Heslop v Accident Compensation Corporation (Treatment Injury)

IN THE DISTRICT COURT AT WELLINGTON (2018] NZACC 54 ACR 266/16 UNDER THE ACCIDENT COMPENSATION ACT 2001 IN THE MATTER OF AN APPEAL UNDER SECTION 149 OF THE ACT BETWEEN DONAL HESLOP Appellant AND ACCIDENT COMPENSATION CORPORATION Respondent Hearing: 15 November 2017 Appearances: Mr Lynskey for the appellant Mr Bisley for the respondent Judgment: 27 March 2018 RESERVED JUDGMENT OF JUDGE JH WALKER [Treatment Injury s 32 Accident Compensation Act 2001] [1] This is an application for appeal pursuant to s 149 of the Accident Compensation Act 2001 ("the Act"). 2] The appellant in these proceeding is Donald Heslop and the respondent is the Accident Compensation Corporation ("the Corporation"). The appeal relates to a decision made by the Corporation dated 18 June 2015 in which it declined the appellant's cover for a treatment injury. The treatment injury in issue is an infection causing a loosening of Mr Heslop's left total knee joint replacement. [3] The matter was the subject of a review which was heard before Reviewer Ms Ritchie and was commenced on 1 June 2016 and concluded on 6 July 2016. The reviewer issued a decision on 3 August 2016 dismissing the application. Background [4] On 23 March 2008, Mr Heslop underwent a left total knee replacement surgery. At that stage, he was aged 72. [5] Mr David Walton who carried out the surgery reported that the procedure was standard and that prophylactic antibiotics were given. [6] On 24 November 2008 Mr Walton recorded that Mr Heslop was doing very well however x-rays indicated a slight lucency under the tibial tray (that is between the tibial cement and bone). [7] Subsequently on 12 November 2009 Mr Heslop had a further x-ray of his knee and Dr Andrew Gooding, radiologist, found that the prosthesis remained in an unchanged position, and the lucency between the tibial cement and bone was unchanged in extent to the previous study and measured up to 5 millimetres in size. [8] It is noted that Mr Walton on the same day recorded that the lucency under the tray had not changed since March 2009 and concluded the situation looked stable however he recommended further x-rays in 12 months time or sooner if Mr Heslop experienced difficulties with his knee. [9] On 9 December 2014 Mr Heslop saw his general practitioner Dr Michael Hobbs after experiencing a "niggle in his knee" [10] A further x-ray was carried out which stated: ... evidence of collapse of the tibial component into the proximal tibia associated with posterior and lateral tilting. Lucency around the central tibial peg ... The femoral component is unchanged in position however there may be some subtle increased lucency beneath the anterior aspect. There is a joint effusion. [11] Mr Heslop was referred to Professor Gary Hooper, professor of orthopaedic surgery. [12] In a report from Professor Hooper to Dr Hobbs he suggested that the collapse of Mr Heslop's tibia and the loosening of his femoral component could be the result of a underlying infection. Professor Hooper arranged for knee aspiration, assessment of erythrocyte sedimentation rate ("ESR"), creative Protein ("CRP") inflammatory markers, and revision surgery. [13] In his report Professor Hooper also recorded: ... Don had his knee replaced in about 2008 by Dave Walton and says he has never really been very happy with it. It has always given him pain and I note Dave has been taking sequential xrays of his knee and there was a very early appearance of a radiolucent line under his tibia which has progressed. 14] Five pathology reports were obtained dated from 12 December 2014 to 9 January 2015. The pathology reports were not indicative of an infection. However, Professor Hooper continued to express an opinion that "loosening this early following his procedure is highly suggestive of infection". [15] On 16 January 2015 Professor Hooper performed a revision left total knee replacement. At the time of the operation, nearly 7 years after the first operation, Mr Heslop was aged 78. [16] Professor Hooper noted in a report to Dr Hobbs that both knee compartments were grossly loose and there was no overt infection. He had sent multiple samples for assessment to try to establish the diagnosis. [17] The operation note of 16 January 2015 notes that both the tibial and femoral components were grossly loose, but there was no overt infection or pus. Professor Hooper prescribed a two-day course of antibiotics. [18] Professor Hooper followed up with the appellant on 23 January, 26 January, 30 January, 2 February, 2 March and 9 March 2015. He recorded that overall the appellant recovered satisfactorily from surgery. [19] Dr Hobbs, GP, lodged a treatment injury claim on behalf of the appellant. The respondent states that on 5 May 2015 Dr Hobbs lodged a treatment injury claim on behalf of Mr Heslop. However, it would seem that in fact cover was not separately applied for and the date of the receipt of the application appears to be 1 1 May 2015. [20] The treatment injury claim form sought treatment of the 28 March 2008 left total knee joint replacement and stated: Underwent (L) total knee replacement injury in 2008 developed discomfort in (L) leg in 2009 and was noted to have lucency under tibial tray on xr in 2008 and subsequently. [21] The injury is noted as "collapse of (L) total left knee replacement. [22] On 5 June 2015 Professor Hooper provided a report to the Corporation expressing an opinion that the loosening was most likely caused by an infection at the time of the initial surgery. [23] He advised that at the time of revision they had taken numerous samples around his knee but there was failure in growing any organism. His working diagnosis continues to be infection as there is no other obvious cause for his problem. [24] On 15 June 2015 Mr Taine, orthopaedic surgeon and medical advisor to the Corporation, provided a report stating that a low-grade infection was certainly possible. He states however: The presence of the described cells is consistent with the response to wear rather than infection. I would therefore disagree that infection is a probable cause of the loosening in this case. It is possible but cannot at this stage be regarded as probable, based on the above considerations. [25] On 18 June 2015, the Corporation declined Mr Heslop's claim for cover and he subsequently filed an application for review on 17 July 2015. [26] On 6 November 2015 Mr Taine provide further comment describing the nature and possible causes of loosening (including low grade infection, malalignment, inadequate cementing and body wear). He concluded: ... there is a physical change ("injury") caused by the loosening, but not a causal link to treatment. [27] Professor Hooper produced a further report dated 15 December 2015. His diagnosis continued to be the failure of the knee replacement secondary to an unrecognised infection on the basis that Mr Heslop had developed a "significant" radiolucent line around his tibial component within "a few years" of the surgery. [28] He also noted: The only other comment I would like to make is that sometimes it can be extremely difficult to make the diagnosis of infection following a hip or knee replacement, particularly an indolant infection with an organism which is extremely difficult to grow in the laboratory. [29] The Corporation referred the claim to Mr Pai, independent assessor for the Corporation and orthopaedic surgeon, who produced a report on 28 February 2016. He expressed an opinion that the infection did not cause the loosening, taking into consideration the ESR, CRP, aspiration and histology results. He said that 90- 95 percent of total knee replacements survive for ten years but there were numerous causes for early failure. He stated that: The recent evidence based literature suggests that there are newer investigation tools such as sonification for the prosthesis that can yield more positive results as to whether there is any infection. [30] On 30 March 2016 Professor Hooper provided a third report responding to Mr Pai's. He referred to a sonification study in Christchurch in which infection was found in over 15 percent of previously unrecognised aseptic cases (that is, in which the diagnostic testing did not identify an infection). He stated: However, the fact that the samples do not grow an organism, as mentioned previously, does not rule out the fact that the implant maybe seeded with an unusual organism which is difficult to grow in the laboratory. [31] He also states: However, there is also no doubt that even in the presence of low inflammatory markers and negative aspirates infections can still be present. [32] He stated that the appellant's BMI of 32 was not excessive. [33] On 30 June 2016 Professor Hooper produced a fourth report which stated that although the implant was placed correctly during surgery, it had subsequently shifted into varus malalignment at some undetermined point. It is significant that all three medical experts agree that varus malalignment may cause aseptic loosening. [34] The review was heard on 1 June 2016 and subsequent dates by Reviewer Ms Ritchie. [35] Professor Hooper's last report became available during the course of that hearing. [36] Reviewer Ritchie released her decision on 3 August 2016 dismissing the application for review. [37] On 8 January 2017 Mr Pai produced a second report for the Corporation responding to Professor Hooper's third and fourth reports which confirmed his earlier opinion that the failure of Mr Heslop's total knee replacement was caused by aseptic loosening [38] He said that the presence of both femoral failure and tibial failure is not indicative of an infection, and cited medical studies in support of that proposition. Instead, he suggested that such a failure was caused by varus malalignment or failure to osseointegrate, resulting in stress concentration and accelerated poly wear. The Law [39] The matter before the Court relates to a treatment injury. This is set out under s 32 of the Act which states: 32 (1) Treatment injury means personal injury that is- (a) suffered by a person- (i) seeking treatment from 1 or more registered health professionals; or (ii) receiving treatment from, or at the direction of, 1 or more registered health professionals; or iii) referred to in subsection (7); and (b) caused by treatment; and (c) not a necessary part, or ordinary consequence, of the treatment, taking into account all the circumstances of the treatment, including- (i) the person's underlying health condition at the time of the treatment; and (ii) the clinical knowledge at the time of the treatment. (2) Treatment injury does not include the following kinds of personal injury: (a) personal injury that is wholly or substantially caused by a person's underlying health condition: b) personal injury that is solely attributable to a resource allocation decision: C) personal injury that is a result of a person unreasonably withholding or delaying their consent to undergo treatment. (3) The fact that the treatment did not achieve a desired result does not, of itself, constitute treatment injury . .. [40] Section 33 of the Act defines the word "treatment" which states: 33 Treatment (1) For the purposes of determining whether a treatment injury has occurred, or when that injury occurred, treatment includes- (a) the giving of treatment: (b) a diagnosis of a person's medical condition: (c) a decision on the treatment to be provided (including a decision not to provide treatment): (d) a failure to provide treatment, or to provide treatment in a timely manner: (e) obtaining, or failing to obtain, a person's consent to undergo treatment, including any information provided to the person (or other person legally entitled to consent on their behalf if the person does not have legal capacity) to enable the person to make an informed decision on whether to accept treatment f the provision of prophylaxis: the failure of any equipment, device, or tool used as part of the treatment process, including the failure of any implant or prosthesis except where the failure of the implant or prosthesis is caused by an intervening act or by fair wear and tear), whether at the time of giving treatment or subsequently: h) the application of any support systems, including policies, processes, practices, and administrative systems, that- (i) are used by the organisation or person providing the treatment; and ii) directly support the treatment. (2) Subsection (1) does not affect the application of the definition of treatment in section 6(1) for purposes other than those stated in subsection (1). (3) Subsection (2) is for the avoidance of doubt [41] Personal injury is defined in s 26 and includes: 26 Personal injury (1) Personal injury means- the death of a person; or 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 injuries 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 personal injury of a kind described in section 20(2)(e) to (h). (3) Personal injury does not include a cardiovascular or cerebrovascular episode unless it is personal injury of a kind described in section 20(2)(i) or (j). Position of the Appellant [42] Submissions were filed by Anna Kokje, counsel, on 3 November 2016 and counsel spoke to the submissions at the hearing. In addition, counsel filed submissions in reply on 21 March 2017. Mr Heslop in particular relies on the reports from Professor Hooper who also undertook the surgery. [43] Professor Hooper's first report of 12 December 2014 in particular states: .. Don had his left knee replaced in about 2008 ... His tibia now has collapsed medially, his femoral component is grossly loose. This really looks like an underlying infection ... It is likely we will need to involve the Infectious Disease Department as well. [44] Counsel notes the report provided by Professor Hooper to ACC was dated 30 March 2016. Counsel states: 14. Professor Hooper explained that, if the prosthesis is placed in varus, then ibial loosening is a well recognised complication. However, femoral loosening is not a common complication of the prosthesis being placed in varus. He therefore explained that the fact of Mr Heslop having both tibial and femoral component loosening is not consistent with loosening due to varus orientation. This is particularly so given that the prosthesis wasn't placed in varus. Mr Pai disagrees as his opinion is that "whether it is tibial loosening or femoral loosening" it can still be explained on the basis of aseptic loosening. However, it is submitted that, as Professor Hooper has explained, this is a clinical finding that supports the presence of an infection. [45] Professor Hooper's report of 16 January 2015 reported there was no overt infection and that he had proceeded with surgery being "revision left total knee replacement". [46] In his clinical notes (p 21 BofD) Professor Hooper noted that Mr Heslop developed some cellulitis post operation in his distal tibia and some and some weeping areas in his knee. He stated: I want him to be on antibiotics to avoid any infection of this implant so I have started him on Flucloxacillin today. [47] On 5 June 2015 Professor Hooper provided a further report to the Corporation stating: I understand that this is contentious and I understand that it is difficult to make a definitive diagnosis in these cases but on the balance of probabilities it is more likely that this knee has become infected or was infected right from the time of surgery. It is likely that the organism was a low grade organism which is notoriously difficult to culture. The things against infection are the fact that his infected parameters have never been significantly raised. [48] It was noted that Mr Taine for the Corporation provided a report of 15 June 2015 acknowledging low grade infection was possible but was of the view that other indications and histology were not indicative of infection and that figured with longevites being probabilities only and that some knee replacements do statistically, loosen earlier than the average. [49] Subsequent to this the Corporation reached its decision by letter to the appellant on 18 June 2015. This states: We're sorry, we can't approve your claim. Thank you for your patience while we assessed your claim for a treatment injury. After careful consideration, we're sorry to say your claim has not been approved. Unfortunately, this also means we are not able to help with the cost of treatment or other support for this injury. Why we can't approve your claim We're unable to approve your claim because it does not meet the criteria for a treatment injury. [50] Included was the Treatment Injury Report, which refers to the report of Professor Cooper and also the view of Mr Taine. The report concludes: For ACC to accept a treatment injury claim for cover, there must have been clinical evidence of a physical injury caused by treatment which will then be assessed against the remaining criteria for treatment injury, for example, whether an injury is a necessary part or an ordinary consequence of treatment. After assessing the available information, ACC declines the claim for cover as there has been no definite diagnosis established in this case. [51] The basis of Mr Heslop's Application for Review, dated 17 July 2015 states: The claimant disagrees with the ACC decision on the basis that they suffered an injury which should be covered. [52] This led to the further report from Mr Taine dated 6 November 2016 and two further reports from Professor Hooper. [53] Professor Hooper provided a report dated 15 December 2015 at the request of Counsel for Mr Heslop. Professor Hooper explained the following: As mentioned, my diagnosis continues to be failure of his knee replacement secondary to an unrecognised infection. The reason why I continue with that diagnosis is that this gentleman had his knee replacement in situ for only a few years before he started to develop significant radiolucent line around his tibial component. The longevity of the type of knee replacement this gentleman had in situ, particularly the cemented from of this implant is approximately 95% survival at 15 years (Christchurch figures), and so it is unlikely that this would fail because of aseptic loosening this early. ... The only other comment I would like to make is that sometimes it can be extremely difficulty to make the diagnosis of infection following a hip or knee replacement, particularly an indolant infection with an organism which is extremely difficult to grow in the laboratory ... [In part and emphasis added]. [54] Subsequent to Professor Hooper's 15 December 2015 report a report was obtained by the Corporation from Mr Vasudeva Pai responding to Professor Hooper's reports. Excerpts of his report are found in Counsel's original submissions. Mr Pai opining: ... In my opinion, infection is more than likely not the cause for the failure in Mr Heslop's case considering the normal CRP, normal ESR, fluid aspiration being negative for infection, and multiple histology take at the time of surgery that is a gold standard for infection also being negative. The recent evidence-based literature suggests that there are newer investigation tools such as sonification for the prosthesis that can yield more positive results as to whether there is any infection ... ... In Ritter's opinion the important factor for failure is such situations is a varus orientation of total knee, and with any varus less than 2 degrees the predication of failure is quite high, and the second important aspect of failure is a BMI of more than 33.7 ... ... In my opinion, considering all these factors, although there is suggestion of a possibility or unrecognised infection as a causation of failure of total knee replacement in Mr Heslop's case I cannot conclude this with the present available technology in New Zealand, and even if the sonication techniques are available, only 10 percent of aseptic loosening can be diagnosed as septic loosening. Therefore, 90 percent still remains as an aseptic loosening ... [In part and emphasis added]. [55] Counsel refers to Professor Hooper's response to Mr Pai's report in his next report of 30 March 2016. In Counsel's submissions excerpts of Professor Hooper's report are referred to as follows: Mr Pai may not be aware of the significant sonification study done in Christchurch where two hundred patients were reviewed where their implants were sonificated and an incident of over fifteen percent demonstrated infection in previously unrecognised infected cases. There are further studies within the literature which suggest up to sixty to seventy percent of cases of aseptic loosening in fact may not be aseptic loosening but maybe related to an organism ... ... the fact that the samples do not grow an organism, as mentioned previously, does not rule out the fact that the implant maybe seeded with an unusual organism which is difficult to grow in the laboratory ... . . . .. As previously mentioned if there was a varus orientation of the tibial component I would expect to see tibial loosening not tibial and femoral loosening. Mr Heslop was not grossly obese and my assessment of the initial xrays was that the alignment was satisfactory ... ... As previously mentioned in my first report, it is difficult to categorically say that this gentleman's implant was infected, particularly in view of the fact that his CRP, ESR and aspirates were all negative. However, there is also no doubt that even in the presence of low inflammatory markers and negative aspirates infections can still be present. I believe that to make this diagnosis we have to look at the overall clinical situation, the time span and how the component loosening has developed. The points in favour of infection in this gentleman are the short time span to development of radiolucent line around the tibia to the extremely short time space for both femoral and tibial components to be grossly loose. This scenario is rarely seen in cases of aseptic loosening ... [Bold emphasis added (by counsel)]." [56] The review hearing before Ms M Richie took place on 2 June 2016 and concluded on 6 July 2016, dismissing the application. [57] In the course of the hearing on 3 June 2016 the Corporation provided further comment sought in a letter dated 30 May 2016 to Professor Hooper. Professor Hooper, responding to the questions in the letter, stated: To my knowledge no sonification study was done on this gentleman's implant. We have done a trial of sonification in Christchurch and found that overall results of sonification of implants does not improve the clinical diagnosis. This would be in keeping with more recent literature which would confirm that. The varus angulation of this gentleman's knee has developed since his immediate post operative xrays in 2008. His immediate xrays looked well aligned in valgus with no obvious abnormality underneath the tibial component but have subsequently shifted into a varus malalignment and subsequent loosening. . .. As I have previously mentioned, my main concern for the explanation for this was infection. Certainly patients elevated BMI can be a contributing factor, although with a BMI of approximately 32 that is not excessive. [58] He concluded: I really have no explanation for his rapid failure of his joint replacement and hat is really the underlying reason why this has been applied for through ACC is it is an unusual and unexpected outcome following this type of surgery, particularly when the implants have been implanted in a satisfactory position. [59] Subsequent to the review decision of 3 August 2016, a notice of appeal was filed by John Miller Law on behalf of the appellant dated 17 July 2015 stating that the decision was wrong in fact and law. Counsel notes underlying(sic) is original emphasis. [60] In written submissions related to the appeal, counsel refers causation in ACC cases and specifically to ACC v Ambros,? where the Court of Appeal laid down the test for causation in Accident Compensation context. [61] Counsel states: 29. Accordingly, the totality of the Court's reasoning laid down a three-stage enquiry for causation. 30. First, it must be asked: "but for" the alleged cause, would the injury or condition have resulted? This can otherwise be expressed as the factual causation enquiry. 31. Secondly, it must be asked: does the injury or condition fall within the appropriate scope of liability? It was noted that this enquiry would identify and consider policy considerations arising generally in the accident compensation context. 32. Thirdly, there is an enquiry into the proximity or remoteness of the damage to the cause, which is often assessed under stage two. It was noted that this third stage is likely to be more important in accident compensation cases. [62] Counsel highlights that it was accepted that some measure of uncertainty in the medical context was inevitable. He refers to paragraph [53] of the Ambros decision which states: Uncertainty can arise because of the objective limitations of scientific medical knowledge about a particular biological process (particularly where there may be multiple possible causes) or from the difficulty of providing a scientific explanation for the sequence in a particular case. While a causal relationship may exist between a possible cause in a certain percentage of cases in the population as a whole, it may be impossible to say whether it causes the condition in the case at hand. [63] Counsel continues: 36. The Court further noted that, when weighing the evidence, the probability of causation is not usually assessed in percentage terms. Rather, the Courts "proceed on their general impression of the sufficiency of the lay and scientific evidence to meet the required standard of proof". 37. It was said that the legal approach to causation differs from the medical or scientific approach, in that the "legal method looks to the presumptive inference which a sequence of events inspires in a person with common sense" 2 Accident Compensation Corporation v Ambros [2007] NZCA 304; [2008] 1 NZLR 340. 38 This distinct methodology allows the Courts to draw inferences of causation in circumstances where a medical or scientific approach cannot. Accordingly, in uncertain cases, the Courts can draw a robust inference on causation, albeit not on the basis of supposition or conjecture. [64] Counsel also states: 41. The final comment made by the Court was to suggest that the "generous and unniggardly" approach advocated in Harrild v Director of Proceedings may support the drawing of inferences in individual matters of causation in the accident compensation context. [65] Accordingly, counsel submits that the appellant had sustained a treatment injury in that: [a] He has sustained a personal injury, being an infection of his knee replacement leading to a loosening of his knee replacement; [b] The treatment, being a total knee replacement in 2008, was provided by a registered health professional; [c] The infection and the loosening is causally linked to the surgery, and [d] The infection and the loosening was not an ordinary consequence of the surgery. [66] In support of the position that Mr Heslop has sustained a person injury, being an infection and loosening of his knee replacement, Counsel's submission is that the crux of this appeal is the importance that is placed on the diagnostic testing that Mr Heslop underwent, that is, whether the clinical diagnosis of infection is proven to the balance of probabilities, notwithstanding the results of the diagnostic testing. (67] Counsel states that if Professor Hooper's diagnosis of infection is accepted it follows from the medical opinions on file, that the infection caused the loosening. Accordingly, Counsel focuses on the infection diagnosis. [68] Counsel states Professor Hooper has consistently considered that Mr Heslop developed an infection and within days of the referral, Professor Hooper had found that "this really looks like [an] underlying infection" and that surgery was urgently needed. [69] Counsel states: Secondly, the possibility of an infection based on all of the evidence, including the diagnostic testing, is accepted by the ACC's advisors. [70] Counsel refers to Mr Taine's comment in support that "a low-grade infection acquired at the time of his initial surgery is certainly possible ... but cannot at this stage be regarded as probable". Counsel states that Mr Pai also accepts that the infection is possible but states "I cannot conclude this with the present available technology in New Zealand" Counsel states thirdly that there is further common agreement between Professor Hooper, Mr Taine and Mr Pai that the testing that the appellant underwent is not determinative of infection because it is unreliable. [71] Counsel refers to the fact that Professor Hooper has explained a number of times that this is because the organisms in the circumstances are "notoriously difficult to culture" and that "the fact that we cannot grow an organism does not rule out this diagnosis". [72] Counsel refers to the fact that Mr Pai agrees, and explained that there is a study of subjects who had been diagnosed with aseptic loosening, i.e. "failure of the bond between an implant and the bone in the absence of infection". [73] He refers to Mr Pai's explanation of a study of subjects diagnosed with aseptic loosening where a newer form of technology (sonification) found that, in fact, 10 percent of them had an infection. [74] Counsel, in his submissions (para [52]), refers to the factors that Professor Hooper said influenced his opinion that there was in fact an infection: The first factor is the simple fact that no one has been able to identify another cause for the loosening. Whilst Mr Taine lists a number of other possible causes for Mr Heslop's knee condition (including "malalignment, inadequate cementing technique, third body wear, to name a few"), there is no evidence to suggest that Mr Heslop's condition has been the result of these other causes. Indeed, Mr Taine does not suggest that another cause is apparent on the evidence, which includes the surgical findings, but merely lists other causes. [75] Counsel submits that neither Mr Taine nor Mr Pai are able to identify any other probable cause. While Mr Pai theorised that the varus orientation of Mr Heslop's knee may have been an important factor in the failure of the knee replacement, Professor Hooper stated he never identified varus orientation in Mr Heslop's knee. [76] Counsel states: There is, therefore, not a plausible cause for the loosening of Mr Heslop's knee replacement other than an infection. Whilst it is accepted that the onus is on Mr Heslop to prove that there was an infection, the fact that there is no other plausible diagnosis assists that onus. The relevance of the fact that there is no plausible alternative cause was noted by His Honour Judge Joyce in Brash: 3 [83] But here - in the absence in my view of any truly competing explanation for what happened - and in the presence of medical evidence properly treatable as supportive, and a factual narrative that fits in that respect, the fact that the medical profession equally (in fact more commonly) recognises the possibility of a spontaneous event, so I find, of no significant movement. [Emphasis added] [77] The second factor that Counsel for Mr Heslop relies on is the presence of a radiolucent line under the tibial implant in 2009 which is indicative of infection. It is counsel's view that all three of the medical witnesses are in agreement. [78] In respect to the third factor, counsel states: 58. The third factor is the prematurity of the failure of the replacement. In fact, Professor Hooper categorically says that "to see gross loosening of gross components within six years of a knee replacement means that almost certainly the implant is infected. "Mr Pai agrees that early loosening is more often caused by hidden [periprosthetic joint infection] than late loosening". [79] Counsel refers to the Corporation's medical evidence based on Mr Pai's reasoning, that 90 percent of diagnoses of aseptic loosening are upheld even with 3 Brash v ACC [2013] NZACC 23. further investigation and it is therefore more likely that Mr Heslop did not have an infection. [80] Counsel questions this reasoning on the basis: First it is questionable that the appellant has ever been diagnosed with aseptic loosening. Reliance on statistics is an insufficient basis for disputing causation. It would appear the ACC is seeking a degree of certainty that is not required in that they have ostensibly considered there must be positive laboratory results to prove an infection. However, it is difficult to reconcile this position with ACC's advisors' acceptance that the laboratory results are unreliable. [81] Counsel also refers to the fact that it must be borne in mind that Professor Hooper is Mr Heslop's treating surgeon and as such he has had an opportunity to assess Mr Heslop on a number of occasions as well as to perform the revision surgery. It is noted he is also a professor of orthopaedic surgery and his clinical practice is largely confined to adult elective surgery, the majority being joint replacement surgery. [82] Counsel also refers to a number of Professor Hooper's publications in this area. [83] Counsel states Mr Heslop's infection and loosening was not an ordinary consequence of surgery. [84] He refers to the decision of Holden v ACC at paragraph [44] which states: An "ordinary consequence" encompasses unintended consequences or side effects that are expected to occur in numbers of cases. Consequences that are rare would not be ordinary consequences, but there will inevitably be a range of consequences that are more difficult to classify. [2012] NZACC 170. [85] In his submissions, Counsel also refers to the case of ACC v McEnteers which states (at para 73): It is apparent that difficulties in establishing the cause of injury claimed to have been caused by "medical misadventure" contributed to the cost of claim processing, and arbitrary outcomes. Hence Parliament's intention to move away from a definition that involved any need to enquire into responsibility for an injury. The scope of cover was to be for the unanticipated adverse outcomes arising from treatment. The comment on the definition of "treatment injury" in the notes accompanying the introduction specified that it will not cover injuries that are an anticipated part or consequence of the treatment, such as a surgical incision during an operation, or which result from the claimant's underlying health condition. [86] Counsel submit (at para 75): It is submitted that a practical and common sense approach should be adopted when applying the treatment injury provisions under the Act. In this case, the question is whether there were consequences of Mr Heslop's total knee replacement that were not ordinary. When considering the answer to this question, it cannot be said that an infection and loosening of a knee replacement is an ordinary or anticipated consequence of a knee replacement. [87] In the Conclusion, counsel submits this is essentially a case as to diagnosis, with the Corporation contending that the loosening is probably not caused by infection although there is no evidence supporting any other cause. [88] Counsel submits that as Professor Hooper was the treating surgeon in the revision operation, there must be compelling reasons to set aside Professor Hooper's opinion, particularly regarding diagnostic issues. The Position of the Respondent [89] Mr Bisley, counsel for the Corporation, also filed submissions prior to the hearing and spoke to these at the hearing. [90] He states in his written submissions that to have cover for a treatment injury Mr Heslop must prove that he has suffered a personal injury caused by treatment, and the injury was not a necessary or ordinary consequence of the treatment. 5 HC Wellington CIV-2008-485-1800, 1 December 2008 at [17] per Dobson J. [91] Counsel notes that in terms of ss 26(1)(e) and 33(1)(g) the failure of an implant or prothesis caused by wear is not a personal injury or treatment under the Act (see Para [40] above). [92] Counsel states: Therefore the essential question in this appeal is the cause of the failure of Mr Heslop's total knee replacement. [93] In his submissions (para 1.4), Counsel goes on to refer to the fact that there are two competing theories as to cause: (a) Mr Heslop submits that the failure was caused by an infection arising out of the surgery in which the total knee replacement was fitted. Although there is no diagnostic evidence of an infection, Mr Heslop relies on the opinion of Mr Gary Hooper, orthopaedic surgeon, who considered that infection was the most likely cause given the timeframe in which the total knee replacement failed. b) The Accident Compensation Corporation (ACC) submits that it is more likely that the failure was caused by wear. ACC relies on the diagnostic evidence, statistical evidence, and the opinions of two orthopaedic surgeons, Mr Bill Taine and Mr Vasu Pai, both of whom concluded that it was unlikely that Mr Heslop suffered an infection. 94] Counsel transverses the background, including the reporting letter from the original surgeon Mr Walton, to Mr Heslop's general practitioner of 28 March 2008. This reports that the procedure was standard and that prophylactic antibiotics were given (bundle of documents (BofD) page 1). [95] Counsel notes that on 24 November 2008, some months later, x-rays indicated a slight lucency under the tibial tray (that is between the tibial cement and bone) but x-rays a year later showed no change to the lucency or the position of the prosthesis. The plan was to review Mr Heslop at 12 months or sooner if necessary. [96] It was noted that it was some six and a half years later after the initial surgery that Mr Heslop first presented to his GP after experiencing "a niggle in his knee". At that stage, the x-rays revealed: ... evidence of collapse of tibial component into the proximal tibia associated with posterior and lateral tilting. Lucency around the central tibial peg ... the femoral component is unchanged in position however there may be some subtle increased lucency beneath the anterior aspect. There is a joint effusion. [97] As noted, Dr Hobbs referred Mr Heslop to Professor Hooper who met with him on the first time on 12 December 2012 and reported back suggesting that the collapse of Mr Heslop's tibia and the loosening of the femoral component could be the result of an underlying infection. [98] Steps were taken but the pathology results were not indicative of infection. Notwithstanding, counsel notes that Professor Hooper continued to think that "loosening this early following his procedure is highly suggestive of infection". [99] Subsequently on 16 January 2015, Professor Hooper carried out a single stage revision surgery. [100] There was a report made to Dr Hobbs dated 16 January 2015 which noted that both the tibial and femoral components were grossly loose, but there was no overt infection or pus. It was also noted that Professor Hooper prescribed a two-day course of antibiotics. [101] Counsel states at 2.6 of his submissions: Multiple samples (taken from beneath the prosthesis and throughout the knee joint) were sent for histology and microbiology. None of those samples revealed any infection. [102] A Treatment Injury ACC claim was lodged by Dr Hobbs on 4 May 2015, some 5 months after the single stage revision surgery, describing the injury as "collapse of [left] total knee joint replacement". It also refers to Mr Heslop having osteoarthritis in his left knee. [103] As noted, on 5 June 2015 Professor Hooper provided a report to the Corporation and expressed the opinion that the loosening was most likely caused by an infection at the time of the initial surgery. [104] He acknowledged that "infected parameters have never been significantly raised". Counsel states: However, he [Professor Hooper] considered that an infection was the most likely cause of the loosening. Because the detailed inquiries he had carried out had not returned any evidence of infection, that conclusion was based on only two facts; first, the relatively short period in which the total knee replacement failed (six and a half years), and secondly, the apparent absence of any other "obvious cause". [105] Counsel refer to Mr Taine, an orthopaedic surgeon and medical advisor to the Corporation, who provided a report dated 15 June 2015 where he noted that infection was a possible but not probable cause of the loosening. Counsel stated: The early note of a radiolucent line beneath the tibial implant and the loosening of the implant before the expected lifespan were suggestive of an infection. However, the stable nature of the radiolucency, the low ESR and CRP, the absence of identification of infection on the aspirate and multiple surgical specimens, and the histology were not indicative of an infection. The figures for longevites are probabilities only: some knee replacements do, statistically, loosen earlier than the average. [106] Therefore, Counsel notes that Mr Taine gave an opinion that the loosening was more likely due to wear. [107] Counsel notes that subsequent to ACC declining cover Mr Taine provided a further report of 6 November 2015 for the purpose of the review in which he describes the nature and possible causes of loosening (including low grade infection, malalignment, inadequate cementing and body wear). At this stage Mr Taine concludes: There is a physical change ("injury") caused by the loosening, but not a causal link to treatment. [108] Counsel also refers at paragraph 2.12 of his submissions to Professor Hooper's second report of 15 December 2015 where his diagnosis revalued the failure of the knee replacement secondary to unrecognised infection, on the basis of Mr Heslop developing a "significant " radiolucent line around his tibial component within "only a few years" of the surgery. [109] However, Counsel states: Notably, this diagnosis appears considerably to overstate the evidence: as explained at paragraph [2.2] above, the contemporaneous x-ray reports refer to a 's[l]ight lucency'. [110] At 2.13 of his submissions Counsel refers to Mr Pai, an independent orthopaedic surgeon, who provided a report to the Corporation. Mr Pai expresses an opinion that the infection did not cause the loosening, taking into consideration the ESR, CRP,' aspiration and histology results. [1 1 1] Counsel refers to Mr Pai's report and states: He said that 90-95% of total knee replacements survive for 10 years, but there are numerous causes of early failure. These causes include progression of radiolucency under the tibial tray or medial bone collapse caused by varus alignment of the total knee, a high body mass index (BMI) and poly wear. Mr Pai cited a study in which the presumed infection accounted for only 10% of cases in which the clinical markers indicated aseptic loosening [112] At 2.14 counsel refers to Professor Hooper's third report responding to Mr Pai's report where Professor Hooper refers to a sonification study in Christchurch in which infection was found in over 15 percent of previously unrecognised aseptic cases (that is, cases in which the diagnostic testing did not identify an infection). [113] Counsel states that Professor Hooper also refers to other cases (not cited or provided) in which the incidence of infection was 60-70 percent. Counsel also refers to the fact that Professor Hooper also said that the loosening of both the tibial and femoral components due to varus malalignment was unlikely in cases of aseptic loosening and that Mr Heslop's BMI of 32 was not excessive. [1 14] In his submission, Counsel also refers to Professor Hooper's fourth report of 30 June 2015, where he said that although the implant was placed correctly during surgery, it had subsequently shifted into varus malalignment at some undetermined point. It is significant that all three medical experts agree that varus malalignment could cause aseptic loosening. Counsel refers to the third Hooper report of 30 March 2016, ESP is an abbreviation for Erythrocyte Sedimentation Rate, a blood test that detects and monitors nation in CRP is an abbreviation for C-Reactive Protein Test - A plasma protein which rises in the blood with inflammation. the second report of Mr Taine of 15 December 2016, the first report of Mr Pai dated 28 February 2016 together with a second report from Mr Pai dated 8 January 2017. [115] Counsel also refers to the second report of Mr Pai of 8 January 2017 responding to Professor Hooper's third and fourth reports. Counsel states that Mr Pai confirmed his earlier opinion that the failure of Mr Heslop's total knee replacement was caused by aseptic loosening. He pointed out that even in the sonification report referred to by Professor Hooper (and indeed Mr Pai in his first report) presumed infection accounted for only 10/15 percent of cases and that the remainder were of cases of aseptic loosening. [1 16] Counsel also states, referring to Mr Pai's report of 8 January 2017: Mr Heslop was not treated like an infected patient in 2014 (that is, only a single-stage revision surgery and a two-day course of antibiotics was required). Mr Pai said that the presence of both femoral failure and tibial failure is not indicative of an infection, and cited medical studies in support of that proposition. Instead, he also suggested that such failure was caused by varus malalignment or failure to osseointegrate, resulting in stress concentration and accelerated polywear. [1 17] In respect of the Law counsel makes the following submissions: Treatment Injury Provisions 3.1 Mr Heslop seeks cover for a "treatment injury". The onus is on Mr Heslop to prove, on the balance of probabilities, that all of the requirements for cover are met. 3.2 A "treatment injury", as defined in section 32, means a "personal injury" that is: (a) suffered by a person receiving or seeking treatment from 1 or more registered health professionals; and (b) caused by "treatment"; and (c) not a necessary part or ordinary consequence of the treatment, taking into account all the circumstances of the treatment and the clinical knowledge at the time of the treatment. 3.3 The fact that the treatment did not achieve a desired result does not, of itself, constitute a treatment injury. In particular, the failure of treatment does not necessarily mean that treatment caused an injury. 3.4 The issues in this appeal are whether Mr Heslop suffered a personal njury, and, if so, whether that injury was caused by treatment. The remaining aspects of the test follow from the conclusions to those issues, which are therefore determinative. Personal injury 3.5 "Personal injury" is defined in section 26 and includes, relevantly: (a) Physical injuries suffered by a person, including a sprain, or a strain. The Supreme Court has defined physical injury as bodily harm or damage having some appreciable and not wholly transitory impact on the person. Mere pain or physical change is not a physical injury. (b) Damage (other than wear and tear) to prostheses that replace a part of the human body. There is limited judicial or legislative guidance about the meaning of section 26(1)(e). However: (i) Section 26(1)(e) was inserted into the Act on the recommendation of the Transport and Industrial Relations Select Committee. The majority of the Select Committee said that their intention was for cover to be available when, were it not for the prosthesis, cover would have been available for a personal injury suffered instead. (ii) Therefore, submitted: (1) Section 26(1)(e) applies when a prosthesis has been harmed or damaged with some appreciable and not wholly transitory impact on the person, such that, in the absence of the prosthesis, the same harm or damage to the replaced body part would be a physical injury . (2) When there has been harm or damage to the body surrounding the prosthesis, but not the prosthesis itself, section 26(1)(b) is relevant. 3.6 "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)-(h). Section 20 provides for cover for personal injury. Relevantly, section 20(2)(f) provides that cover is available for "personal injury caused by a gradual process, disease, or infection that is a treatment injury suffered by the person". Treatment 3.7 "Treatment" is defined in section 33 and includes, relevantly: (a) The giving of treatment. (b) The failure of any equipment, device, or tool used as part of the treatment process, including the failure of any implant or prosthesis (except where the failure of the implant or prosthesis is caused by an intervening act or by fair wear and tear), whether at the time of giving the treatment or subsequently. [1 18] In respect to causation, Counsel in his submissions also refers to the decision of ACC v Ambros. He states that in cases of uncertainty the Court may draw robust inferences of the causal link between the treatment and the personal injury. However, any inference must be based on facts supported by evidence and not based on supposition or conjecture. [119] In respect to whether injury was as a result of a "Necessary part or ordinary consequence" counsel states at paragraph 3.9 of submissions: 3.9 As to whether a personal injury was a necessary part, or ordinary consequence, of treatment for the purposes of section 32(1)(c), relevantly for present purposes: a) The two exclusions are alternatives rather than cumulative. Therefore, cover will not be available if either exclusion applies.8 (b) In determining whether either exclusion applies, the Court must take into account all the relevant circumstances, including the person's underlying health condition and the clinical knowledge at the time of the treatment.' [120] Counsel in submissions under his Analysis section, submits: 4.1 ACC submits that: (a) The failure of Mr Heslop's total knee replacement was caused by wear, and therefore was not a personal injury or, it follows, a treatment injury. (b) There is no evidence that the failure of the total knee replacement caused any physical injuries. In any event, because the failure was caused by wear, any such injuries would not be caused by treatment or, therefore, treatment injuries. (c) The only other injuries were the surgeries and the presence of the total knee replacement, which were necessary parts and ordinary consequences of treatment. [121] Counsel then goes on to discuss the matter of a Personal Injury as relevant to this case. In respect to the nature of the injury he states: 8 Accident Compensation Corporation v McEnteer, HC Wellington, CIV-2008-485-1800, 1 December 2008 at [23]. 4.2 Since Mr Heslop lodged his claim, the relevant injury has not been described consistently or with clear reference to the types of personal injury listed in section 26(1) of the Act. 4.3 For example, Dr Hobbs described the injury as "collapse of [left] total mee joint replacement". The reviewer referred to "the tibial collapse of the left knee, the failure of the knee replacement". On appeal, Mr Heslop describes the injury as "an infection of his knee replacement leading to a loosening of his knee replacement". The medical experts refer interchangeably to the "collapse", "failure", or "loosening" of the total knee replacement. 4.4 As such, there has been uncertainty as to whether the injury should be treated as: (a) A physical injury (section 26(1)(b)), with the treatment being the failure of the prosthesis (section 33(1)(g)). (b) The failure of a prosthesis (section 26(1)(e)), with the treatment being the total knee replacement surgery in 2008 (section 33(1)(a)). [122] Counsel submits in either case, cover is not available as failure by wear (and tear) is excluded under s 26(1)(e) and s 33(1)(g). He states this is consistent with the scheme of the Act. He also states that if failure of the prosthesis was caused by wear there is no cover however if caused by infection, it is within the scope of s 32. [123] Counsel refers to the position of Mr Heslop being that the cause of the failure was an unrecognised infection arising out of the 2008 surgery. The Corporation, on the other hand, submits that there is no evidence of infection and that it is more likely than not that the failure was caused by aseptic loosening (namely wear caused by the implant shifting into varus malalignment over time). In respect to this position, in his submissions, Counsel states (para 4.6): (a) There were no signs of infection immediately following that surgery. Although a "slight lucency " under the tibial tray was noted eight months after the surgery, there was no change a year later. (b) There was no evidence of infection when Mr Hooper carried out the revision surgery in 2014. Pathology and histology results were inconsistent with the presence of an infection. In fact, the histology results were, according to Mr Taine, consistent with a response to wear. Section 32(1)(c). (c) Only a single stage revision surgery and a two-day treatment of antibiotics was required. Mr Pai considered that was "not a normal course of an infected total knee replacement as treatment of low virulent total knee replacement is difficult to irradiate". [124] Counsel also submits (at 4.7): The experts agree that it is possible that the loosening was caused by an undetected infection. [125] It is accepted that diagnostic testing does not always detect an infection; a low- grade infection is difficult to culture and the sonification study referred to indicates an infection may be present in 10-15 percent of cases which were previously treated as aseptic loosening. 10 [126] Counsel refers to the fact that both Mr Taine and Mr Pai consider it more likely that the loosening was caused by wear than infection as opposed to Professor Hooper's view. [127] Counsel submits that the Court should prefer the conclusions of Mr Taine and Mr Pai because of the following reasons: (a) Mr Hooper's evidence appears to be driven by a firm view that cover should be available for Mr Heslop. To explain: (i) Mr Hooper consistently suggested that the early failure of total knee replacement was a personal injury for which cover is available, regardless of the cause of the failure. However, the fact that treatment did not have the desired result does not, in itself, constitute a treatment injury, and failure caused by wear is expressly excluded under the Act. (ii) Mr Hooper has over-stated certain facts. For example, he said that Mr Heslop developed "significant" lucency under the tibia following the initial surgery, whereas Mr Walton (the treating surgeon) referred to "a very slight" lucency. Mr Hooper also said that failure within four to five years was unlikely, when the failure was more than six years after the initial surgery. (iii) There are also inconsistencies across Mr Hooper's four reports, including in relation to the average longevites of total knee replacements. Unlike Mr Pai, he has not cited or provided reference material for his statistical assertions. 10 Mr Pai's second report dated 8 January 2017. (b) Mr Hooper thought that an infection might be the cause when he first met Mr Heslop, but before any tests had been done. The only justifications he gave for maintaining that opinion after the diagnostic testing and revision surgery failed to reveal any infection were that (i) Mr Heslop's total knee replacement failed earlier than expected, having regard to average longevites (with approximately 90-95% of total knee replacements lasting 10 years). However, that is both factually and legally inconclusive, for the following reasons: (1) Total knee replacements can fail early, for a number of reasons. It is impossible to predict how long a particular implant will last. (2) Identifying the exact cause of failure in any case is not always possible. Infection is one of the possible causes of total knee replacement failure. (3) The sonification study found that aseptic loosening was the cause of early failure in 85-90% of cases in which there was no diagnostic evidence of an infection. Therefore, even if the Court had no other facts specific to Ms Heslop of which it could rely, based on probability alone it would be more likely that the failure was caused by aseptic loosening than an infection. (ii) In this case, every possible step has been taken to explore the possibility of an infection. In particular, consistently with the sonification study, deep tissue samples have been taken and examined. The study (of which Mr Hooper was an author) says, at page 6, that "the best method to confirm or exclude infection was deep tissue samples at the hive of [revision] surgery". Those samples did not show any infection. (iii) There was "no other obvious cause" for the failure. In fact, both Mr Taine and Mr Pai put forward a credible alternative, namely wear. In particular, Mr Pai explained that varus malalignment puts stress on a total knee replacement, causing wear and loosening. Mr Hooper initially said that varus malalignment was unlikely because both the tibial and the femoral components had loosened. However, Mr Hooper later accepted that that implant had shifted into varus malalignment after the 2008 surgery, and Mr Pai said that the failure of both components could occur in the case of aseptic loosening. (c) Mr Heslop submits that weight should be given to Mr Hooper's opinion as the treating surgeon and as an expert in joint replacement surgery. However, Mr Hooper only became Mr Heslop's surgeon for the revision surgery. Given that no evidence whatsoever of infection emerged during that surgery, Mr Hooper is in no better position than an independent expert to review the relevant medical records and give an opinion on causation. Mr Pai and Mr Taine are also experienced orthopaedic surgeons who are suitably qualified to give an opinion on causation [128] Counsel submits accordingly (at 4.10): Therefore, submitted, the diagnostic evidence, medical opinion and statistics support the conclusion that the failure on Mr Heslop's total knee replacement was caused by wear, not an infection. As such, Mr Heslop cannot establish a personal injury under section 26(1)(e). Treatment Injury [129] Counsel goes on to state at paragraph 4.11 and 4.12: 4.11 In the absence of a personal injury, Ms Heslop cannot establish a treatment injury. 4.12 Equally, if Mr Heslop could identify a physical injury caused by the failure of the total knee replacement, that personal injury would not be a treatment injury because the failure was due to wear, and therefore was not treatment (see [4.5] above). Submissions in Reply [130] Counsel for Mr Heslop has filed submissions in reply to the Corporation's submissions filed on 21 March 2017. [131] First, Counsel deals with Mr Pai's observation that the appellant's revision surgery in 2015 was a one stage revision and he was given antibiotics for two days only. [132] Counsel in response states at paragraph 3: However, Professor Hooper performed this revision surgery and indeed he wrote on 16 December 2014 that "there has been no growth at 48 hours which is encouraging. It is suggested in fact that his knee is not grossly infected so it is more likely that we will be able to do a one stage procedure than a stage two procedure. [133] Secondly, in respect to lucency Counsel notes that the Corporation in submissions refers to the letter from Mr Walton, six months following the initial surgery where he explains that Mr Heslop has a very slight lucency at that time as seen on the first of the three x-rays. The Corporation submitted this is inconsistent with Professor Hooper's report of December 2015 and that he had overstated the position. Counsel submits in response: 6. However, as Professor Hooper said on 12 December 2014, Mr Heslop initially presented with a very early appearance of radiolucent line which progressed. Professor Hooper's December 2015 report confirmed that it was some years following the surgery - in other words, some years following Mr Walton's report of "slight lucency" - before "he started to develop significant radiolucent line around his tibial component". 7 . Thus, Professor Hooper was explaining that Mr Heslop's lucency developed to be significant in the years following the surgery; this is not inconsistent with Mr Walton's early finding of slight lucency. [134] Thirdly, Counsel takes issue with the topic of longevity of replacements and counsel states that the Corporation submitted that Professor Hooper had overstated that: Failure within four to five years was unlikely, when the failure was more than six years after the initial surgery. [135] Counsel for the appellant states the complete sentence in his December 2015 report stated: ... loosening of the implant because of wear and tear of the implant is highly unlikely within four to five years, which is the time that he started to present with significant radiological findings. [136] Accordingly, Counsel submits that Professor Hooper was referring to Mr Heslop beginning to demonstrate significant radiological findings rather than the later collapsing of the tibia. [137] Counsel also takes issue with the submission of ACC heading "Failure to achieve the desired result". [138] Finally, with regard to "varus orientation, Counsel states: 12. It is correct that Professor Hooper has expressed that he understood the cause of the loosening to be irrelevant to a treatment injury claim. Professor Hooper's proficiency is of course in medicine; it is not uncommon for medical specialists to be less familiar with the ACC legislation. It is submitted that this is irrelevant to the question at issue as Professor Hooper is nonetheless clear in his opinion that, notwithstanding what is or is not required for a treatment injury claim, the cause of the oosening was an infection. 13. Finally, Professor Hooper explained that the varus orientation is relevant to the placement of the component at surgery. In Mr Heslop's case, Professor Hooper explains that his prosthesis was not placed in varus and was implanted in a satisfactory position, as seen in his post-operative x- rays. 14. Professor Hooper explained that, if the prosthesis is placed in varus, then tibial loosening is a well recognised complication. However, femoral loosening is not a common complication of the prosthesis being placed in varus. He therefore explained that the fact of Mr Heslop having both tibial and femoral component loosening is not consistent with loosening due to varus orientation. This is particularly so given that the prosthesis wasn't placed in varus. Mr Pai disagrees as his opinion is that "whether it is tibial loosening or femoral loosening" it can still be explained on the basis of aseptic loosening. However, it is submitted that, as Professor Hooper has explained, this is a clinical finding that supports the presence of an infection. 15. Professor Hooper's final report explained that Mr Heslop developed a varus malalignment some time after the placement of the prosthesis, however he did not consider this to be the cause of the loosening in Mr Heslop's case. 16. It is submitted that the clinical picture is one which demonstrates that an infection is the most likely cause of the loosening and that this Appeal should respectfully be allowed. The Hearing [139] It was expressed by Counsel at the hearing that the essence of this case, to some extent, is whose medical opinion is preferred. Professor Hooper's opinion, as stated by Counsel, is that there is evidence of infection. This is not laboratory evidence but based on his professional judgement. Counsel referred to the fact that he has been author of a number of articles and had specialisation in this area. Counsel submitted that Ambros was relevant to the decision given Professor Hooper's evidence over the four reports. [140] Ms Bisley, Counsel for the Corporation submitted that Professor Hooper was expressing himself to some extent as an advocate and that the presence of lucency was not in itself any evidence of infection. [141] He also submitted that this is not a case where there is a question only between two causes of causation, rather there are a number of possible causes. [142] Counsel also points to the fact that there was no evidence at the time of surgery and it is of note that Professor Hooper states in his report of 30 May 2016 at page 47 (Bof D): I really have no explanation for the rapid failure of his joint replacement and that is really the underlying reason why this has been applied for through ACC as it is an unusual and unexpected outcome following this type of surgery, particularly when the implants have been implanted in a satisfactory way. Discussion and Analysis [143] The starting point of the events is Mr Heslop's operation, which occurred on 28 March 2008, which was undertaken by Mr D I Walton, consultant orthopaedic surgeon of Christchurch and was a left total knee replacement. He noted that the kneecap was replaced and the medial compartment was exposed and this was grossly arthritic and deformed. It is stated that he undertook the standard procedure and he prescribed prophylactic antibiotics that were given. [144] There is a follow up letter from Mr Walton to the GP on 24 November 2008 which states: The x-rays are pretty good although there is a very slight lucency under the tibial tray rather than the stem on one or two of the views. [145] Mr Walton stated that he would keep an eye on this and would see Mr Heslop in six months for further x-rays. [146] It appears there was a follow up x-ray taken a year later on 12 November 2009. The radiologist Dr Gooding, states that the previous films were available and the prosthesis remain in an unchanged position. The lucency between the tibia cement and bone is unchanged to extend to the previous study and measures up to 5 mm in size. [147] As a result, Mr Walton wrote to the GP, Dr Hobbs, and advised there had been no change since March 2009 and that he should be seen again in 12 months with x-rays or sooner if there are any symptoms or concerns. [148] There is no evidence of any further difficulty either in x-rays or by way of information provided by Mr Heslop. [149] The next information is an x-ray taken in December 2014, some six years after the original surgery. Dr Gooding's report states that there is evidence of collapse of the tibial component into the proximal tibia associated with posterior and lateral tilting. Lucency around the central tibial peg. No significant fracture demonstrated. The meniscal bearings are not well profiled however probably remain positioned normally. The femoral component is unchanged in position however there may be some subtle increased lucency beneath the anterior aspect. [150] As a result Dr Hobbs referred Mr Heslop to Professor Hooper on 9 December 2014. Mr Heslop had been aware of a niggle in the leg for some time but it had become increasingly symptomatic over the previous six months. [151] Professor Hooper wrote back to Dr Hobbs on 12 December 2014 to advise of a diagnosis of "Loose Query Infected Left knee replacement". [152] In paragraph 2 of that letter Professor Hooper says: This really looks like and (sic) underlying infection. I am going to aspirate his knee today and I am going to organize to get ESR and CRP markers done, but he needs an urgent revision. If it is grossly infected then he will need a two stage revision, if it is not grossly infected then I would proceed with a single stage revision. It is likely we will need to involve the Infectious Disease Department as well. [153] Dr Hobbs' notes indicate that the aspirate from the knee took place on 12 December 2014. It was noted under Culture "No growth after 48 hours. The specimen has been incubated and a further report will follow". [154] Professor Hooper wrote to Dr Hobbs on 16 December 2014 noting that there had been no growth in 48 hours which he considered encouraging and stated: It is suggestive in fact that his knee is not grossly infected so it is more likely that we will be able to do one stage procedure than a two stage procedure. [155] There is a letter of the same day to Dr Chambers (from Professor Hooper) that stated: I saw this gentleman with his loose knee replacement. Loosening this early following his procedure is highly suggestive of infection altough(sic) all her parameters are down and really his aspirate is negative. I am just highlighting him to let you know that he may come your way if something turns up positive at the time of his surgery. I am planning to do a one stage procedure. [156] A pathology report done on 9 January 2015, reported "culture no growth". The revision surgery took place on 16 January 2015. [157] Professor Hooper's report of 16 January stated he had sent multiple samples off to try and establish a diagnosis but there was no overt infection. He went on to state: We will continue with IV antibiotics for the first 48 hours till the samples are returned. Postoperatively routine. [158] A Histology report of 16 January 2015 from Dr Shona Mcdowell states: Diagnosis: knee joint tissue Chronic synovitis in keeping with prosthetic joint wear. [159] A report from the anaesthetist Dr Menon to Dr Hobbs states: I revised this gentleman's knee today. Both compartments were grossly loose. There was no overt infection. I have sent multiple samples off to try and establish the diagnosis. [160] There were details of ongoing follow up consultation notes with Professor Hooper. His note of 2 March 2015 states: I saw Donald again today. He is doing very well with his knee, he is flexing it through to at least 100 degrees now but he has developed some cellulitis in his distal tibia and some weeping area in his knee. I want him to be on antibiotics to avoid any infection of this implant so I have started him on Flucloxacillin today. I will review him again next week just to make sure everything is satisfactory ... [161] A further letter of 9 March 2015 to Dr Hobbs from Professor Hooper states: ... He has still got a little bit of cellulitis in his distal tibia and I want him just to complete his antibiotics as scheduled. ... I will review him again in three months time with an xray and keep you informed of his progress. [162] Accordingly, approximately two months after the surgery, Mr Heslop developed a bacterial skin infection and was subsequently put on antibiotics (fluxcloxacillin). [163] Professor Hooper's report of 5 June 2015 on this matter was sent to the Corporation. In this he makes a tenuous link only between the loosening of the implant and possible infection: We know that the knee replacement that he had inserted has an excellent longevity and our Christchurch figures would show that approximately 92% of those are still surviving at 20 years so for it to loosen within six years is highly suggestive that it is infected. The problem with infected arthroplasties is that sometimes they can be extremely difficult to diagnose as is the case with this gentleman. I note he has put in a claim for treatment injury because his implant has not survived as he had anticipated and I think that is not unreasonable. Whether his problem is secondary to loosening of his implant because of other factors or whether it is due to infection is probably irrelevant bearing in mind that the anticipated longevity of this prosthesis is significantly more than six years. [164] He concludes: I understand that this is contentious and I understand that it is difficult to make a definite diagnosis in these cases but on the balance of probabilities it is more likely that this knee has become infected or was infected right from the time of surgery. [165] Mr Taine provided a response on 15 June 2015. It is noted that a work up (ESR of 15 and CRP of 4) for infection showed no growth on an aspirate and no rise in inflammatory markers. At the revision on 16 January 2015 both implants were found to be loose, but with no frank evidence of infection. Specimens were stated to show no growth. The histology reports are available from that surgery, reporting histocytes and multinucleated giant cells with refractile material consistent with a chronic response to plastic wear debris. No leucocytes were present. A single revision was performed. [166] Although Mr Taine says that "a low grade infection acquired at the time of his initial surgery is certainly possible", he continued: ... In favour of that as a cause of the loosening is the early note of a radiolucent line under the tibial implant. There is loosening of the implant before the average or expected lifespan, although it is appropriate to point out that figures for longevity are probabilities only: it is not possible to predict exactly how long a single implant will survive before revision, only that most survive for a satisfactory period (as a rough guide, about 1% per year will fail for some reason) [167] He states conclusively that histology is not indicative of infection. He states: The presence of the described cells is consistent with a response to wear rather than infection. I would therefore disagree that infection is the probable cause of the loosening in this case. [168] In the treatment injury report issued with the decision of 18 June 2015 it is stated: For ACC to accept a treatment injury claim for cover, there must have been clinical evidence of a physical injury caused by treatment which will then be assessed against the remaining criteria for treatment injury, for example, where the injury is a necessary part or an ordinary consequence of treatment. After assessing the available information ACC declines the claim for cover as there has been no definite diagnosis established in this case [169] Mr Taine's further report of 10 November 2015 (page 33 BofD) explained why he was of the view that Mr Heslop's condition came from other sources. He states: Loosening of an implant causes a physical change in the tissues compared to a securely fixed implant: fibrous tissue formation between bone and cement, sometimes fragmentation of cement, various degrees of bone loss, and sometimes movement of the implant (subsidence, migration, etc) There will be a degree of inflammatory response depending on the cause of the loosening. I dislike the use of the term "injury" in this context, as it implies an accidental cause, but in terms of consideration by ACC, there is a physical change i.e. injury. The labelling of that change as injury does not imply anything regarding causation In cases like these (this is not an uncommon scenario across the claims I have commented on in the past), the issue is the cause of the loosening. There are a number of possible causes: low grade infection, malalignment, inadequate cementing technique, third body wear, to name a few. However, in many cases a specific cause is not apparent. The histology suggests plastic wear as a contributor but not a cause of that wear. The presence of the TKR" is a necessary part. As discussed in the previous advice, it would be expected that this TKR should have lasted for a longer period, but this is an expectation made on the basis of probability, not with regard to an individual case. In other words, of 100 TKRs, 5-10% will require further surgery in the 10 years following surgery (an indicative figure), but which specific ones will be problems cannot be predicted. If a specific reason for the loosening was identified, and that reason was due to treatment, then a causal link would be sustainable. But if no particular cause is able to be provided, it cannot be said that there is a causal link between the mere presence of the implant (as above, this is a necessary part). [170] Professor Hooper's second report of 15 December 2015 for the review hearing (page 34 BofD) states: In direct response to your questions - 1. As mentioned, my diagnosis continues to be failure of his knee replacement secondary to unrecognised infection. The reason why I continue with that diagnosis is that this gentleman had his knee replacement insitu for only a few years before he started to develop significant radiolucent line around his tibial component. The longevity of the type of knee replacement this gentleman had insitu, particularly the cemented form of this implant is approximately 95% survival at 15 years (Christchurch figures) and so it is unlikely that this would fail because of aseptic loosening this early. 2 . I do not believe that this gentleman has had a personal injury. I believe that the claim that should be proposed for this gentleman is an unsuccessful outcome following his knee replacement secondary to presumed infection, which would be a treatment injury. 3. There is a possibility that this gentleman's overall inflammatory state could have been suppressed by his testicular problem, although I have no evidence to suggest that. [171] However, despite this report, I find there appears to be no basis to suggest that the loosening of the implant is caused by infection. [172] Professor Hooper himself says at paragraph 4: The only other comment I would like to make is that sometimes it can be extremely difficult to make diagnosis of infection following a hip or knee replacement, particularly an indolant infection with an organism which is extremely difficult to grow in the laboratory. 11 TKR - peri prosthetic total knee arthroplasty. [173] Accordingly, I do not find definitive evidence to support that any part of the procedure at the operation in 2008 resulted in an infection developing. [174] This is not to say that subsequently, such as in the example of cellulitis as occurred in the second operation, an infection may occur. [175] Mr Pai in his Treatment Injury Advice of 28 February 2016 specifically goes through the events of 28 April 2008, the date of surgery by Mr Walton and the further notes from Mr Walton of 24 November 2008 and 11 November 2009. Apart from the lucency which had been earlier identified and measured there was no change and he was to be seen again in 12 months. [176] It is accepted when Ms Heslop did make contact with another health provider and some five years later x-rays showed the medial component to be collapsed into the tibia however there is no medical information other than conjecture, to support the diagnosis made by Professor Hooper on 12 December 2014 that his presentation was because of infection which occurred at the date of surgery in 2008. All tests undertaken in 2014 including the blood markers (ESR and CRP) and the knee aspiration were negative for infection. [177] In addition, the results of the second surgery did not indicate any infection and none of the specimens taken at the time came back positive. [178] Mr Pai stated: From the available documents, it is not possible for me to objectively consider infection as the causation of his total knee replacement failure. [179] He referred to the fact that there were some early failures and some of these early failures may not be related to infection at all. 180] He also refers to the tests undertaken (the CRP, the normal ESR and the fluid aspiration), as being negative for infection, together with the multiple histology taken at the time of surgery that is a "... gold standard for infection also being negative". [181] Mr Pai's first report of 28 February 2016 goes through the patient's management and documents in detail. He also referred to the investigation tools such as sonification for the prosthesis (that can yield more positive results as to whether there is any infection). He referred to the study of 190 diagnosed cases of aseptic loosening with an early failure, and that the detection rate of infection as causation using a sonification fluid of prosthesis and tissue samples yielded around 10% being positive for infection. This would mean that 90 percent of the group still had still aseptic loosening. It is Mr Pai's opinion that, on the balance of probabilities, Mr Heslop's failure is more likely related to aseptic loosening rather than an infection. [182] Professor Hooper's further report of 30 March 2016 refers to a Christchurch study using sonification where the results indicated 15 percent demonstrated infection in previously unrecognised infected cases. [183] Professor Hooper still maintains his view that: ... to see gross loosening of gross components within six years of a knee replacement means it is almost certainly the implant is infected. The fact that we cannot grow an organism does not rule out this diagnosis. [184] Professor Hooper agrees that: There is no doubt that knee replacements can fail if they are mal-aligned, that is if the component is placed in varus then there is an incidence of tibial loosening. [185] It appears to be Professor Hooper's opinion that: As previously mentioned if there was a varus orientation of the tibial component I would expect to see tibial loosening not tibial and femoral loosening [186] In paragraph 5 he relies on his diagnosis of infection and states: The points in favour of infection in this gentleman are the short time span to development of radiolucent line around the tibia to the extremely short time 12 Published in BMC Infectious Diseases 2015, Volume XV. span for both femoral and tibial components to be grossly loose. This scenario is rarely seen in cases of aseptic loosening. [187] He appears to acknowledge both possibilities but because of the timeframe of six years he believes the more likely explanation is that it is infection related. Subsequently he goes on to state: I find this discussion difficult to comprehend when this gentleman is applying for ACC as treatment injury because of a poor or unexpected outcome following total knee replacement surgery. I think that all of the adjudicators would be of the opinion that six years following a total knee replacement is an extremely poor outcome is certainly not what is expected from the procedure and is by definition a treatment injury. [188] Unfortunately, this comment adds nothing to the discussions. The fact that the outcome after six years of the knee replacement is poor is not per se basis of eligibility for ACC assistance. [189] As noted above this view has also been referred to in Professor Hooper's last report of 30 May 2016 when he accepts that he has no explanation in respect to the rapid failure of the joint replacement. [190] He also states in the report of 30 May: The varus angulation of this gentleman's knee has developed since his immediate post operative xrays in 2008. His intermediate xrays looked well aligned in valgus with no obvious abnormality underneath the tibial component but have significantly shifted into a varus malalignment and subsequent loosening. [191] This certainly appears to be something that is not commented on in the 2008 or 2009 x-rays however there are no x-rays available subsequently until 2014 so it is extremely difficult to judge when this occurred. It is noted that Mr Heslop on presentation in 2014 said he had niggles in his left knee for at least six months. This aseptic loosening is a failure of the bond between an implant and bone in the absence of infection and it would seem that the risks increase as the age of the patient at the time of the implant. It is seen that it can occur in the absence of infection and that there can be interface with lucency. [192] Mr Pai in his supplementary report of 8 January 2017, responds to Professor Hooper's letter and reports. [193] He states: My report was based on the documents and investigations provided to me and on reviewing literature, and there was no conflict of interest as I am not his treating specialist. [194] He states, as noted in his earlier report, he is aware of significant sonification studies done in Christchurch. [195] He states: I would like to pointe out that I am aware of this study from 2006 as we had a lecture from world renowned professor on infection either during our training program or in NZOA conference, I have also gone through Professor Hooper's paper as stated under (Reference 6) - this paper has suggested they recommend routine intraoperative sampling for patients and did not support routine use of ultrasound sonification for detection of infection. The fact remains that even if sonification studies are done, it detects only 10% - 15% of aseptic loosening and therefore the other 85%- 90% are still due to aseptic loosening. The sonfication studies is done mainly in Christchurch and I am not sure as to why this test was not carried out on Mr Heslop when there was difficulty in diagnosing an infection and it was suspected. It is clear in literature that in major centres multiple site biopsies are used and in my opinion multisite biopsy is still the gold standard and this test that was performed did not confirm an infection in Mr Heslop. With regards to early failure of arthroplasty related to aseptic loosening (I have provided literature references in my original report and in addition further reference articles have been provided that aseptic loosening is still the cause for early failure), this is not uncommon, and whether it is tibial loosening or femoral loosening or presence of tibial and femoral aseptic loosening, in my opinion, this is not necessarily related to an infection and can still be explained on the basis of aseptic loosening (Reference 2,3, and 5). I am in agreement with Mr Taine's report of 15/06/2015 that at this stage infection cannot be regarded as a probable cause for his aseptic loosening. In my opinion, presence of femoral failure in addition to tibial failure can occur in aseptic loosening and that as such does not confirm loosening is secondary to infection. In my Heslop's case his revision surgery in 2015 did not reveal overt infection or positive multiple site culture and was treated like an aseptic revision replacement (one stage revision) and IV antibiotics were give only for two days in other words he was not treated like an infected total knee replacement where V antibiotics are given for a total period of six weeks. From the documents available to me it appears that there is no evidence of deep infection or any clinical evidence of failure of revision surgery. This in my opinion is not a normal course of an infected total knee replacement as treatment of low virulent total knee replacement is difficult to irradiate. [196] Mr Pai also answers a number of questions put to him and states: In this particular case femoral component loosening is a part of aseptic loosening and it is more than likely related to poly wear particles or failure to osseointegrate. Although I cannot quantify with the available documents as stated in my original report, Mr Hooper's letter does suggest that there was some varus of the tibial component and this orientation can cause more wear particles and an active person leading loosening and in some situations osteolysis. 2 When there is failure of total knee replacement either due to varus alignment or failure to osseointegrate there is bound be increased wear due to stress concentration. This leads to accelerated poly wear. The poly particular because of increased osteoclastic activity causes loosening and increases deformity and which further increases stress on the poly. This is a dynamic situation and requires revision surgery. The poly wear definitely changes the subchondral bone and the stability of the prosthesis fixation is affected. Such changes can occur in early stages of knee replacement (Reference 1,2,3 and 5). It is still considered to be the significant cause for early failure. [197] He also refers to three references: Haddad. The epidemiology of failure in total knee arthroplasty: avoiding your next revision. Bone Joint J. 2016 Jan; 98-B(1 Supple): 105-12 High Early Failure Rate of the Columbus@ Posterior Stabilized High- flexion Knee Prosthesis CORR 470: 1472 Mechanisms of failure of the femoral and tibial components in total knee arthroplasty. Insall. CORR 1989 248 [198] It is noted that the last reference referred to by Mr Pai is an article by Professor Hooper. In this article, it is stated: We recommend routine intraoperative sampling for patients having revision for aseptic loosening, but we do not support the routine use of ultrasound sonication for its detection. Conclusion [199] The requirement of a treatment injury in this case is the determination that it was caused by the treatment received at the time of the original operation in March 2008. [200] Mr Heslop's position is that the femoral component loosening was caused by infection contracted from the 2008 surgery. 201] Although is it accepted infection can be a factor, Mr Heslop must be able to establish this claim on the balance of probabilities. [202] No evidence of infection being contracted at the time of the operation has been eluded to and at best, Mr Heslop retained slight lucency between the tibial cement and the bone being noted on subsequent x-rays subsequent in November 2008 and November 2009. [203] No specimen at the time of surgery or subsequent has identified any infection and it is noted that even if there is aseptic loosening there can be thin interfaces identified. It is noted by 2014 there had been some subtle increase. [204] Professor Hooper stresses the expected life of the longevity of Mr Heslop's knee replacement but accepts the view of the other specialists that this is speculative only and that wear is the likely factor. [205] In applying the Ambros principles, I do not find sufficient material pointing to proof of injury in the left knee, either at or after the surgery in 2008. [206] As noted, shortly after the 2014 surgery, Mr Heslop required treatment for an infection which was treated at that time by antibiotics. [207] Although Professor Hooper was the treating surgeon, his position in June 2015 of the failed knee implant was because "there is no other obvious cause for the problem". However, the lack of indicators of infection in 2008, 2009 and later in 2014 do not support Professor Hooper's opinion that infection was the cause. [208] I accept the submission of Counsel for the Corporation that Professor Hooper, being the surgeon of the revision operation is in no better position to give evidence as to an infectious origin, particularly given no evidence of infection whatsoever emerged for the second surgery. [209] I accept that both Mr Taine and Mr Pai have put up a credible and reasoned alternative, namely wear. As noted, Mr Pai explained that varus malalignment puts stress on a total knee replacement, causing wear and loosening. [210] I find that the Corporation made the correct decision on 18 June 2015 declining cover for a treatment injury and the review decision of 3 August is upheld. [21 1] According, this appeal is dismissed. [212] There is no order as to costs. District Court Judge Solicitors: John Miller Law, Wellington for the appellant Buddle Findlay, Wellington for the respondent