein the deposition of bone can be slow. Fusion eventually, however, is very good. Fusion rates vary depending on technique but with an interbody procedure can be well over 90%, however depending on the series may well be 5% to 10% of patients which fail to fuse. This can be salvaged by other techniques which, in Mr Hannigan's case, was revision posterolaterally done at 8 months because of the presented clinical condition". [21] I come now to the report of Mr Rietveld. He wrote on 1 November 2010 to say that the radiology was demonstrative of solid lower back fusion both anteriorly and posteriorly, but that Mr Hannigan nevertheless had: ... incredible hypersensitivity in his back and legs and severe pain syndrome. His view was that this ought to be dealt with by a desensitisation programme and significant pain management, not further surgery. [22] He said that he would review Mr Hannigan after further MRI, this to exclude the unlikely possibility that a recurrent disc problem had been missed. That he did on 16 November 2010, thereafter writing to Mr Finnis to say that the further MRI had not shown anything of significance. [23] He then rehearsed in firm terms his resistance to any idea of further surgery and opined that Mr Hannigan's was purely a pain management issue. Thereafter, on 13 January 2011, the Corporation made and communicated the decision now in contention. [24] Mr Hannigan's GP subsequently referred his patient for a third opinion, this time from Mr James Burn, orthopaedic surgeon. He has written more than one report but in that dated 1 February 2012 he said that: Brent's problem all stems from the fact that he had an anterior fusion with a Synfix cage at the 5/1 level and he had INFUSE put in there. ... INFUSE has been shown to cause a marked inflammatory change, bone resorption and often a vasculitis. It was never designed to be used anterioraly and this was an off label use but promoted by the reps who were selling the material here in New Zealand. He then because of his pain was explored and had posterior surgery and in my experience of approximately 130 of these cases I have never had to go to the S/1 level to a posterior instrumentation and nor have I ever used INFUSE, I use only patient's own bone which has been highly satisfactory, [25] He wanted to pursue what was described as an 'MUA/BUA lower legs' with gentle stretching under anaesthetic to monitor pain levels - the monitor to be an anaesthetist whom he identified. He thought this process would be helpful in discovering whether there was any psychological overlay. He also wondered whether there were nerve rootlets actually adherent to the dura like an arachnoiditis. [26] Mr Finnis was then asked to comment on the use of INFUSE (BMP) for the purpose of establishing a fusion, in particular on the assertion that it should not have been used anteriorly. Responding Mr Finnis wrote that: The only approved clinical use for INFUSE is in anterior spinal surgery. This remains the only FDA approved use of this product. Any other use of INFUSE has not been granted FDA approval and therefore is considered "off label" ... .. there has been a trend away from using tapered screws ... to using square cages ... supported by a screw fixation. The use of INFUSE to establish a fusion has therefore been carried on in the use of these cages, given the potential benefits and good theoretical justification for this. Because the FDA study was not in the variety of cages now specifically used, the use of INFUSE in these cages is therefore considered "off label". It is generally considered however, that with proper scientific support, surgeons may use products in an "off label" manner. ... A pain disorder caused by inflammation of the arachnoid, one of the membranes that surrounds and protects the nerves of the spinal cord. I have used, in my practice, in anterior lumbar interbody fusion operations various products to establish a fusion including ... INFUSE ... the choice of fusion material would be dependent on the clinical situation. I have been happy with the fusion rates and clinical outcome of my patients with using BMP for the fusion process with little complication. ... learning this surgical procedure I have travelled overseas to participate in this operation with other surgeons and have observed their use of and recommendation for BMP to establish a fusion. ... The use of INFUSE in PEEK cages is probably the most common way to establish fusion now in New Zealand with the operation of anterior lumbar interbody fusion. It is also my impression with workshops attended in Australia ... that this remains the most common way to establish fusion in Australia as well ... it has been of interest of ACC to establish an anterior lumbar interbody fusion registry. ... this registry has been specifically established to monitor the use of INFUSE in anterior fusion surgery, At the most recent ACC Spine Working Group meeting on 4 March 2011 the continued use of BMP in an anterior lumbar fusion was raised and there was unanimous support by the surgeons present that this should continue. ... You have asked for further comment on whether INFUSE caused bone absorption. It is a recognised potential that INFUSE can induce osteoclastic activity as well as osteoblastic activity; that is, absorption of bone as well as the deposition of bone .. the clinical significance of the vast majority of this is probably very small ... my own, and other surgeon's, clinical experience ... is that most of the radiologically observed radiolucency caused by bone resorption is clinically insignificant, commonly seen, and does not require any specific management. There is generally resolution of these changes in due course, but this can sometimes but up to a year Although you have not asked for my comment on the issue of radiculitis with the use of BMP, it has come to my attention that this has been raised in the current case. I will not address this in detail, however certain points need to be recognised. Most of the descriptions of radiculitis following the use of BMP is with posterior surgery where the passage of the substance past the nerve root or the substance in close proximity to an exposed nerve may induce this. No BMP was used in the case under consideration for this report posteriorly. With anterior surgery the BMP is bound to a collagen sponge and placed in a contained cage (PEEK separating the BMP from the disc). Furthermore, the posterior part of the disc is intact, with a retained annulus, which acts as another barrier. These structures limit radiculitis as a problem with anterior surgery. [27] He finally said: "The answer to your question "why INFUSE was used in the manner it was" is that it is indeed the only approved use for this product in spine surgery, and has been shown to cause excellent fusion rates with minimal complications. ... [28] I should add that these observations were interspersed with various, and seemingly apposite, references to medical literature and that, as the Court is aware, Mr Finnis is a highly experienced surgeon in his field. [29] Mr Burn wrote to the GP again on 4 April 2012, this time to report that a manipulation of the back under general sedation with monitoring of brain function showed there was a significant pain generation as opposed to a psychological overlay, but that some progress had potentially been made with flexibility of movement. [30] This note included the observation that: (Mr Hannigan) ended up with a lot of pain following (the surgery) and the working diagnosis is that he has arachnoiditis from his INFUSE which was used off label and this has now been reported widely throughout the world, - but no, supportive of this seemingly bold assertion, literature was cited. [31] In a later 15 May 2012 report Mr Burn rehearsed his concerns about the use of INFUSE in terms that it was possible that that had caused a chemical irritation and dural adhesions i.e. a degree of arachnoiditis. [32] He then added: In my opinion Brent has suffered a treatment injury in that his pain has been made a lot worse since he had his surgery and I will copy through all our notes ... I attended one of the spine sessions at the American Academy of Orthopaedic Surgeons in February and this topic of INFUSE was brought up. There was an incidence mentioned of between 30-40% of patients having complications as a result of this being used off-label in the interbody spinal fusion application . No guidelines to the treatment of these patients was forthcoming at that session. [33] Mr Burn subsequently prepared an ARTP dated 5 November 2012 proposing "steroid epidural with manipulation under anaesthetic", as an endeavour to "free up" No specifics as to the precise nature of the remedial fusion surgery said to have been discussed were offered by Mr Burn. nerve roots, this on the foundation of his opinion that Mr Hannigan's then current problems related: ... directly to a complication of the medications used to try and generate bone at the time of his anterior fusion and the technical aspects to the surgery themselves were not correctly carried out'. [34] A 10 December 2012 report from Workwise Wellington (completed by Dr Cheesman who specialises in occupational medicine) includes this; The recent suggestion from Mr Burn is that the current symptoms all relate to treatment injury from INFUSE causing arachnoiditis. This does not well explain the prior symptoms and is a provisional diagnosis. This has not been the conclusion of the other treatment providers involved. ... Mr Burn suggests an unconventional pain management procedure based on his review of literature. This procedure is not addressing the original covered injury (which I also do not see as established) and has not been proposed by the pain specialist involved. The evidence provided does not provide compelling support for the novel treatment provided and the likelihood of further intervention or procedures resolving symptoms given the lack of success to date with such interventions is likely to be low. ... The current surgeon suggests the presentation now relates to treatment injury from the fusion. This being arachnoiditis. This does appear well supported. The other suggestions are that this is a pain syndrome and as Dr Xiong noted at the last medical case review this may not be linked to injury ... (Emphasis added) [35] A 7 May 2013 report from Mr M MacFarland, neurosurgeon, who saw Mr Hannigan on 2 May 2013, included the observation that: ... I said that I did not think that any further direct surgery to the L5/SI level of his lumbar spine would benefit him in any way in terms of pain relief and I went on to say that I thought this was now a situation where we need to seriously consider the technique of neuromodulation/spinal cord stimulation and I think that he would be a good candidate for undertaking this procedure. With this in mind therefore I will refer him to Rick Acland and also Sandy MacLeod ... for their expert input ... 3 I again note a clear absence of any reasoned explanation or justification for that contention. Mr Taine [36] The Corporation, so I understand, was responsible for the involvement of Mr Taine, orthopaedic surgeon. His original advices were comprehensive but (bearing in mind the considerable history) reasonably contained, so I set them out in full: 38 year old man who had an anterior 15/SI fusion on 23.3.09 utilising a cage (Synfix) with a bone morphogenic protein, or BMP (INFUSE) preoperative symptoms were of back pain and left leg pain. Ongoing pain led to a CT revealing a non-union and an MRI showing oedema in the vertebral bodies of LS and SI, at the surgical site. The report stated there was no epidural or paravertebral inflammatory change. There was mention of reaction possibly secondary to the INFUSE made by the surgeon on 9.9.09. He underwent a posterior on 17.12.09 using pedicl screws and a mix of antilogous bone (from the patient) and extra bone material (progenic). Persistent back pain and symptoms in the left leg led to an MRI on 14. 10.10 with no comment on adverse reaction about the material used. There is a statement in a letter that the MRI showed "no inflammatory complication around the 15/SI level". Inflammatory markers were normal. A CT in 2010 was considered to show union anteriorly. A report by the surgeon on 15, 10.10 and a musculoskeletal physician concluded that there was no injury caused by treatment. Since then the question (has arisen) as to whether the use of INFUSE in the anterior vertebral column has led to adhesions ("like arachnoiditis": see Mr Burn's comment at 1.2.12) about the neural elements in that area. Investigations have been performed: an MRI on 30.11.11 reported "no features of bone destruction" and "no intradural abnormality". A bone scan reported mild increased uptake on the LS vertebral body and a CT on 16.5.12 reported anterior fusion but incomplete (I presume this means that there is fusion across the front part of the disc but not all of it), and no posterior here is no mention of significant resorption or collapse around the anterior column, There are a number of articles included in the file that indicate that the INFUSE product was FDA approved for use in the anterior lumbar spine, albeit with a specific metal device. A literature search around this and the other elements used in the procedures here are summarised below. INFUSE was utilised in conjunction with a Synfix device. This is made from a polymer (PEEK) supported by a titanium plate and screws. It appears that the main complication of PEEK devices in the anterior lumbar spine is non-union and resorption (Schimmel et al, J. Spinal Disord, 2010). There is no mention of neural complications. There is an article specifically addressing the use of PEEK device with BMP (Vaidaya et al. Spine, 2008). Resorption and non union are complications, but not neurologic problems. Progenix was used in the second procedure, for posterior spinal fusion. This s a demineralised bone matrix or DBM. It is a graft extender, intended for use together with the patient's own bone (autograft), as here. Its use here meets the accepted recommendations for use. Although it could elicit a local reaction, it seems unlikely that it would cause a neurologic problem. With regard to the use of INFUSE, it appears that although it was approved for use in conjunction with a particular device in the anterior lumbar spine, it has been widely used in other regions and with other devices. Although not apparent in the past, a number of complications have arisen as a result of its use, including excessive bone formation, inflammatory reactions, and neurologic problems. These complications appear to be when it is used in the cervical spine and the posterior lumbar spine. Although an advisory has been issued for this it appears that its use in the anterior lumbar spine is still authorised. A specific complication, retrograde ejaculation, appears to be more common with BMP use. This is a recognised complication of anterior approaches to the lumbar spine in males albeit rare. It is due to injury to the Autonomic nerves to the front and side of the vertebrae, not the caudaequina or nerve roots. Its increased incidence with BMP is presumably due to an issue anterior to the spinal column, not to the nerve roots Strictly speaking, its use in this case was "off-label", in terms of approval in the USA. This relates only to its use with a device other than that approved by the FDA. However, its use would seem to have been appropriate in the circumstances here. Recognised problems relate to Autonomic nerve problems, resorption and non union, not to spinal canal or nerve root adhesions. A note on arachnoiditis: this is a very specific pathology, with adhesions within the dural sac, most often associated with oil based contrast media used for Myelograms. Use of these ceased in the late 1970s. It is characterised by clumping on nerve strands in the caudaequina and is clearly seen on MRI. It is not present here. I note that Mr Burn used the term as "like arachnoiditis": the intention was not doubt to indicate that he considered the problem to be adhesions outside the dural sack and about the nerve roots to be a pain generator, but not actually arachnoiditis In relation to the claim of an adverse reaction to the INFUSE, this is not supported by the imaging available. There was clearly early oedema in the vertebral bodies, adjacent to the surgery, as one would expect. Subsequent MRIs and CTs have not indicated substantial resorption or subsidence in the anterior column. A less than complete union has now been demonstrated, which could be a pain generator, but this is a complication with a number of factors involved in its causation, including host factors such as smoking. The same comment applies to the apparent posterior lack of bone formation. Although imaging may be limited by hardware, there has been no comment on adhesions or scarring about the exiting roots at the surgical level. An MRI with gadolinium (which I do not think has been operatively) may help in showing or ruling out adhesions or fibrosis. I note that the patient's symptoms remain as similar to that experienced prior to the treatment: predominantly back and left leg pain. They are currently worse han the pre-operative level. This represents a clinical failure, in that the intention of the intervention is not met. Such a failure is a recognised risk of fusion for degenerative spinal disease. There is however no confirmation that a specific element of treatment has played a role in this. The common complications of the use of a BMP with a PEEK implant in the anterior lumbar spine are not apparent. Adhesions, fibrosis or scarring from the exiting nerve roots have not been conclusively demonstrated as caused by these materials. A less than complete union has however been confirmed. Clinical correlation as to whether this is a substantial contributor to the patient's symptoms is required. (Emphasis added) [37] Lastly from Mr Taine is an email response of 15 July 2013 referable to a communication to him from Mr Castle conveying to him the more recent medical reports recorded in this judgment. Responding to that he wrote this: After writing my last comment, I found on my arrival home that night the latest copy of the Journal of Bone and Joint Surgery. It contained a study of single level lumbar fusions using BMP - 2 (INFUSE) with a PEEK device (Micholsen et al: 2013: JBJS Am: 95A. P 873). The surgery was performed from the posterior approach, not anterior as in Mr Hannigan's case, but otherwise was the same. Only a small number of cases though, but the complications related to resorption and osteolysis with slow (but sure) union. In particular, there were no cases of radicular pain associated with the technique. It might be expected that there would be a greater risk of nerve problems with the posterior approach than an anterior one. The further information in the attachments of your email does not add substantially to the case other than the review by Mr Macfarland. I would generally agree with his formulation, I would note his recording of pre-treatment left leg symptoms, and the current state of left leg pain without a mechanical compression of nerve tissue. It appears the clinical problem is pain. Spinal surgery does not always succeed in settling pain particularly if arising from extrinsic reasons, i.e. from within the nerve root or adjacent spinal cord regions, as opposed to extrinsic reasons (compression from a disc protrusion). Adhesions and scarring will develop in many cases but is not always associated with pain. I note that to date, no significant adhesion formation has been demonstrated. There are limitations here though: the presence of metal may interfere with MRI imaging, and gadolinium has not been utilised as far as I can see. It would seem from Mr MacFarlane's report that his left leg pain is from the left S1 nerve root, but that this is predated treatment, albeit with lower level symptoms. Treatment if it played a role at all has been at most only an aggravating factor, not a cause. Extra dural adhesions is raised as a possible cause of his symptoms but not demonstrated to date. In particular, a causal link from the use of INFUSE to radicular symptoms is not able to be supported (Emphasis added) [38] I add for completeness that the medical material put before me included the file notes of Mr Burn which I have read. It is obvious from these that he is convinced that the problem lies with the INFUSE. Case for Mr Hannigan [39] The submissions provided by Mr Wadsworth came down to these: Mr Burn had identified a treatment injury arising from the surgery carried out by Mr Finnis. The working diagnosis was of arachnoiditis (inflammation of the membranes of the nerve) a physical injury caused by the treatment. Mr Finnis had not commented on Mr Burn's working diagnosis of arachnoiditis'; and Arachnoiditis, as diagnosed by Mr Burn, could not be considered a normal or necessary consequence of the treatment provided by Mr Finnis; and The difference between Mr Hannigan's presentation in February 2008 as reported by Dr Xiong in a medical case review and his condition in 2012 provided significant evidence of such a difference before and after for - Dr Xiong had reported in 2008 that Mr Hannigan was independent in basic daily living but after the surgeries he had become dependent upon his partner including for showering and getting about - he frequently using a wheelchair, [40] Mr Wadsworth went on to submit that; ... there (was) sufficient evidence to draw a robust inference on causation. While it may be a rare complication, Mr Burn had conducted a thorough investigation and his findings had not been challenged'. He considered the use of INFUSE had caused a physical injury. But Mr Taine had, and in terms discounting in reasoned terms any possibility of this being a sustainable diagnosis. Of course, the reverse is the case in this respect. See both Mr Finnis and Mr Taine. Case for Corporation [41] Mr Castle submitted that: Mr Hannigan's worsening pain syndromes arose in the context of a long history of significant neck and back pain including the consequences of a serious neck injury sustained in 1992 when cement fell on him. Dr Mac Vicar, while certainly not a specialist in spinal surgery, was well placed to comment on pain and had been involved with Mr Hannigan since 2007. His 15 November 2010 report was noteworthy for the comment; I do not believe that the surgery has caused a physical injury, rather what has happened is that surgical fusion has failed to provide Mr Hannigan with symptom relief, and there may be a link to this and non union of the fusion. in other words - no new injury, When Mr Finnis saw Mr Hannigan before any surgery he had noted (15 October 2010) that: Mr Hannigan was in considerable distress because of the pain and walked very slowly with a limp. Mobility of his back was severely restricted because of pain, ... Mr Burn (who was obviously self-convinced that INFUSE was the problem) had set out to justify his conclusion in quite general terms, this in contrast with Mr Finnis's discussion of INFUSE and likewise that of Mr Taine; so although - This really was a case where a lot was going on; nevertheless - It was a claim bound for failure, first, because the evidence did not establish a new discrete physical injury and, secondly, because the post surgery symptomology reflected that which was apparent before and any worsening in that was not shown to be causally linked to the treatment. The law [42] Section 20 says that: (1) A person has cover for a personal injury if - (a) He or she suffers the personal injury in New Zealand ...; and (b) The personal injury is any of the kinds of injuries described in section 26(1)(a) or (b) or (c) or (e); and (c) The personal injury is described in any of the paragraphs in subsection (2) (2) Subsection (1)(c) applies to - (a) Personal injury caused by an accident to the person: (b) Personal injury that is a treatment injury suffered by the person: (c) Treatment injury in circumstances described in section 32(7): Personal injury that is a consequence of treatment given to the person for another personal injury for which the person has cover: (e) Personal injury caused by work-related gradual process, disease, or infection suffered by the person: Personal injury caused by a gradual process, disease, or infection that is treatment injury suffered by the person: (8) Personal injury caused by a gradual process, disease, or infection consequential on personal injury suffered by the person for which the person has cover: h) Personal injury caused by a gradual process, disease, or infection consequential on treatment given to the person for personal injury for which the person has cover: Personal injury that is a cardio vascular or cerebrovascular episode that is treatment injury suffered by the person: Personal injury that is cardio-vascular or cerebrovascular episode that is personal injury suffered by the person to which section 28(3) applies (3) Subsections (1) and (2) are subject to the following qualifications: (a) Section 23 denies cover to some persons otherwise potentially within the scope of subsection (1); (b) Section 24 denies cover to some persons otherwise potentially within the scope of subsection (1) and (2)(e) (4) A person who suffers personal injury that is mental injury in circumstances described in section 21 has cover under section 21, but not under this section. Those circumstances have no application to this case. [43] Mr Castle contended that the truly relevant provision for the purposes of s 20(1)(c) was in fact subs (2)(d) - Personal injury that is a consequence of treatment given to the person for another personal injury for which the person has cover, [44] Section 26 deals with personal injury by saying that: (1) Personal injury means- (a) The death of a person; of ( b ) Physical injuries suffered by a person, including, for example, a strain or a sprain; or (c) Mental injury suffered by a person because of physical injury suffered by the person; or (d) Mental injury suffered by a person in the circumstances described in section 21; or (da) Work-related mental injury that is suffered by a person in the circumstances described in section 21B; or (e) Damage (other than wear and tear) to dentures or prostheses that replace a part of the human body. (2) Personal injury does not include personal injury caused wholly or substantially by a gradual process, disease, or infection, unless it is a personal injury of a kind described in section 20(2)(e) to (h). 3) Personal injury does not include a cardio-vascular or cerebrovascular episode unless it is personal injury of a kind described in section 20(2)(i) or (j). (4) Personal injury does not include - (a) Personal injury caused wholly or substantially by the ageing process; or (b) Personal injury to teeth or dentures caused by the natural use of those teeth or dentures. [45] Mr Castle reminded me of the frequently cited case of Ambros were, amongst a host of other observations, the Court said at [65] and [67]: The requirement for a plaintiff to prove means that the plaintiff must show that the probability of causation is higher than 50%. However, courts do not usually undertake accurate probabilities calculations when evaluating whether causation has been proved. They proceed on their general impression of the sufficiency of the lay and scientific evidence to meet the required standard of proof. ... 7 ACC v Ambros [2007] 1 NZLR 340 (CA) The different methodology used under the legal methods means that a court's assessment of causation can differ from the expert opinion and courts can infer causation where experts cannot. This has allowed the court to draw robust inferences of causation in some cases of uncertainty ... however, a court may only draw a valid inference based on facts supported by the evidence and not on the basis of supposition or conjecture, [46] Mr Castle very fairly referred to Harrild which had spoken of a "generous and unniggardly approach" to causation. But subsequently, in Ambros, the Court of Appeal had said that it was not appropriate to use this approach to modify the causation test. Instead it might support the drawing of robust inferences in individual cases. [47] This too from Ambros at [70]: In regards to pain: pain in itself is not a physical injury that can be covered under the ACC Act 2001. In particular the case of Teen (244/02) is directly relevant to this point. Further in the case of Goss (279/2008) Judge Beattie confirmed ... there is significant jurisprudence on whether pain and pain syndromes, such as fibromyalgia and chronic pain can be considered to come within the definition of personal injury. The clear position is that the various pain syndromes, no matter what their description, cannot be considered a physical injury in themselves, but the cover can extend to such condition if it can be established that the pain syndrome is caused by or arises out of physical injury ... Discussion [48] The first question is whether or not the probabilities favour the conclusion that in the course - or arising out - of his surgeries Mr Hannigan suffered 'personal injury'. Section 26(1)(b) says that includes 'physical injuries suffered by a person'. [49] If an individual suffers personal injury in New Zealand, including physical injury as just mentioned, and if that injury is one of those identified in s 20, then that person has cover for that personal injury. [50] The cover provisions of s 20 include in that respect (subs 2): (b) Personal injury that is a treatment injury suffered by the person: Harrild v Director of Proceedings [2003] 3 NZLR 289 (CA) c) Treatment injury in circumstances described in s 32(7); (d) Personal injury that is a consequence of treatment given to the person for another personal injury for which the person has cover; ". [51] In Allenby the Supreme Court said at [56] that: As the illustration provided by s 26(1)(b) indicates, "physical injuries" are those suffered by the claimant which have some appreciable and not wholly transitory impact on the person but which are not necessarily long-lasting or ones that cause serious bodily harm. [52] Here, and with reference to the medical evidence, I note that: The background to the events in question here is one of Mr Hannigan having a long history of significant neck and back pain substantially to be related to a serious neck injury he suffered in 1992 when some cement fell on him. Dr Mac Vicar is of the view that the surgery had not caused a physical injury: rather, what has happened is that the fusion had failed to provide symptom relief. It was Dr Mac Vicar who had earlier reported that various (including anaesthetic block) attempts to identify the reason for - the source of - Mr Hannigan's pain had been unproductive. Mr Finnis (who operated on Mr Hannigan on both occasions) reported that both operative procedures were technically satisfactory even be it the second (that achieved fixation in the sacrum on the right side) required bone cement; and - Significantly, he noted that the underlying nature of Mr Hannigan's pain was not entirely clear, could be multi-factorial, and that difficulties of delayed fusion were recognised in a proportion of those undergoing this surgery and that Mr Hannigan's ongoing pain problems were similar to those he was suffering before each of the surgeries; moreover - Complete fusion was never assured and, where there was a failure, alternative techniques (such as that eventually used with Mr Hannigan) could permit salvage. Mr Rietveld, like Mr Finnis a specialist spinal surgeon, was called in by Mr Finnis for a second opinion. He had reported what he called incredible hypersensitivity in the back and legs and a severe pain syndrome. However he did so in terms that the radiology was demonstrative of solid lower back fusion anteriorly and posteriorly and there was no suggestion from him of any injury-related explanation for that hypersensitivity and the pain syndrome. [53] Mr Hannigan relies on the views of Mr Burn, also an orthopaedic surgeon, who, as I have already shown, was and remains of the firm view that the problem lay with the use ("off label" he says) of INFUSE in the initial, which was anterior, spinal surgery. He has contended that: Mr Hannigan's problems all stemmed from an anterior fusion with a Synfix case with INFUSE, INFUSE had been shown to cause a marked inflammatory change, bone resorption and often vasculitis, and was never designed or approved for use anteriorly; and Mr Burn has 'also wondered whether there were nerve rootlets actually adherent to the dura like an arachnoiditis'. [54] Responding to these suggestions (and as I noted in full detail earlier) Mr Finnis: Allenby y Middlemore Hospital & ACC [2012] NZSC 33. Asserted that the only approved clinical use for INFUSE was in anterior spinal surgery Its use here had been "off label" in the sense only that the kind of cage used had not been one included in the FDA study reading to INFUSE approval, but there was scientific, and experiential, support for that use. In learning the procedure in question he had travelled overseas and operated with other surgeons there observing their use of, and learning of their recommendation for, INFUSE as a means of establishing fusion. He went on to say that - The use of INFUSE in PEEK cages was probably the most common way to establish fusion in New Zealand in the anterior surgery context, and that was also so in Australia. He accepted that INFUSE could cause bone absorption but this would almost always be insignificant. Then - Turning to the issue of radiculitis with the use of INFUSE, he observed that most descriptions of that followed use with posterior surgery where the passage of the substance past the nerve root or the fact of the substance being in close proximity to an exposed nerve might induce that; but No INFUSE (more often than not described by Mr Finnis as BMP) had been used with Mr Hannigan posteriorly; moreover - Physical structures limited radiculitis as a potential problem with anterior surgery; his end statement being that - The only approved use of INFUSE was in spine surgery where it had been shown to cause excellent fusion rates with minimal complications. [S5] I now come back to the words of Mr Taine who was subsequently approached for advice by the Corporation. He must be seen in this case as an independent, as well as a very experienced and highly regarded, specialist. Points of significance in his comprehensive advices included that: MRI following the first surgery did not disclose any epidural or paravertebral inflammatory change, Another post-the-second operation MRI evinced no evidence of adverse reaction and, indeed, it was specifically reported that there was "no inflammatory complication around the L5/SI level" - to the contrary, inflammatory markers were normal. There was no mention in the relevant radiology of significant resorption or collapse around the anterior column. Literature surrounding the use of a Synfix device with INFUSE indicated that the main complication in the anterior lumbar spine was non-union and resorption; there was no mention of neural complications. In the second posterior spinal fusion operation a different material, a DBM which is a graft extender, was used in terms meeting the accepted recommendations for that use. And it seemed unlikely that this could cause a neurological problem. Such problems as had arisen with the use of INFUSE appeared to have occurred when that use was in the cervical spine and the posterior lumbar spine and its use in the anterior lumbar spine still appeared authorised. Specifically on the "off label" issue: This relate(d) only to its use with a device other than that approved by the FDA. However, its use would seem to have been appropriate in the circumstances here. Recognised problems ... (did not) ... relate to ... spinal canal or nerve root adhesions. Arachnoiditis was a very specific pathology that was patently not present here and the claim of an adverse reaction to the INFUSE was not supported by the imaging. Signs of oedema in the vertebral bodies early on were to be expected and subsequent radiology had not indicated any substantial resorption or substance in the anterior column. A complete union had not been demonstrated. While that could be a pain generator it was simply a surgical complication and some number of factors could be involved in its causation; moreover The imaging offered no comment on adhesions or scarring about the exiting roots of the surgical level; and Although the symptomology was similar to (indeed worse than) that experienced prior to treatment - thus identifying a clinical failure in that the intention of the intervention was not met - that was a recognised risk of fusion; and fundamentally - There was no confirmation that any element of treatment had played a role in this; moreover - The common complications of the use of a BMP with a peak implant anteriorly were not apparent. There was less than complete union as such but no clear demonstration of adhesions, fibrosis or scarring from the exiting nerve roots. [56] Lastly (see [37] above) I rehearse Mr Taine's 15 July 2013 email response to Mr Castle arising from his consideration of a recent study of single level lumbar fusions using BMP-2 (INFUSE) with a peak device where, in fact, the approach had been posterior. [57] Only a small number of cases were apparently involved in the study in question but it had been shown that complications related to resorption and osteolysis with slow (but sure) union, and no cases of radicular pain associated with the technique had come to light. Mr Taine also said that: It might be expected that there would be a greater risk of nerve problems with the posterior approach than an anterior one... [58] In that same email, after referring favourably to the views of Mr Mcfarland, Mr Taine wrote: It appears the clinical problem is pain. Spinal surgery does not always succeed in settling pain ... Adhesions and scarring will develop in many cases but is not always associated with pain ... no significant adhesion formation has been demonstrated ... Treatment if it played a role at all has been at most only an aggravating factor, not a cause. Extradural adhesions is raised as a possible cause of the symptoms but not demonstrated to date. In particular, a causal link from the use of INFUSE to radicular symptoms is not able to be supported. [59] In overall effect Mr Taine's views, if accepted, trump the singular in this case opinions of Mr Burn. Those opinions were already sitting uncomfortably with the carefully explained views of Mr Finnis. But what renders Mr Taine's views particularly instructive comprises not only their clarity (and every sign that they are well-founded) but also that Mr Taine has no personal investment in the case. [60] While Mr Taine and, before him, Mr Finnis back up their views by reference to literature (including in the case of Mr Taine such as was very much up to date) the mark of Mr Burn's advices is their generality and the absence of adequate identification of research-based support. [61] Even when Mr Burn referred to an experience at a surgical conference in the United States he did not do so in terms descending to the sharing of any information such as might have suggested a clear connection between the subject matter of the conference discussion and the case here at hand. [62] So although, as I have quite regularly said in other judgments, there is no numbers game in cases of medical differences, the simple fact here is that Mr Burn must be identified as a lone voice whose views lack usefully identified by him substance. [63] Returning now to the observations of the Supreme Court concerning what constitutes physical injury, and applying those observations to this case, the problem is that there is no credible or reliable evidence of physical, and thus personal, injury arising during or in consequence of the surgeries, either of them. [64] That in fact means that Mr Hannigan's case falls at the first hurdle. Beyond the merely speculative observations of Mr Burn there is simply no evidential sign of surgery-related personal injury at all. [65] Commonsensical presumptive inference precepts cannot be called in aid of Mr Hanningan's plight when the requisite evidential premises are absent. There is not even a sufficiency of supportive medical evidence such as would justify the Court in taking a robust approach to the injury issue, [66] All that is evidentially apparent is that he has the same kinds of pain problems that he had before (so, as Mr Taine observed, the surgery has not effected the desired cure) and although that pain may now be worse - certainly was that for a time - pain itself is not an injury - there must be evidence of a physical cause and in this case there is none. [67] For the sake of completeness I add that had s 32 been in the picture , then for its purposes it would have also been necessary to show that the injury was not an ordinary consequence of the treatment, taking into account all the circumstances including Mr Hannigan's underlying health condition at the time of treatment, And on the evidence before me that would have been a difficult hurdle to surmount, [68] Most certainly, the mere fact that the treatment did not achieve the desired result would not have helped, for s 32(3) quite specifically says that: 10 In terms that s 20(2)(b) rather than s 20(2)(d) was engaged. The fact that the treatment did not achieve a desired result does not, of itself, constitute treatment injury. [69] So, even if s 32 had been held relevant, Mr Hannigan would have had to cross some quite significant barriers in order to gain treatment injury cover. [70] But, as I have found that the fundamental of personal injury ["physical (injury or) injuries"] is not established on the evidence, those kinds of issues do not arise. Result [71] The appeal is dismissed. Roderick Joyce QC District Court Judge