Commentary
Although the detail of this judgment is for diabetologists and gastroenterologists it is of general interest with regard to its treatment of statistical evidence, reliance on odds ratios and relative risk ratios and the position taken by the court on the relationship between the civil standard of proof and confidence intervals.
Case
This was the trial of the Claimant's claim for damages for clinical negligence arising out of an endoscopic retrograde cholangiopancreatography procedure (ERCP) performed on 16 July 2019.
Experts:
Professor Ian Gilmore, Consultant Physician and Gastroenterologist on behalf of the Claimant.
Dr Jeremy Woodward, Consultant Gastroenterologist on behalf of the Second Defendant.
Dr B, Consultant Physician and Gastroenterologist on behalf of the First Defendant.
History
In May 2019 the Claimant suffered intermittent upper right abdominal pain. On 28 May 2019, he underwent an ultrasound scan of his abdomen at the Hospital. On 14 June 2019 he had a consultation with the Second Defendant as an outpatient. There was a discussion about endoscopic retrograde cholangiopancreatography (ERCP). In cross-examination, the Second Defendant said that he did not at this consultation advise the Claimant that he was going to receive a diclofenac suppository rectally. On 18 June 2019, the Claimant had a magnetic resonance cholangiopancreatography scan at the Hospital, which showed a stone in the common bile duct. The Second Defendant advised that an ERCP be performed. On 16 July 2019, the Claimant underwent an ERCP at the Hospital, performed by the Second Defendant. The ERCP revealed an 8 mm stone in the distal common bile duct, which was removed.
The Second Defendant said in his witness statement,
"It is my usual practice to always prescribe Diclofenac either before or during a ERCP procedure in all cases in order to reduce the risk of post-procedure pancreatitis"
At 21:30 the Claimant was discharged from the Hospital. At around 23:40 the Claimant's wife contacted the Hospital. She reported that the Claimant was suffering from abdominal pain. Advice was given to take simple analgesia and observe.
On 17 July 2019 at 02:29, the Claimant attended the Accident & Emergency Department at University Hospital Lewisham with severe abdominal pain. Severe acute necrotising post ERCP pancreatitis was diagnosed and he was admitted. He was initially managed by the Surgeons of Lewisham & Greenwich Hospital and the Clinical Critical Care Outreach Team.
On 19 July 2019 the Claimant was admitted to the Critical Care Unit as his condition had deteriorated and a CT scan showed acute necrotising pancreatitis. He required intravenous feeding, artificial ventilation, haemofiltration and showed evidence of multi-organ failure.
On 14 August 2019 the Claimant underwent surgery because of deteriorating sepsis. At surgery, he was found to have a perforated colon with gross faecal contamination and peritonitis and was severely unstable. A double-barrelled ileostomy was created.
On 13 November 2019, having spent almost four months in hospital, the Claimant was discharged from King's College Hospital. However, he was readmitted with a further gastrointestinal bleed approximately three days later. He underwent endoscopy and angiography, which were unable to identify the source of the bleeding, but there was concern about the risk of developing a pulmonary embolism and an inferior vena cava (IVC) filter was placed.
On 17 July 2020 an attempt was made to reverse the stomata at the Hospital. However, this was abandoned because of significant blood loss of two and a half litres and closure of the stomata was not attempted. On 18 July 2020 a further re-look laparotomy was carried out. On 19 July 2020 a colonic re-anastomosis was performed, with a covering loop ileostomy on 21 July 2020. However, a further CT scan on 3 August 2020 showed that there was a splenic and partial superior mesenteric vein thrombosis resulting in left-sided portal hypertension. The Claimant was discharged on 8 August 2020.
In October 2020 the Claimant developed an enterocutaneous fistula (communication between the gut and the skin). He was placed on home intravenous feeding to manage this condition.
The Claimant required a further admission with an acute hyperglycaemic state due to having developed diabetes necessitating Insulin treatment.
On 11 July 2021 the Claimant underwent a successful operation to close the fistula and reconnect the small bowel to the sigmoid colon. Following his recovery from this operation, the Claimant was able to return to full-time work.
Counsel for the Claimant said in his opening skeleton argument,
"The most significant complications of C's PEP [post-ERCP pancreatitis] are that he has altered bowel function and has developed type 3c diabetes. Diabetes was first apparent on 8 May 2021 when he required admission for an acute hyperglycaemic state. He is now insulin dependent with a reduced life expectancy and at a risk of diabetic complications."
Finding as to diclofenac
The court found that all the lay and expert evidence was in agreement that the Claimant should have been informed as part of the consenting process that diclofenac would be administered rectally.
The court was concerned that Dr B thought it appropriate to say in his report that he would not expect any discussion about the use of a diclofenac suppository. The court found that this position was untenable and would constitute substandard and negligent practice. It found that Dr B's evidence on this point diminished his credibility as an independent expert and resulted in the court being able to place less weight on his evidence. The court also noted that Dr B did not comment on the fact that the Second Defendant did not complete the sections in the consent form relating to anaesthesia and sedation.
The court found the Defendants' argument that the Claimant would not remember being informed that diclofenac would be administered rectally or remember diclofenac being administered rectally was fundamentally flawed.
The court rejected the Second Defendant's evidence at trial that he consented the Claimant for the administration of diclofenac. The court found that the Claimant was not advised by the Second Defendant as part of the consenting process on 16 July 2019 that diclofenac would be administered rectally if he underwent an ERCP.
The court found that a prescription for diclofenac should have been recorded in the Drug Prescription and Administration Chart and that diclofenac was not administered to the Claimant on 16 July 2019. The Second Defendant intended to prescribe diclofenac but failed to communicate this to the First Defendant's nurses and was thereby in breach of his tortious and contractual duty of care to the Claimant in failing to instruct the First Defendant's nurses to administer diclofenac. Having regard to the fact that the Second Defendant did not provide a written prescription or an oral request to the First Defendant's nurses to administer diclofenac, it followed that the claim against the First Defendant had to be dismissed.
Causation:
The Claimant submitted that on the balance of probabilities causation was proved on two bases:
i) "But for" the breach of duty of care of the Second Defendant, the Claimant would have avoided pancreatitis.
ii) The Claimant has suffered a material contribution in that the severity of his pancreatitis would have been reduced but for the negligence of the Second Defendant in failing to prescribe diclofenac.
The Defendants referred the Court to the following authorities on the application of statistical evidence: Gregg v Scott [2005] 2 AC 176 and Sienkiewicz v Greif (UK) Ltd [2011] UKSC 10.
In the joint statement of the Consultant Gastroenterologists, they said:
"They agreed that the relative risk ratio rather than the odds ratio would be the correct statistical tool, equating to 'balance of probabilities' although the odds ratio and relative risk ratio would likely approximate closely."
The court had regard to:
Dumonceau JM et al; European Society of Gastrointestinal Endoscopy. Prophylaxis of post-ERCP pancreatitis: European Society of Gastrointestinal Endoscopy (ESGE) Guideline - updated June 2014. Endoscopy. 2014 Sep;46(9):799-815.
Buxbaum JL. American Society for Gastrointestinal Endoscopy guideline on post-ERCP pancreatitis prevention strategies: summary and recommendations. Gastrointest Endosc. 2023 Feb;97(2):153-162.
Issues on "but for" causation:
i) Is it appropriate to rely upon sub-group analysis?
ii) If so, is it appropriate to rely on the studies selected by Professor Gilmore?
iii) Should the Court decide the case on the balance of probabilities, i.e. more than 50%, or on the scientific standard, namely a confidence interval of 95%?
Is it appropriate to rely upon sub-group analysis?
Professor Gilmore considered subgroups from two meta-analyses, namely:
i) The meta-analyses laid out in the supplement to the European Guideline
ii) Lyu et al's study: What is the impact of nonsteroidal anti-inflammatory drugs in the prevention of post-endoscopic retrograde cholangiopancreatography pancreatitis: a meta-analysis of randomized controlled trials the ESGE (2020) guidelines. BMC Gastroenterology (2018) 18:106.
Professor Gilmore said in his report,
"The best review of the evidence is in the ERCP Adverse Events Guideline of the European Society of Gastrointestinal Endoscopy (2020), and the relevant table 3s is included as an appendix to this report. Overall, the table confirms a benefit from administration of an NSAID, but it is not clear that it makes the prevention of pancreatitis more likely than not. However, the evidence review includes different NSAIDs and routes of administration, and in the specific case of rectally administered diclofenac there is a better than 50% chance of preventing pancreatitis in 9 of 11 available studies, reducing the risk usually to about 40%. From this I conclude that, if the court finds that diclofenac was not given, it is more likely than not that pancreatitis would have been prevented by its rectal administration."
Professor Gilmore said in the joint statement:
"While accepting that the meta-analyses laid out in the supplement to the ESGE (2020) guidelines utilise overlapping studies, it is notable that in the 7 meta-analyses that looked at the risk ratio of rectal diclofenac, all found a value of less than 0.50 (0.24-0.41) and superiority over indomethacin.
ITG concludes that, if the court finds that rectal diclofenac was not given and should have been, on the balance of probabilities pancreatitis would have been avoided."
Professor Gilmore said in the joint statement that Lyu et al is the only study to look specifically at individual drugs and routes of administration. He said,
"IG agrees with JW [Dr Woodward] that a recent and helpful meta-analysis is provided by Lyu et al, and notes also that of Liu et al (both are published in 2018 and include 21 and 19 randomised controlled trials (RCTs) respectively). Lyu et al is the only one to look specifically both at individual drugs and routes of administration, and so there are data specifically on rectal diclofenac, the drug and route under consideration in this case. 330 patients received diclofenac and 338 patients placebo. The risk ratio with rectal diclofenac was 0.38, giving a risk reduction of 62%."
In the joint statement, Professor Gilmore said,
"IG understood JW's wish to bring scientific validity to the question by relying on the confidence intervals of 95% ie the conclusion is likely to be correct 19 times out of 20. In order to achieve that, JW has amalgamated data from other drugs, usually indomethacin, and other routes of administration, such as oral or by injection. However, in IG's opinion this fails to answer the question posed by the court, namely if it finds that diclofenac 100 mg per rectum was not administered and should have been, if it had been administered pre-operatively would the claimant have avoided pancreatitis on the balance of probabilities. In his view, this means that evidence regarding rectal diclofenac only, as opposed to other drugs and/or other routes, should be considered. The test should be that the relative risk (preferably) or odds ratio should be reduced below 0.5. It will be for the court to decide whether or not to deviate from the principle of balance of probabilities and instead take into account the possibility that the balance of probabilities does not reach the 95% confidence limit. IG does not accept that the different routes of administration and different drugs used can be amalgamated. Indeed the current evidence suggests that diclofenac is more effective than indomethacin and that the rectal route is more effective than others."
Applying the guidance of Lord Nicholls at paragraph 32 of Gregg v Scott (supra),
"The value of the statistics will of course depend upon … the closeness of their position to that of the claimant."
the court found that the statistics closest to the position of the Claimant were those involving the administration of diclofenac rectally preoperatively. It accepted Professor Gilmore's evidence that it is appropriate to carry out a sub-group analysis in order to answer the question before the Court, ensuring that the statistics being used are close to the position of the claimant.
It accepted Professor Gilmore's evidence that the number of patients (668) in the five studies in Lyu et al which provide data specifically on rectal diclofenac and the number of patients in the seven studies which provided data specifically on rectal diclofenac in Table 3s to the European Guideline are numbers which are often used in meta-analysis and there is nothing unusual about them. Professor Gilmore said of the numbers in the five studies in Lyu,
"That's the sort of numbers that are often used in meta-analysis. There's nothing unusual - there's nothing illegal or unusual about using meta-analysis on 600 patients."
Distinguishing between different routes of administration of diclofenac:
All three Consultant Gastroenterologists agreed that the only effective method of administering NSAIDs is rectally. It found that the Court should consider studies where NSAIDs were administered rectally because these studies are closest to the position of the Claimant, and the question before the Court was whether on the balance of probabilities if 100 mg diclofenac had been administered rectally to the Claimant, this would have prevented pancreatitis.
Distinguishing between diclofenac and indomethacin:
Professor Gilmore referred to table 3s in the appendix to the European Guideline. In table 3s there are seven meta-analyses which deal specifically with rectal NSAIDs and are split by drug, namely diclofenac or indomethacin. Table 3s shows that in all seven, the risk ratio or odds ratio was below 0.5 for rectal diclofenac. Table 3s shows that the risk ratio or odds ratio was above 0.5 for rectal indomethacin:
The court accepted Professor Gilmore's analysis that if one considers the seven studies at table 3s of the European Guideline and the five studies in Lyu et al in which diclofenac was administered rectally, it can be seen that diclofenac administered rectally is more efficacious in preventing pancreatitis and reduces the relative risk ratio below 50%.
The court found that if one uses the meta-analyses, one is using drugs and means of administration which are known to be either less effective than the administration of diclofenac rectally or not effective at all. The meta-studies have used NSAIDs administered intramuscularly, orally and intravenously, which are known not to reduce pancreatitis and indomethacin, which can be seen to be less efficacious than diclofenac.
The court found that studies involving diclofenac administered rectally should be considered because they are closest to the position of the Claimant and the Court was tasked with considering whether 100 mg diclofenac administered rectally to the Claimant would on the balance of probabilities have prevented pancreatitis.
Was it appropriate to rely on the studies selected by Professor Gilmore?
For the Defendants it was submitted that three of the five studies in Lyu et al which involved only diclofenac administered rectally, namely Khoshbaten, Lua and Murray, concern specifically selected high-risk cases, whereas the Claimant was low to medium risk or low risk. It was submitted that these three studies did not match the Claimant's risk. They constitute 464 of the total 668 patients, or 69% of the patients in the five studies relied upon by the Claimant
Dr Woodward said in cross-examination that to compare the Claimant with patients who were high risk was like comparing the Claimant's case with a different procedure. However, Dr Woodward also said that based on the literature it was not possible to say whether the efficacy of diclofenac was different in different risk groups.
It was also submitted that in Lyu et al the authors said that no differences were observed between diclofenac and indomethacin,
"Our meta-analysis showed that the rectal administration of NSAIDs might be the most effective in decreasing the incidence of PEP. Further subgroup analysis showed that indomethacin and diclofenac were able to reduce the incidence of PEP significantly compared with the placebo control. However, no differences were observed between the two drug groups.
It was further submitted that there is a difficulty with the Otsuka study in Lyu et al as to how pancreatitis was defined.
Regarding the first point, Professor Gilmore said in cross-examination:
i) "I know from the evidence that there is no or unlikely to be any difference in the efficacy between high risk and average risk."
ii) "There is no strong evidence that there's any difference in response between high and low risk"
Professor Gilmore said in re-examination,
"MR DUFFY: My question is this: is there any reason that diclofenac would be more effective in a high-risk case than an average-risk case?
PROFESSOR GILMORE: Clearly the high-risk case is by definition more at risk of getting pancreatitis, but I don't know any particular reason why diclofenac would be less or more effective.
MR DUFFY: Thinking about it this way, if you're at low risk of getting pancreatitis, I mean, would that make it easier or harder for diclofenac to work, if I can put it like that?
PROFESSOR GILMORE: I don't think I've got the evidence to answer that, but I would expect, if anything, diclofenac would be less of a hard job to do if there wasn't high risk. For example, you know, if the patient's pancreatic duct is cannulated or something like that, that would put them at higher risk and the diclofenac would have a bigger job to do.
MR DUFFY: Just follow that through mechanically. So why is – why would it be harder?
PROFESSOR GILMORE: Because there would be an inflammatory -- more likely be an inflammatory reaction triggered that needed dampening down."
Regarding the second point that in Lyu it was said that no differences were observed between diclofenac and indomethacin, Professor Gilmore said that this statement had to be looked at in context. The meta-analysis carried out by the writers of the Lyu et al paper did not compare diclofenac administered rectally and indomethacin administered rectally. They say,
"No study has compared rectal indomethacin and diclofenac in a head-to-head trial to see if there is any difference in the efficacy between these 2 agents."
The Defendants relied upon figure 5 in the Lyu report (G/64/p. 153), which provides a combined risk ratio of 0.54. However, this table differentiates by route rather than by drug and includes diclofenac and indomethacin (and also naproxen in the case of the Mansour-Ghanaeiv study) without differentiating between them. As Professor Gilmore said in cross-examination, the greater number of indomethacin patients had "swamped the diclofenac data" in figure 5. Professor Gilmore relied upon the seven studies in the European Guideline and the five studies in Lyu et al which show that the results for diclofenac were better than those for indomethacin in terms of risk ratios and odds ratios.
The court therefore found firstly that the evidence given in the present case showed on the balance of probabilities that there is no difference between the efficacy of diclofenac in high-risk cases and low to medium-risk or low-risk cases.
Secondly, it found that the statement in Lyu et al that no differences were observed between diclofenac and indomethacin had to be looked at in context. As the writers of the Lyu et al paper say, "No study has compared rectal indomethacin and diclofenac in a head-to-head trial to see if there is any difference in the efficacy between these 2 agents". It found that Professor Gilmore had demonstrated cogently and convincingly that if one analyses the seven studies in the European Guideline and the five studies in Lyu et al, it could be seen that diclofenac administered rectally is more efficacious than indomethacin and reduces the risk of pancreatitis below 50%.
Thirdly, the court found that the Otsuka study did apply the usual Cotton criteria to define pancreatitis. In cross-examination it was put to Professor Gilmore that one would expect the Cotton criteria to be used to define pancreatitis. Professor Gilmore pointed out that in the Otsuka study it is said,
"The primary outcome measure was the occurrence of PEP, defined by the criteria of Cotton et al. as the development of abdominal pain and elevation of the serum amalyse level to greater than three times the upper normal limit within 24 h after ERCP."
Regarding the additional point that the Otsuka study in Lyu et al involved patients who received no more than 50 mg, whereas the Claimant's intended dosage was 100 mg, the court found:
i) Neither Dr Woodward nor Dr B submitted that the dosage of the NSAID was material. There was no evidential foundation for the submissions.
ii) In any event, the Second Defendant's intended dosage of diclofenac was 100 mg, which is the optimal dosage. This was the dosage recommended in the European Guideline and the American Guideline. Therefore, the probability of the Claimant developing pancreatitis would have been less likely than in the patients in the Otsuka study who received no more than 50 mg. The risk ratio for the patients in the Otsuka study developing pancreatitis was 0.21, which is far in excess of 50%. As the Second Defendant intended the optimal dosage of 100 mg to be administered to the Claimant, the Claimant's risk of developing pancreatitis would have been even lower.
Standard of proof and confidence interval:
The court found that the standard of proof in a civil clinical negligence case is the balance of probabilities, namely more than 50%. The standard is not the scientific standard of a confidence interval of 95%. Dr Woodward was applying the standard required to submit a paper to a medical journal. In so doing, he failed to address the question to be answered by the Court of whether the Claimant would have avoided pancreatitis on the balance of probabilities.
The court noted that counsel for the first defendant accepted that the crucial issue in this case was "whether administration of rectal Ddiclofenac is likely to have reduced the risk of pancreatitis by more than 50% in this case".
The court accepted Professor Gilmore's evidence that:
i) The five studies in Lyu et al in which diclofenac was administered rectally show a risk ratio of 0.38, giving a risk reduction of 62%, and therefore satisfy the balance of probabilities test.
ii) The seven studies in table 3s annexed to the European Guideline in which diclofenac was administered rectally show that the risk ratio or odds ratio was below 0.5 and therefore satisfy the balance of probabilities test.
Conclusion on "but for" causation:
The court rejected the Defendants' contention that this case should be judged by the scientific standard of a confidence interval of 95%.
The meta-studies, being medical papers, all consider the issue of the prevention of pancreatitis through the lens of the scientific standard of 95%.The court accepted Professor Gilmore's evidence that this is not the enquiry with which the Court was engaged. The Court was tasked with addressing the issue of whether if the Claimant had been administered 100 mg diclofenac rectally, as the Second Defendant intended, his pancreatitis would, on the balance of probabilities, have been avoided.
Following the guidance of Lady Hale in Sienkiewicz (supra), the Court must consider all the evidence when considering causation, including the studies and statistics, and the interpretation put on those studies and statistics by the gastroenterologists.
Following the guidance of Lord Nicholls in Gregg v Scott, "The value of the statistics will of course depend upon … the closeness of their position to that of the claimant."
The court found that the Court should consider statistics where diclofenac was administered rectally because that most closely fits the factual causation in the present case. Further, it was inappropriate to consider the studies involving the administration of NSAIDs orally, intravenously or intramuscularly because it was common ground between all three gastroenterologists that those routes of administration are ineffective and only the rectal route is effective.
The court found that it is inappropriate to consider statistics relating to NSAIDs without distinguishing between diclofenac and indomethacin for two reasons. Firstly, indomethacin was not the factual causation case which was before the Court. Secondly, if one considers the seven studies in table 3s of the European Guideline and the five studies in Lyu et al where diclofenac and indomethacin could be compared separately, it could be seen that diclofenac is more efficacious. The court found that the Lyu et al paper shows that in respect of the patients administered rectal diclofenac, the risk ratio was 0.38, giving a risk reduction of 62%. Using the European Guideline, the seven studies in which Diclofenac was administered rectally show that the risk ratio or odds ratio was below 0.5.
The court concluded that it was appropriate to rely upon the seven studies in Table 3s of the European Guideline and the five studies in Lyu where diclofenac was administered rectally as this is the factual causation case which is before the Court.
For the aforementioned reasons, the court concluded that the claimant had proved on the balance of probabilities that if he had been administered 100 mg diclofenac pre-operatively rectally, as the Second Defendant intended, it is likely that he would have avoided pancreatitis.
In the light of the finding on "but for" causation, it was unnecessary to consider material contribution. However, as the court was addressed on it, the judge made findings on this issue having regard to Bailey v Ministry of Defence [2008] EWCA Civ 883 and CNZ v Royal Bath Hospitals NHS Foundation Trust [2023] Med LR 59.
Material contribution
In a study by Patil and others entitled "Role of Rectal Diclofenac Suppository for Prevention and its Impact on Severity of Post-Endoscopic Retrograde Cholangiopancreatography Pancreatitis in High-Risk Patients" (Gastroenterol Res. 2016;9(2-3): 47-52), it is said,
"Various theories about pathogenesis of post-ERCP pancreatitis have been proposed. But the most accepted theory is mechanical trauma to papilla or pancreatic sphincter causing transient obstruction to outflow of pancreatic juice. … Regardless of mechanism, the cascade of events is initiated resulting in activation of proteolytic enzymes causing autodigestion of pancreas and impaired acinar secretion. This results in activation of inflammatory cascade causing both local inflammation and systemic effects [5, 6]. The interventions for prevention of post-ERCP pancreatitis aim at breaking this cascade. Non-steroidal anti-inflammatory drugs (NSAIDs) are potent inhibitor of phospholipase A2 which is thought to play a critical role in early inflammatory cascade [7]."
Professor Gilmore said in his report,
"It is also thought that an NSAID is likely to reduce the severity of post-ERCP pancreatitis, and so it seems more likely than not that its use would on the balance of probabilities have rendered it less severe if it had not been prevented."
The court accepted the evidence of Professor Gilmore that Diclofenac is,
"An anti-inflammatory and it works by locking the prostaglandins and the interleukins that fuel inflammation, then one would expect it to have an effect, a material effect, but we haven't got the data to show it".
In short, the three gastroenterologists were in agreement that diclofenac was on the balance of probabilities likely to have made a contribution that was more than negligible to reduce the inflammatory process which leads to pancreatitis.
If the court had found that in the case of a patient who was low to medium-risk, there was insufficient data to state on the balance of probabilities whether the administration of diclofenac rectally pre-procedure would have prevented pancreatitis, the court would have found that the Claimant had proved on the balance of probabilities that he suffered a material contribution.
The court found that the expert evidence of all three gastroenterologists in this case was that on the balance of probabilities:
i) The contribution of diclofenac, if given, would have been material and would have reduced the inflammation.
ii) That contribution cannot be apportioned between the negligent and non-negligent components.
iii) As a consequence of ii) above, the Claimant is entitled to 100% of his damages.
Further, for completeness' sake, the court found that the following evidence supported its finding that the contribution of diclofenac, if given, would have been material and would have reduced the inflammation.
Firstly, the court accept the submissions that
"The meta-analyses include data which would support an effect on severity of pancreatitis with administration of diclofenac which supports a material contribution, even where PEP is not avoided. In Liu table 2 supports a reduction in severe PEP with RR of 0.45 for diclofenac and 0.45 overall (all routes all drugs). Individual studies all demonstrate benefit of more than 50%: see Patil for instance".
The court noted that the subgroup involved all routes of administration and all NSAIDs (diclofenac, indomethacin and other).
Secondly, the court noted that the European Guideline, the American Guideline and the Japanese Society of Hepato-Biliary-Pancreatic Surgery all recommend routine rectal administration of 100 mg of diclofenac or indomethacin in all cases, whether high, medium or low risk. The second point above is supported by the paper by Lyu et al, in which it is said (G/64/p. 157-158),
"The definition of risk of ERCP were varied in the included studies. However, we accepted the original author's classifications. In this study, we found no difference in the incidence of PEP between average-risk and high-risk patients. A similar conclusion has been drawn in the studies of Shen et al. and Patai et al. This conclusion provides the basis for the recommendations of the ESGE and Japanese Society of Hepato-Biliary-Pancreatic Surgery in its guidelines for the prevention of PEP."
Thirdly, in an individual study by Patil and others entitled "Role of Rectal Diclofenac Suppository for Prevention and its Impact on Severity of Post-Endoscopic Retrograde Cholangiopancreatography Pancreatitis in High-Risk Patients" (Gastroenterol Res. 2016;9(2-3): 47-52), it is said,
"Methods: We conducted a single-centre, prospective, open-labelled, randomized trial for evaluating the use of rectal diclofenac in prevention of post-ERCP pancreatitis in high-risk patients. We assessed 526 patients coming for ERCP for different indications. Four hundred patients were eligible for the study. Those not fitting the high-risk criteria and with acute pancreatitis were excluded. These patients were randomized in two groups: 200 patients received rectal diclofenac prior to or during the procedure, while 200 patients received placebos. …"
…
Conclusion: Rectal diclofenac prior to or during ERCP in high-risk patients reduces the incidence as well as severity of post-ERCP pancreatitis compared to placebo."
The Patil paper was not included in table 3s of the European Guideline or in Lyu et al. There was no evidence before the Court as to why it was not included and the court found that it is a paper that could be taken into account, although it did no more than confirm a finding that the court already made that the Claimant satisfied on the balance of probabilities the material contribution test.
Conclusion
The court ordered that:
i) There be judgment for the Claimant against the Second Defendant in the sum of £550,000.
ii) The claim against the First Defendant be dismissed.
References
Buxbaum JL, Freeman M, Amateau SK, Chalhoub JM, Coelho-Prabhu N, Desai M, Elhanafi SE, Forbes N, Fujii-Lau LL, Kohli DR, Kwon RS, Machicado JD, Marya NB, Pawa S, Ruan WH, Sheth SG, Thiruvengadam NR, Thosani NC, Qumseya BJ; (ASGE Standards of Practice Committee Chair). American Society for Gastrointestinal Endoscopy guideline on post-ERCP pancreatitis prevention strategies: summary and recommendations. Gastrointest Endosc. 2023 Feb;97(2):153-162. doi: 10.1016/j.gie.2022.10.005. Epub 2022 Dec 12. PMID: 36517310.
Dumonceau JM, Andriulli A, Elmunzer BJ, Mariani A, Meister T, Deviere J, Marek T, Baron TH, Hassan C, Testoni PA, Kapral C; European Society of Gastrointestinal Endoscopy. Prophylaxis of post-ERCP pancreatitis: European Society of Gastrointestinal Endoscopy (ESGE) Guideline - updated June 2014. Endoscopy. 2014 Sep;46(9):799-815. doi: 10.1055/s-0034-1377875. Epub 2014 Aug 22. PMID: 25148137.
Khoshbaten M, Khorram H, Madad L, Ehsani Ardakani MJ, Farzin H, Zali MR. Role of diclofenac in reducing post-endoscopic retrograde cholangiopancreatography pancreatitis. J Gastroenterol Hepatol. 2008 Jul;23(7 Pt 2):e11-6. doi: 10.1111/j.1440-1746.2007.05096.x. Epub 2007 Aug 7. PMID: 17683501.
Lua GW, Muthukaruppan R, Menon J. Can Rectal Diclofenac Prevent Post Endoscopic Retrograde Cholangiopancreatography Pancreatitis? Dig Dis Sci. 2015 Oct;60(10):3118-23. doi: 10.1007/s10620-015-3609-9. Epub 2015 Mar 11. PMID: 25757446.
Lyu Y, Cheng Y, Wang B, Xu Y, Du W. What is impact of nonsteroidal anti-inflammatory drugs in the prevention of post-endoscopic retrograde cholangiopancreatography pancreatitis: a meta-analysis of randomized controlled trials. BMC Gastroenterol. 2018 Jul 4;18(1):106. doi: 10.1186/s12876-018-0837-4. PMID: 29973142; PMCID: PMC6032784.
Patil S, Pandey V, Pandav N, Ingle M, Phadke A, Sawant P. Role of Rectal Diclofenac Suppository for Prevention and Its Impact on Severity of Post-Endoscopic Retrograde Cholangiopancreatography Pancreatitis in High-Risk Patients. Gastroenterology Res. 2016 Jun;9(2-3):47-52. doi: 10.14740/gr672w. Epub 2016 Jun 18. PMID: 27785324; PMCID: PMC5040543.