Skip to content
laws.sgby aturio
Grounds publishedDecided 24 Aug 2016
Contents

General Division of the High Court

Koo Quay Keong (Administrator of the Estate of Lee Lee Chan, Deceased) v Ooi Peng Jin London Lucien

[2016] SGHC 168Case no. Suit No 714 of 2014

What happened, in plain English

There is no plain-English summary of this judgment yet.

Ask about this judgment

Grounds of decision

Official text from eLitigation

  1. Introduction

  2. [1] This is a claim in negligence regarding the post-operative care of a patient who underwent a surgical procedure known as a Whipple operation. Unfortunately, the patient encountered post-operative complications and eventually passed away. The Defendant was the surgeon who performed the operation, and was in charge of the patient's post-operative care.

  3. [2] The Plaintiff alleges that the Defendant ought to have, at various time-points after the operation, suspected or diagnosed a post-operative complication and performed a Computed Tomography (“CT”) scan of the patient’s chest, abdomen, and pelvis (an “abdominal CT scan”). This would have led the Defendant to perform certain interventional procedures earlier than was in fact done. Those procedures would have saved the patient's life.

  4. [3] The Defendant submits that there was no reason, at the time-points put forward by the Plaintiff, to suspect or diagnose the complication, and, in any event, it was reasonable not to have performed the abdominal CT scan earlier. Moreover, the Defendant denies that an earlier CT scan would have altered the course of the patient's post-operative management.

  5. Background

  6. [4] The deceased, Mdm Lee Lee Chan (“Mdm Lee”), was 59 years of age at her passing on 28 July 2011. The Plaintiff is the widower of Mdm Lee and is also the administrator of her estate. The Defendant, Dr London Lucien Ooi Peng Jin, is a Senior Consultant Surgeon at Singapore General Hospital (“SGH”) specializing in hepato-pancreato-biliary (“HPB”) surgery, ie, surgeries involving the liver, pancreas, and bile ducts.

  7. [5] On 4 July 2011, Mdm Lee underwent a Whipple operation performed by the Defendant to remove a tumour on the head of her pancreas. While under post-operative care, she suffered several complications, and passed away on 28 July 2011. Her final cause of death was as follows:

  8. [6] The Plaintiff now claims against the Defendant in negligence in his capacity as the administrator of Mdm Lee’s estate.

  9. [7] All material events took place in July 2011. For brevity, I will omit the year in subsequent descriptions of Mdm Lee's treatment, and will where appropriate identify treatment dates by their post-operative day (“POD”) numbers, ie, POD1 (4 July) through POD25 (28 July).

  10. Whipple operation

  11. [8] The pancreas is an organ located behind the stomach. At its head is the duodenum (ie, the first section of the small intestines) and the bile duct (ie, the tube joining the liver to the pancreas).

  12. [9] In a Whipple operation, the pancreatic head, the gallbladder, and part of the bile duct, stomach, and small intestines are removed. The remnants of the pancreas, bile duct, and stomach are then anastomosed (ie, reconnected) to the gastrointestinal tract to facilitate normal digestion and absorption of food.

  13. [10] Mdm Lee had three anastomoses during her Whipple operation:

  14. Allegations and issues

  15. [11] In his Statement of Claim (Amendment No 2) (the “SOC”), the Plaintiff alleged that the Defendant:

  16. [12] At or around the beginning of the trial, the Plaintiff narrowed his claim to a single allegation, ie, that the Defendant failed to offer timely and appropriate care between the Whipple operation and two subsequent surgeries.

  17. [13] The Plaintiff submitted that by 7 July (POD3), Mdm Lee had suffered “an anastomotic leak of the anastomotic sites of the Whipple Operation”, which the Defendant failed to investigate, diagnose, and treat timeously. By the time the anastomotic leak was diagnosed on 17 July, it had deteriorated severely. Thereafter, despite knowing of the anastomotic leak, the Defendant failed to manage Mdm Lee appropriately. Up until 24 July, he failed to investigate whether her fluid collections were adequately drained by the existing abdominal drains, and to intervene via a percutaneous drainage procedure (ie, inserting a small tube through the skin to drain fluid in the abdomen) into any undrained fluid collections. By the time the Defendant decided to perform a percutaneous drainage procedure on 24 July, Mdm Lee’s condition had become unsalvageable.

  18. [14] The Defendant submitted that Mdm Lee did not have an anastomotic leak until 2300hrs on 16 July, and was managed appropriately both before and after that time:

  19. (a) Before 2300hrs on 16 July, it sufficed for the Defendant to have simply suspected, but neither investigated nor diagnosed, an anastomotic leak. In any event, his management and treatment of Mdm Lee would have been substantially the same even if she had been diagnosed with an anastomotic leak.

  20. (b) Following the occurrence of an anastomotic leak at 2300hrs on 16 July, it was appropriate for the Defendant to have managed Mdm Lee with non-surgical measures. She was too unstable for invasive intervention, whether in the form of a laparotomy (ie, opening up the abdomen to wash it out and close the disrupted anastomosis), or a percutaneous drainage procedure to remove the undrained fluids in the abdomen. Hence, she was managed non-surgically to optimise her condition for the definitive intervention of a laparotomy.

  21. The law on negligence

  22. [15] The applicable legal principles are not in dispute. In Singapore, the leading authority on medical negligence is Khoo James and another v Gunapathy d/o Muniandy and another appeal [2002] 1 SLR(R) 1024 (“Gunapathy”). The test is as stated in Bolam v Friern Hospital Management Committee [1957] 1 WLR 582 (“Bolam”) at 587:

  23. The parties and their witnesses

  24. [18] The Plaintiff called four witnesses:

  25. [19] The Defendant called eight witnesses:

  26. Suspicion of anastomotic dehiscence: 7 to 16 July

  27. Post-operative Pancreatic Fistula

  28. [20] On 7 July (POD3), Mdm Lee had amylase levels of 1,361U/L and 1,942U/L respectively in her right and left operatively-inserted drains. These levels were respectively 7.3x and 10.4x of her normal serum amylase level of 186U/L. Citing Claudio Bassi et al, “Postoperative pancreatic fistula: an international study group (ISGPF) definition” (2005) 138(1) Surgery 8 (the “Bassi Article”), the Plaintiff submitted that, in view of these results, the Defendant should have suspected or diagnosed an anastomotic leak. This was initially denied by the Defendant.

  29. [21] A patient with a POD3 drain amylase level of over 3x her normal serum amylase level has, by definition, a post-operative pancreatic fistula (a “POPF”). This authoritative test for a POPF is found in the Bassi Article at p 10 (“the ISGPF Test”):

  30. Because there was much argument over the meaning of the Bassi Article, I have annexed a copy of it to this Judgment.

  31. [22] Based on her drain amylase level on 7 July, Mdm Lee had a POPF. Although Dr Mak accepted this, he did not equate the POPF with an anastomotic leak. He added only that the prognosis on 7 July was that the POPF would resolve spontaneously. The disputes between the parties were thus as follows:

  32. (a) whether a POPF was to be equated with an anastomotic leak in the context of a Whipple operation; and

  33. (b) if so, whether the Defendant should at least have suspected that Mdm Lee might have had such an anastomotic leak given her drain amylase levels on 7 July.

  34. [23] The Plaintiff submitted that in the accepted medical literature, the terms “leak” and “POPF” are synonymous and interchangeable. Accordingly, Mdm Lee had by 7 July suffered from, and should have been diagnosed with, an anastomotic leak. He referred to the following passage from the Bassi Article (at p 10) to support his contention:

  35. [24] The Defendant submitted that the terms “POPF”, “anastomotic dehiscence”, and “anastomotic leak” are distinct. Even if Mdm Lee had a POPF on 7 July, she did not necessarily have an anastomotic leak. I set out the Defendant’s definitions of these terms:

  36. (a) POPF: a leakage of pancreatic fluid into the abdominal cavity that results in drain output on or after POD3 with an amylase content of over 3x the serum amylase activity, without any indication of the mechanism by which the leakage actually occurred.

  37. (b) Anastomotic dehiscence: a rupture or disruption in an anastomosis (the PG anastomosis in Mdm Lee’s case) that results in a gap in the anastomosis through which contents of the stomach and pancreas can leak into the abdominal cavity.

  38. (c) Anastomotic leak: an anastomotic dehiscence that results in leakage of the contents of the stomach and pancreas into the abdominal cavity.

  39. Since a anastomotic dehiscence is the mechanism through which an anastomotic leak occurs, I will refer to both the complications of an anastomotic leak and an anastomotic dehiscence simply as a “Dehiscence”.

  40. [25] The Defendant claimed in his closing submissions that he had from POD3 (7 July) suspected a Dehiscence based on Mdm Lee’s elevated drain amylase level, even though there was no need to have documented it:

  41. [26] This submission stood in stark contrast to the Defendant’s pleadings and evidence-in-chief. In the Defence, Mdm Lee’s elevated POD3 drain amylase levels were omitted entirely from the Defendant’s account of his observations and decisions from 4 to 16 July. This was even though the SOC had described the drain fluid amylase level of 1,361U/L (for the right drain) on 7 July, as a finding that “reflected a clear possibility of a pancreatic leak.”

  42. [27] As chronicled in the Defence, Mdm Lee was first observed for indications of a Dehiscence late in the night of 16 July when she complained of sudden abdominal pain. An abdominal CT scan in the morning of 17 July showed a possible Dehiscence. Only thereafter was she “diagnosed” with “septic shock likely secondary to anastomotic leak”. The Defendant then pleaded that “[o]n 17 July 2011, an anastomotic leak was suspected.” This suggested that he did not even suspect a Dehiscence until 17 July.

  43. [28] The Defendant pleaded also that from 4 to 16 July, “[t]he volume of fluid drained from [Mdm Lee’s] abdominal drains, the fluid amylase levels, and the occasional pain were not unusual.” He stated also in his affidavit of evidence-in-chief (“AEIC”) that her drain amylase level on 7 July “was slightly elevated but was not unusual”. These were remarkable statements. The fluid amylase levels in [Mdm Lee’s] right and left drains on 7 July were respectively 7.3x and 10.4x her normal serum amylase levels (see [20] above). These far exceeded the [3x] threshold for identifying the existence of a POPF under the Bassi Article. Even Dr Mak described Mdm Lee’s drain amylase levels on POD3 as “high”.

  44. [29] The Defendant stated further in his AEIC that amylase-rich pancreatic fluid spilt during the Whipple surgery “may not have been totally washed away during surgery … will persist for a few days post-operatively and … therefore be measureable or detectable in the abdominal drains.” It was thus “more relevant to monitor for a possible anastomotic leak through observing the colour and volume of the drain [fluids].” For Mdm Lee, “the colour of the drain fluid was not worrisome or suggestive of any form of anastomotic leak until 17 July”.

  45. [30] In his oral evidence, the Defendant put forward two additional explanations for Mdm Lee’s elevated POD3 drain amylase level: (i) seepage of pancreatic fluid through the suture holes along anastomoses, and (ii) release of amylase-rich exudate due to pancreatitis or irritation of the pancreas. In support of these contentions, the Defendant cited Vladimir D Dugalic et al, “Drain amylase value as an early predictor of pancreatic fistula after cephalic duodenopancreatectomy” (2014) 20(26) World Journal of Gastroenterology 8691 (the “Dugalic Article”) at 8695.

  46. [31] Hence, the Defendant offered four explanations in total for Mdm Lee’s elevated POD3 drain amylase level: (i) retention of pancreatic fluid spilt during surgery; (ii) seepage of pancreatic fluid through anastomotic sutures; (iii) release of amylase-rich exudate due to pancreatitis or irritation of the pancreas, and (iv) a Dehiscence. I will refer to the first three explanations as the “benign explanations”.

  47. [32] Although the Bassi Article acknowledged (at p 8) that, conceptually, a POPF describes not only a Dehiscence but also a parenchymal leak not directly related to an anastomosis, it was written with the objective of producing a universal test to identify a Dehiscence. The focus of the research was on the link between an elevated POD3 drain amylase level and the existence of a Dehiscence after pancreatic resection. In my view, the Bassi Article was written to set out an objective international biochemical test to diagnose, or at least to suspect, the existence of a Dehiscence. Although the existence of a POPF does not conclusively prove the existence of a Dehiscence, the Bassi Article treated the two concepts synonymously. Given the undisputed authoritativeness of the Bassi Article on the post-operative management of patients after a Whipple operation, I find that the Defendant should have at a minimum suspected the existence of a Dehiscence based on the fact that Mdm Lee had, by definition, a POPF on 7 July.

  48. [33] The benign explanations for Mdm Lee’s POPF put forth by the Defendant (at [31] above) did not assist him. As Dr Mak clarified, one would expect the benign explanations for an elevated drain amylase level to be “self-limited” and settle within a few days of a Whipple operation. The Bassi Article advocated conducting the drain amylase test for identifying a POPF “on or after [POD3] simply because one hopes that by the third day, you might exclude some of the other possible causes for amylase being present in the fluid.” This was also the view of Dr Nikfarjam, who opined that “stitch holes often seal up, and they don't persist beyond day 3. Anastomotic leaks, on the other hand, of any significance would persist.”

  49. [34] In other words, the reason for measuring the drain amylase level on POD3, and not earlier, is precisely to exclude the benign explanations as likely causes of an elevated drain amylase level. Further, although the Defendant relied on the benign explanations to account for Mdm Lee’s elevated POD3 drain amylase level, he could not exclude a Dehiscence as a possible cause of the same. Indeed, based on the Bassi Article, a Dehiscence was not simply a possible but rather a likely cause of her elevated POD3 drain amylase level. Hence, the Defendant should have at least suspected that Mdm Lee had a Dehiscence.

  50. [35] Although the Defendant could not overtly deny that the Bassi Article was relevant and applicable to the circumstances at hand, he downplayed its utility during the trial and in his submissions.

  51. [36] First, the Defendant referred to Yi-Ming Shyr, et al, “Does drainage fluid amylase reflect pancreatic leakage after pancreaticoduodenectomy?” (2003) 27(5) World Journal of Surgery 606, which appeared to suggest that elevated drain amylase levels do not indicate a Dehiscence. However, this article was written two years before the Bassi Article, and its findings had been considered by the authors of the Bassi Article when they formulated their definition of a POPF (at p 11, footnote 49). More importantly, the Defendant had in his submissions accepted the Bassi Article as authoritative.

  52. [37] Secondly, the Defendant submitted that the Bassi Article was intended purely to facilitate comparative study of surgical experiences. However, the Bassi Article “suggest[ed] that use of [the] proposed definition of POPF will confirm its clinical value and allow accurate comparison of different surgical experiences” (at p 12). I note that the Defendant’s submissions quoted only the latter part of this sentence (which referred to the accurate comparison of surgical experiences), but omitted any reference to the former part of the sentence (which confirmed the “clinical value” of the ISGPF Test).

  53. [38] Thirdly, the Defendant relied on Daniel J Moskovic, et al, “Drain data to predict clinically relevant pancreatic fistula” (2010) 12 Journal of the International Hepato-Pancreato-Billary Association 472 to show that elevated drain amylase levels are not of themselves clinically significant. However, the Bassi Article ascribed no clinical significance to the bare fact of an elevated POD3 drain amylase level. Instead, it proposed a three-tier ex post facto grading of the clinical significance of a POPF, with Grade A having no clinical impact (see [46] below).

  54. [39] Finally, the Defendant suggested that the volume and colour of the drain fluids were more important indicators of a Dehiscence than an elevated POD3 drain amylase level (see [29] above). However, the Bassi Article had examined the diagnostic value of the volume and quality of drain output before formulating its objective test for a POPF/Dehiscence by reference to the POD3 drain amylase level (at p 11). This is not to say that the volume and colour of the drain fluid are unimportant, but only that the absence of drain fluid of a large volume or a sinister colour does not detract from the possibility of a Dehiscence when the requirements of the ISGPF Test (at [21] above) are satisfied.

  55. [41] I would add only that the Plaintiff’s oral elaboration of his opening statement added to the confusion. It conflated the criteria for identifying the existence of a Dehiscence with that for assessing its severity. I set out the statement of counsel for the Plaintiff, Mr Tan Chee Meng, SC (“Mr Tan”).

  56. [42] Dr Nikfarjam opined that a Dehiscence could be identified from an elevated POD3 drain amylase level alone. He took the view that Mr Tan’s statement above, that “there are other criteria that one looks at to ensure that [the] possibility [of a leak] is minimised”, was more accurately a statement that the severity of a leak could not be determined solely based on an elevated POD3 drain amylase level.

  57. [43] On the other hand, Dr Beckingham accepted that Mdm Lee’s elevated POD3 drain amylase level gave rise only to a suspicion, and not necessarily a diagnosis, of a Dehiscence.

  58. [44] For this case, the distinction between a suspicion and a diagnosis of a Dehiscence is not material. Both of the Plaintiffs’ experts testified that a Dehiscence is the most frequent major complication after pancreatic surgery. Dr Mak acknowledged that a Dehiscence is the most severe cause of a POPF, “the one which we as surgeons are obliged to consider as our worst case scenario.” I find that it was thus unreasonable for the Defendant to fail to even suspect a Dehiscence based on Mdm Lee’s elevated POD3 drain amylase level. In doing so, he fell below the standard of care expected of him.

  59. [45] I note that Dr Mak opined that “the possible [POPF] that was discovered on [7 July] was not directly related to the later anastomotic dehiscence” that the Defendant eventually diagnosed on 17 July (see [121] below). What Mdm Lee had on 7 July was a “transient or biochemical pancreatic fistulae” that settled spontaneously. Hence, “there was not any dehiscence of the [PG] anastomosis before 16th July”. However, the Defendant did not mention a transient or biochemical pancreatic fistulae in his evidence. Besides, as Dr Beckingham explained (by reference to the Dugalic Article (at p 8697)), the concept of “a fistula that wasn't there to start with and then developing later … after the 5th post-operative day” was doubtful. In any event, Dr Mak conceded that “it [was] difficult to determine, if the fistulae had closed by 16th July”. Accordingly, I do not accept the suggestion that Mdm Lee had on 7 July a [transient] POPF that settled shortly thereafter and that was unrelated to the Dehiscence diagnosed on 17 July. Instead, she likely had a Dehiscence from 7 July. Although this Dehiscence was not clinically significant on 7 July, it deteriorated and had become much worse by 16 and 17 July.

  60. Post-operative management

  61. [46] The Plaintiff’s pleaded case was that an abdominal CT scan should have been performed before 17 July to investigate Mdm Lee’s condition. Had it been done, it would have confirmed the existence of a Dehiscence and revealed a need for intervention: specifically, a percutaneous drainage procedure to drain the collections of fluid in the abdomen that were inadequately drained by the existing operatively-inserted drains.

  62. [47] As a result, Mdm Lee did not have an abdominal CT scan until 17 July, by which time it was “too late” for any intervention to have changed her outcome:

  63. It bears mention that the last sentence is not an alternative plea, and therefore also refers to the same deadline of 17 July for the ordering of a full abdominal CT scan.

  64. [48] In my view, the mere fact that, on 7 July, Mdm Lee had a POPF shed little light on how she should have been managed thereafter. The Bassi Article cautioned that the ISGPF Test was “so inclusive that many asymptomatic patients that fit the definition of POPF may not be clinically ill”. It thus proposed a three-tier grading system for POPF (Grades A, B, C), depending on the extent to which the patient deviated from the normal post-operative course. Further, the grade of severity has little predictive value because it can be determined only retrospectively, ie, at the end of the patient’s post-operative course, whether discharge from the hospital or death (at pp 10–11).

  65. [49] Although the Defendant fell below the standard of care expected of him in failing to even suspect a Dehiscence based on the elevated POD3 drain amylase level, he will not be liable in negligence unless it is shown that this omission caused the injuries suffered by Mdm Lee. Specifically, the Plaintiff must prove: (i) that the Defendant should have managed her differently if he had suspected a Dehiscence from 7 July, and (ii) that a different management would have changed the outcome of her case. Since the Plaintiff did not plead a loss of chance, this must be proved on a balance of probabilities.

  66. [50] Before any evidence was led on the first day of trial, the Defendant applied to amend his Defence to clarify that there would have been little change to Mdm Lee’s management even if a Dehiscence had been diagnosed before 17 July:

  67. [51] I turn now to the post-operative management of Mdm Lee from 7 July according to the time-frames described in the Plaintiff’s submissions.

  68. 7 to 10 July: initial recovery

  69. [52] Dr Beckingham testified that “the earliest point at which it would have been reasonable to consider further intervention would be 11 July, and probably the latest point at which we should have been considering it was 17 or 18 July.” Further, such intervention would have been performed only “on the assumption that you [have] do[ne] a CT scan which shows you something significant enough to warrant percutaneous drainage.” Although Dr Beckingham subsequently retracted his evidence that percutaneous drainage could have been performed as late as 17 or 18 July, he nevertheless refrained from criticising Mdm Lee’s management before 11 July.

  70. [53] Dr Nikfarjam did not challenge Mdm Lee’s management prior to 11 July. From 7 to 8 July, despite her elevated POD3 drain amylase level, she had improved since her Whipple operation and was not unwell. It was thus reasonable for the Defendant to have managed her the way he did. On 9 July, Mdm Lee experienced respiratory compromise (ie, breathing difficulties) and shortness of breath, and drained “haemoserous” fluid. It was thus appropriate to have focussed on controlling these complications on 9 and 10 July.

  71. [54] Accordingly, I am of the view that the Defendant did not fall below the standard of care expected of him in his management of Mdm Lee from 7 to 10 July.

  72. 11 to 13 July: pleural effusion, free fluid in abdomen, and other factors

  73. [57] On 12 July, 2,360ml of abdominal fluid was drained from Mdm Lee. Her ABG level was pH7.301 at 0213hrs (the “12 July 0213hrs ABG”) and pH7.323 at 1108hrs (the “12 July 1108hrs ABG”). She was acidotic but improving.

  74. [58] On 13 July, 2,500ml of abdominal fluid was drained from Mdm Lee.

  75. [59] The Plaintiff submitted that by 11 July (POD7), Mdm Lee had deviated from the usual clinical course after a Whipple operation. The clinical signs and symptoms she displayed necessitated a CT scan of her abdomen to eliminate the possibility of a Dehiscence and to investigate whether further intervention was necessary. Additionally, the clinical indicators that emerged between 11 and 12 July should have heightened the Defendant’s suspicion of a Dehiscence, and alerted him to the need for such an abdominal CT scan. The Plaintiff highlighted 10 indicators when cross-examining the Defendant and Dr Mak, but referred to a total of 12 indicators in his submissions:

  76. [60] The Defendant agreed, in cross-examination and in submission, that Mdm Lee had by 12 July displayed these 10 or 12 indicators. However, he testified that not all of the indicators persisted at the same time. Some had resolved, while others had been attributed to causes other than a Dehiscence. Once an indicator was accounted for, it was “dropped from the equation” and “we move[d] on.” Thus, it was incorrect to manage Mdm Lee by “tak[ing] the ten [or twelve] symptoms and signs and throw[ing] them into one time frame. That is not the way you practice medicine.” I do not understand the Plaintiff’s experts to have been propounding a different approach. Nevertheless, the parties disagreed on the extent to which each indicator had resolved.

  77. [61] The Plaintiff made detailed submissions only on the haemoserous fluid, pleural effusion, persistent low albumin, acidosis, free abdominal fluid, and abdominal distension. I will first examine these indicators individually, and then consider whether Mdm Lee’s management should have been changed in the light of all the indicators, viewed globally.

  78. Pleural effusion

  79. [62] The Defendant in his oral evidence-in-chief attributed the pleural effusion to Mdm Lee’s low albumin level, which allowed fluid to leak out of her blood vessels and into the [third] space in her chest. The fact that she had edema even of her extremities (see [55] above) reinforced the conclusion that the third-spacing was so extensive as to have caused the pleural effusion. This was exacerbated by her atelectasis, which had in turn been worsened by her lack of mobilisation. Given the absence of abdominal complaints, all indicators pointed to the chest – and away from the abdomen – as the source of the pleural effusion.

  80. [63] Dr Mak accepted that a Dehiscence could contribute to fluid in the abdomen, which could in turn lead to a pleural effusion. Given Mdm Lee’s highly elevated POD3 amylase level, the pleural effusion could not have been due only to her low albumin state. The possibility of a Dehiscence must also have been addressed. However, any concerns about a Dehiscence were assuaged by the significant improvement in her clinical condition after her chest was drained.

  81. [64] Dr Nikfarjam opined that on 11 July, the breathing difficulties increasingly experienced by Mdm Lee, coupled with the associated development of a significant pleural effusion, should have raised concerns of a significant intra-abdominal complication. Large pleural effusions rarely occur after a Whipple operation in a patient who, like Mdm Lee, has no significant pre-existing heart failure, chronic liver disease, malnutrition, or chronic kidney disease. However, Dr Nikfarjam then accepted that it was “reasonable to monitor what happens after 11 July to [determine] whether there might be a clinically significant anastomotic leak,” and that on “12 and 13 July [Mdm Lee’s] clinical condition had improved after the drainage of the pleural effusion on the 11th."

  82. [65] Dr Beckingham testified that the pleural effusion was caused by the presence of pancreatic fluid [from a Dehiscence] within the abdominal cavity, which "irritate[d] the pleural viscera directly above the diaphragm.” He cited Mark I van Berge Henegouwen et al, “Incidence, Risk Factors, and Treatment of Pancreatic Leakage after Pancreaticoduodenectomy: Drainage versus Resection of the Pancreatic Remnant” (1997) 185(1) Journal of the American College of Surgeons 18 at pp 21–23, which found that at least 48.3% (14 of 29) of patients with a Dehiscence developed a pleural effusion, and that a pleural effusion pointed to a Dehiscence.

  83. [66] Dr Mak explained that, statistically, the fact that 48.3% of patients with a Dehiscence developed a pleural effusion did not mean that there was a 48.3% chance that Mdm Lee had a Dehiscence by virtue of her pleural effusion. The study (at [65] above) considered only the incidence but not the cause of the pleural effusions in the patients studied. It was unreliable since pneumonia, infections of the chest or lung, and low albumin states were more common causes of a pleural effusion than a Dehiscence.

  84. Acidosis

  85. [67] In his AEIC, the Defendant attributed Mdm Lee’s 11 July ABG results to a “respiratory acidosis”. In his oral evidence-in-chief, however, he amended his explanation of the 11 July ABG results to a “metabolic acidosis possibly secondary to a respiratory cause”. He elaborated that while acidosis from a lung failure was “respiratory acidosis”, acidosis from a lung problem that impaired the transport of oxygen to the tissues and thereby caused the tissues to produce acid was “metabolic acidosis”.

  86. [68] Dr Mak accepted that Mdm Lee had on 11 July suffered metabolic acidosis. The fluid that had accumulated in her chest from her pleural effusion made breathing difficult and prevented the lungs from getting enough oxygen to her tissues. The 11 July ABG result was “quite abnormal”. He thus “ha[d] to exclude various possibilities … for example, of infection and sepsis.” However, he remained “unsure simply on the basis of looking at this test result whether in fact that problem with breathlessness has also led and contributed to this abnormal state where her acidosis is.”

  87. [69] Dr Nikfarjam dismissed any relation between the metabolic acidosis and the lungs. He explained that Mdm Lee’s partial pressure of carbon dioxide (PCO-2) levels, which would have increased if her lungs had been unable to fully expel the carbon dioxide in her body, had remained normal. Since her lungs were working normally, they could not have caused the acidosis.

  88. [70] The Plaintiff submitted that the Defendant’s belated amendment of his evidence-in-chief (see [67] above) showed that he had “failed to apply his mind to properly evaluating the causes of the acidosis”. Instead, he had “jumped to the conclusion that it must have been a result of the pleural effusion at the time.” The metabolic acidosis indicated an intra-abdominal problem, which undermined attempts to localise the clinical indicators to a chest problem.

  89. [71] However, the significance of the 11 July ABG result was not fully borne out in the evidence of the Plaintiff’s experts. Dr Beckingham did not mention the 11 July ABG result in his evidence-in-chief or cross-examination, and discussed it only at the end of his re-examination when Mr Tan brought it to his attention. Dr Nikfarjam agreed that Mdm Lee’s management on 11 July was appropriate, and that any differences between how he would have managed Mdm Lee and how the Defendant had in fact managed her constituted “an acceptable standard of variation.”

  90. [72] It was only after counsel for the Defendant, Ms Kristy Tan (“Ms Tan”), referred Dr Nikfarjam to the 11 July ABG results that he expressed concern about them. Dr Nikfarjam explained that “if [he] had seen this blood test, [he] would have actually been more worried that there's an intraabdominal problem,” and “would have been more concerned that an imaging test should have been done on the 11th, of the abdomen.” This was not convincing. If the 11 July ABG results were as important as he was belatedly making them out to be, he would not have neglected them earlier in his evidence.

  91. Free fluids in abdomen

  92. [74] The Defendant attributed the free fluid above the liver displayed by the 11 July chest CT scan to Mdm Lee’s low albumin state, which facilitated third-spacing in the abdominal cavity. The free fluid was of little concern because it could be drained by her existing abdominal drains. Dr Mak agreed with this assessment, and added that there was in fact no significant collection of fluid shown on the [albeit limited] set of images taken in the 11 July chest CT scan.

  93. [75] The Plaintiff did not dispute Dr Mak’s opinions in his submissions.

  94. [76] Dr Beckingham referred only to a “2-to-3-cm rim of fluid around [Mdm Lee’s] liver which has not been drained”, but gave no evidence as to its significance. This fell short of establishing the Plaintiff’s submission that there was “significant fluid under her diaphragm”, particularly since Dr Nikfarjam had accepted that the free fluid above the liver could have been drained out by the existing abdominal drains.

  95. Haemoserous fluid, persistent low albumin, and abdominal distension

  96. [77] The Defendant gave evidence that haemoserous fluid was a very non-specific sign of a Dehiscence. It indicated only that blood had been mixed with serous fluid. The observations of haemoserous fluid between 9 and 11 July were not accompanied by any abdominal signs, except for a slight abdominal distension recorded at 1600hrs on 11 July that subsequently settled. Further, the drain fluid became serous shortly thereafter. Dr Mak echoed these observations, and emphasised that Mdm Lee’s abdomen would have exhibited signs of disquiet had there been a Dehiscence. The Plaintiff’s submissions did not refer to any evidence from his own experts on the issue of the haemoserous fluid.

  97. [78] The Defendant testified that low albumin was similarly a very non-specific sign of a Dehiscence. Although Dr Mak accepted that a suspicion of a Dehiscence had to be entertained if a low albumin state persisted, he attributed Mdm Lee’s low albumin state from 11 to 14 July to her “somewhat conservative” albumin supplementation of 100ml/day and to her continued fasting. He added that the steady rise in her albumin levels from 11 to 14 July pointed further away from a Dehiscence, which would have affected them adversely. Dr Beckingham accepted that a low albumin state was common after a Whipple operation because a patient would not have been fed for three to four days. Dr Nikfarjam testified only that low albumin was “a reflection of a patient becoming unwell … rather than a true cause”, and was “indicative of an inflammatory process that’s going on.”

  98. [79] The Defendant explained that the slight abdominal distension would have been worrying only if it was accompanied by impaired bowel function, which could have suggested a collection of fluid in the space behind the lining of the abdominal cavity. Mdm Lee had normal bowel functions, which Dr Nikfarjam accepted was reassuring. Once again, the Plaintiff’s submissions did not refer to any evidence from his own experts on the issue of the abdominal distension.

  99. Management

  100. [81] Dr Nikfarjam agreed that the Defendant’s management of Mdm Lee on 11 July was appropriate. He also agreed that her clinical condition had improved on 12 and 13 July. However, he then backtracked and emphasised the 11 July ABG results (see [71] and [72] above).

  101. [83] It was only at the close of his re-examination that Dr Beckingham retracted this evidence and asserted that an abdominal CT scan should have been done on 11 July (see [52] above). I place little weight on this late retraction and assertion.

  102. [84] On balance, I am of the view that the Defendant did not fall below the standard of care expected of him in his management of Mdm Lee from 11 to 13 July.

  103. 14 to 16 July: abdominal pain

  104. [85] On 14 July, Mdm Lee had a low albumin level of 22g/L and drained “chylous” fluid. At 1000hrs, she complained of severe pain over a wound site, and received 75mg of Pethidine, a painkiller. At 1100hrs, she had pain with slight tenderness around her right drain site. At 1200hrs, she had mild abdominal pain with a soft abdomen and generalised tenderness on palpitation. However, she had no guarding, rebound tenderness, or peritonitis. At 1347hrs, her C-reactive protein (“CRP”) value was 135mg/L while her WBC count was 11.53 x 109/L. At 1620hrs, her pain had become “much better” and she was alert and comfortable.

  105. [86] On 15 July, the colour of Mdm Lee’s drain fluid was “light haemoserous” at 0600hrs and “yellowish” at 1520hrs. Across the morning, she drained 1,510ml and 20ml of fluid from right and left abdominal drains respectively. At 0810hrs, she was alert, comfortable, and had no complaints of pain, normal parameters, and a soft and non-distended abdomen. At 1000hrs, she was well enough for physiotherapy, and performed ankle pumps and marched on the spot. When the Defendant saw her at 1320hrs, she was well and had no complaints. She took feeds and passed flatus. She drained serous fluid. At 1530hrs, she underwent another physiotherapy session. Three further physiotherapy sessions were then planned for 16 and 17 July.

  106. [87] On 16 July, Mdm Lee had pain scores of 6/10 at 0930hrs and 3/10 at 1430hrs, but was alert and clinically well. She underwent physiotherapy and did not complain of pain until 2035hrs. Her drain fluids were “yellowish”.

  107. [88] The Plaintiff submitted that, taken together, the clinical indicators that Mdm Lee exhibited pointed to a Dehiscence. As at 14 July, these included her elevated POD3 amylase level, her low albumin level, her elevated CRP and WBC values, her “chylous” fluid, and her abdominal disquiet. As at 15 July, Mdm Lee also displayed an inability to tolerate the soft foods that were commenced on 13 July, an elevated heart rate, and drain fluid of a high volume and light haemoserous or yellow colour. As at 16 July, Mdm Lee suffered from the aforementioned unexplained abdominal pain. The Plaintiff focused on the abdominal signs and “chylous” fluid on 14 July, the “yellowish” fluid on 15 July, and the unexplained abdominal pain on 16 July.

  108. [89] As mentioned (at [60] above), it was not unreasonable for the Defendant to have considered only the indicators that had persisted at a given point of time. Those that had been resolved or otherwise attributed to conditions other than a Dehiscence could be “left out [of] the equation.” Accordingly, I will once again examine the indicators in question individually, and then consider whether Mdm Lee’s management should have been changed in the light of all the indicators, viewed globally.

  109. Abdominal disquiet on 14 July

  110. [90] The primary indicator relied upon by the Plaintiff was Mdm Lee’s abdominal disquiet on 14 July. I set out chronologically the observations on her abdomen, as recorded in her clinical notes.

  111. [91] The Plaintiff submitted that the progression from severe pain at 1000hrs to generalised tenderness at 1230hrs should have alerted the Defendant to the possibility of a Dehiscence and the need for an abdominal CT scan. Although the pain apparently dissipated thereafter, the Plaintiff submitted that Dr Yeo’s observation of “pain much better than earlier today” at 1620hrs meant that the pain had resurfaced and that the underlying abdominal problem remained unresolved. I do not accept this contention. Those words did not mean that the pain had resurfaced, but that Mdm Lee had been getting better, not worse. The Defendant’s review of Mdm Lee at 1920hrs that evening, in which he observed that Mdm Lee was well with no complaints of pain, bear this out.

  112. [92] The Defendant accepted that unexplained abdominal pain accompanied by tenderness and guarding could point to a Dehiscence, but denied that Mdm Lee had such unexplained pain, generalised tenderness, or guarding.

  113. (a) Mdm Lee’s abdominal pain at around 1000hrs was attributable to tugging on the right abdominal drain stitched to her skin. She had localised the pain to her right drain site. At 1100hrs, the Defendant examined her and found her abdomen soft and non-tender except for slight tenderness around the right drain site. Finally, X-rays of her chest and abdomen showed an absence of free air, which pointed away from bowel-abdominal communication.

  114. (b) The record of “generalised tenderness” at 1230hrs was inconsistent with the contemporaneous observations that Mdm Lee’s abdomen was “soft” with “0 guarding, 0 rebound, 0 peritonism”. Tenderness and guarding go together, except at the very onset of an abdominal problem. When tenderness has spread through the entire abdomen to give “generalised tenderness”, guarding must have followed. The discrepancy in the clinical records made it reasonable to have held off action. By the time the Defendant reviewed Mdm Lee at 1920hrs on 14 July, her pain issues had resolved, and her abdomen was soft and non-tender. Her abdomen remained soft and non-tender for the remainder of 14 July, and through 15 and 16 July.

  115. [93] Dr Mak testified that a Dehiscence would have irritated or inflamed the abdomen. However, Mdm Lee’s abdomen was soft and non-tender with no peritonism (ie, inflammation ). At 1030hrs, the clinical entry of “mild guarding” was contradicted by the observation that the abdomen was “soft”. The further observation of “0 rebound” reinforced the conclusion that there was no peritoneal irritation or other serious abdominal problem. At 1230hrs, the record of “no peritonism” in particular showed that the abdomen was well.

  116. [95] Interestingly, Dr Beckingham gave no evidence on the significance of the abdominal signs displayed by Mdm Lee on 14 July.

  117. [96] The observation of “generalised tenderness” was made by a junior doctor. As the Defendant testified, it could have been a product of unwieldy assessment technique. Further, it was undisputed that this observation was generally inconsistent with the contemporaneous observation, made by the same doctor, that Mdm Lee’s abdomen was “soft” with “0 guarding, 0 rebound, 0 peritonism”. More importantly, the Defendant had personally reviewed Mdm Lee at 1100hrs and 1920hrs, shortly after each disputed observation was recorded, and found her abdomen soft and non-tender. The review at 1920hrs, in particular, was carried out after all the effects of the Pethidine had worn off. The Defendant thus attributed the initial pain to tugging on the abdominal drains, and concluded that Mdm Lee’s abdominal signs showed no real cause for concern throughout 14 July.

  118. Chylous fluid on 14 July

  119. [97] Abdominal fluid turns white and milky – ie, chylous – when the fat globules in the lymph vessels leak out into the abdomen.

  120. [98] The Defendant testified that the chylous fluid was attributable to the chyle from the fats in Mdm Lee’s soft diet on 13 July. If a Dehiscence had occurred at that time, Mdm Lee would have been in significant pain. Instead, she exhibited no such indicators at all.

  121. [99] However, the Plaintiff’s experts opined that the amount of fat in the food that Mdm Lee had consumed on 13 and 14 July was insufficient to have produced a chyle leak. Dr Beckingham testified that chyle leaks are unheard of in pancreatic surgery. Hence, the term “chylous” in the clinical notes described only the appearance of the fluid. There was no other evidence of a chyle leak. Dr Nikfarjam added that the definitive test for a chyle leak is a triglycerides test for fat globules, which was not done. Instead, the Defendant simply diagnosed and treated the chylous fluid based on what he believed to be the characteristic appearance of chyle in the abdominal drains. This was misguided. The chylous fluid could in fact have been pancreatic fluid, which could present a similar appearance to the white colouration of chyle. The Defendant also conceded that the white colouration could have concealed pure pancreatic fluid, which was colourless.

  122. [100] Two senior HPB surgeons separately recorded the observations of chylous fluid: Dr Wong Jen San at 0810hrs and the Defendant himself at 1100hrs. The former added that a dissection of a lymph node could have caused the increased chylous output. Dr Mak and Dr Ng Shin Yi testified that chyle has a distinctive appearance that experienced specialists can immediately identify. As observed by the Plaintiff, 14 July was the only time when Mdm Lee’s drain fluid had a chylous appearance. Both parties agreed in the Glossary of Medical Terms that the Plaintiff tendered to the court that “chylous fluid” meant “[f]atty fluids with a white, creamy or milky colour from the lymphatic system [emphasis added]”. The Plaintiff could have put forward his own interpretations of term “chylous”, as he did for other entries in the Glossary. The fact that he did not do so reinforced the conclusion that Mdm Lee had only chyle in her drains.

  123. [101] It was thus not unreasonable to have attributed Mdm Lee’s chylous fluid to a chyle leak, particularly when her drain fluid became serous again shortly after her diet was changed to manage the chyle leakage.

  124. Yellowish fluid on 15 July

  125. [102] The Plaintiff submitted that the “light haemoserous” and “yellowish” colours of Mdm Lee’s drain fluid on 15 July “should have rang alarm bells.” He referred to the Defendant’s evidence that “[a]ny time we see something that has been described differently, we will check”, and to Dr Mak’s evidence that bile is yellowish and that a “mixture of bile and serous fluid can also turn yellowish.”

  126. [103] The Defendant testified that the change in the colour of drain fluid from “chylous” to “light haemoserous” at 0600hrs on 15 July was not in itself of concern. Only if the haemoserous colour had persisted would he have been worried. By 0950hrs, however, the drain fluid had become “serous”. Similarly, concerns about the “yellowish” fluid observed at 1520hrs had been assuaged following Dr Yeo’s review of Mdm Lee that evening. The fluid had become “serous” and was of no concern. Dr Yeo did not see fit to record it.

  127. [104] Dr Mak added that bile is “golden yellow” or “intensely yellow”, rather than the “straw coloured faint yellow tinge” of normal serous fluid. The absence of observations about the quality of the fluid (in particular, that the fluid was bilious), as opposed to simply the colour of the fluid, suggested that Mdm Lee did not have a Dehiscence.

  128. [105] Importantly, neither of the Plaintiff’s experts took issue with the “yellowish” and “haemoserous” observations, whether in their expert reports or their oral testimony.

  129. Continued abdominal pain on 16 July

  130. [106] The Plaintiff submitted that the Defendant would have diagnosed a Dehiscence based on the significant abdominal pain in the morning of 16 July if he had been suspecting a Dehiscence. It was “blinkered” to have attributed the pain to “faecal loading”. The Plaintiff added in his reply submissions that the Defendant should have been alarmed because the residual pain (pain score 3/10) after Mdm Lee passed motion was “unexplained”.

  131. [107] The Defendant testified that Mdm Lee’s pain score fell from 6/10 to 3/10 after she passed motion. The abdominal X-ray on 14 July had showed faecal loading. He thus concluded that faecal loading had caused the abdominal pain at 0625hrs. Dr Mak agreed that the passing of motion substantially resolved what appeared to have been severe abdominal pain. This suggested that the pain had not been caused by a Dehiscence or by significant inflammation or infection within the abdominal cavity.

  132. [108] However, Dr Nikfarjam opined that faecal loading could not explain Mdm Lee’s abdominal concerns. Further, it was worrying that she kept developing abdominal pain even while on antibiotics. Even so, Dr Beckingham accepted that faecal loading could have caused the pain, which was resolved upon the passing of motion. Mdm Lee had not passed motion from 14 to 16 July, and faecal loading was seen on the abdominal X-ray that was performed on 14 July.

  133. Management

  134. [109] The Defendant testified that Mdm Lee improved on 12, 13, and 14 July. This was incompatible with her having suffered a Dehiscence. X-rays of her abdomen and chest showed that the operatively-inserted abdominal drains were in place and that there was no free air in the abdomen, which would have indicated a Dehiscence. The fall in her WBC count from 26.26 x 109/L on 11 July to a near-normal 11.53 x 109/L on 14 July suggested an absence of on-going sepsis and infection in the abdomen. The maintenance of that WBC level from 14 to 16 July suggested further that Mdm Lee was quite well.

  135. [110] Dr Mak found that Mdm Lee’s stable global condition on 14 July was incompatible with that of a patient with a Dehiscence, who would have deteriorated. Less risk-tolerant doctors might have considered an abdominal CT scan, but it was acceptable to have continued to observe Mdm Lee. An abdominal CT scan was necessary only if she had had a fever, worsening abdominal pain associated with rebound tenderness and guarding, or drain fluid that was bilious or purulent.

  136. [112] Having testified that Mdm Lee should have had an abdominal CT scan on 11 July, Dr Beckingham did not give evidence that an abdominal CT scan should have been performed on 14 July.

  137. [113] The abdominal pain that Mdm Lee suffered initially in the morning of 14 July did not persist. The effect of the Pethidine that she had received at around 1000hrs that morning would have worn out after six hours. Yet she did not complain of further pain thereafter. The observation of generalised tenderness was inconsistent with the observation of “0 guarding, 0 rebound, 0 peritonism” (see [96] above). In any event, the Defendant in his subsequent review that evening did not observe any tenderness of the abdomen. Indeed, there was no subsequent observation of generalised tenderness on 14, 15, or 16 July.

  138. Diagnosis of anastomotic dehiscence: 16 to 17 July

  139. [115] At 2300hrs on 16 July, Mdm Lee complained of a sudden intense pain over her abdominal region (pain score 8/10). She was “screaming” in pain. Her right and left abdominal drains drained 2,775ml and 330ml of fluid respectively. She received 50mg of Pethidine, but continued to describe pain “everywhere in the abdomen” (pain score 8/10).

  140. [117] At 0900hrs, Dr Yeo observed that Mdm Lee had deteriorated and that her abdominal tenderness had persisted. An ABG test showed severe metabolic acidosis of pH7.187 requiring correction by bicarbonate infusion. As the Defendant was not at work on 17 July (Sunday), Dr Yeo informed him of these findings over the telephone. Given that Mdm Lee was in severe abdominal pain but had no severe kidney problems or breathing difficulties, the Defendant believed that the acidosis was likely due to an intra-abdominal cause. He thus “instructed Dr Yeo to arrange for an urgent [abdominal] CT scan to investigate if there was any anastomotic leak which may be the cause of a suspected intra-abdominal sepsis.”

  141. [118] At 1230hrs, a CT scan of Mdm Lee’s chest, abdomen, and pelvis (the “17 July Abdominal CT Scan”) was performed. It confirmed a Dehiscence and revealed collections of fluid. The scan report stated:

  142. [119] At 1300hrs, Mdm Lee was moved to the High Dependency Unit (“HDU”) for monitoring and for stabilisation via correction of her acidosis.

  143. [120] By 1320hrs, the Defendant had arrived at SGH, reviewed Mdm Lee, and diagnosed her with a “delayed pancreatic dehiscence”. After a discussion with Dr Ng Shin Yi, the consultant intensivist at the SICU on 17 July, the Defendant moved her to the SICU for resuscitation at 1530hrs.

  144. [121] There is no dispute that Mdm Lee had a Dehiscence on and after 2300hrs on 16 July, if not earlier. The Plaintiff did not challenge the Defendant’s actions on 17 July, save that the diagnosis should have been of a subsisting and not a new Dehiscence.

  145. [122] Every case must be decided on the issues raised by the pleadings, which bind the parties. If other issues are desired to be raised or come to light during the trial, they must be pleaded by way of amendment (Deans Property Pte Ltd v Land Estates Apartments Pte Ltd and another [1994] 3 SLR(R) 804 at [15]). Although evidence at trial can overcome omissions in a party’s pleadings where the opponent is not taken by surprise (Holland Leedon Pte Ltd (in liquidation) v C & P Transport Pte Ltd [2013] SGHC 281 at [150]), it may not establish facts radically different from those pleaded that more than simply vary, modify, or develop what has been alleged (John G Stein & Co Ltd v O’Hanlon [1965] AC 890 at 909).

  146. Management of anastomotic dehiscence: 17 to 28 July

  147. [125] I set out the relevant portions of the Defendant’s AEIC on his plan for Mdm Lee from 17 July:

  148. [127] It is clear to me from the Defendant’s AEIC and oral evidence at trial that, from 17 July, he had wanted to perform a laparotomy but could not do so because Mdm Lee was haemodynamically unstable. The intensivists had to resuscitate and stabilise her first. This was also the evidence of Dr Ng Shin Yi and Dr Ang, who were the consultant intensivists at the SICU on 17 July and from 18 to 25 July respectively. However, it was less clear whether, after Mdm Lee’s condition had improved by 19 July, the Defendant had been continuing to wait for an opportune time to perform the laparotomy, or had changed his mind and had been deferring the laparotomy to see whether her condition would improve to such an extent as to render a laparotomy unnecessary. As it turned out, this lack of clarity was not material, as I will explain later.

  149. [128] As for the Plaintiff’s case, I reiterate that it was Dr Beckingham’s evidence that 17 or 18 July was the latest time for intervention via a percutaneous drainage procedure (after an abdominal CT scan) or a laparotomy (see [52] above). Thereafter, any such intervention would not have “made a difference to the outcome in [Mdm Lee’s] case.” Dr Beckingham maintained this position throughout the trial.

  150. [129] As for Dr Nikfarjam, he opined that if percutaneous drainage was not performed on 17 July, it should have been performed “on the 18th, 19th, 20th, 21st, 22nd, 23rd, till the day [Mdm Lee] had surgery.”

  151. Ms Tan pressed Dr Nikfarjam to give a more specific time-frame for the performance of percutaneous drainage. However, he maintained that the time-frame was simply after Mdm Lee “had gone into the intensive care, was intubated, was given some inotropic supports and was having blood pressure maintained” in the afternoon of 17 July. This was vague.

  152. 17 to 19 July

  153. [134] By 2100hrs on 17 July, Mdm Lee had become more stable. However, she remained on two inotropes. Her abdomen was slightly more distended than earlier in the day but was still soft. Her NGT drained bilious fluid, and both her abdominal drains drained turbid fluid. Her urine output was fair. She had an elevated intra-abdominal pressure due to an accumulation of fluid in her abdomen. Across the day, she experienced severe coagulopathy and acidosis, and had low oxygen saturation that required intubation for ventilation.

  154. [135] On 18 July, Dr Ang assumed the role of consultant-in-charge of the SICU from Dr Ng Shin Yi, and with it, the care of Mdm Lee. At 0830hrs, Mdm Lee remained “very unwell”. In her condition on 18 July, surgery carried a 70–80% risk of death. Accordingly, the objective was to continue resuscitation to improve her condition. By 2020hrs however, Mdm Lee had gradually stabilised. She displayed good drainage from her right drain of 800ml. Her intra-abdominal pressure had fallen further, and her abdomen had become less tense and distended. However, various issues persisted, including low blood pressure, severe coagulopathy, poor oxygen saturation, acidosis, acute kidney injury, raised intra-abdominal pressure, and septic shock.

  155. [136] By 0930hrs on 19 July, Mdm Lee had improved even further. Her oxygen saturation had increased, and a single inotrope sufficed to support her blood pressure. Her lungs were clear and she was able to obey commands. However, her coagulopathy, acute kidney injury, and raised intra-abdominal pressure remained unresolved. The Defendant consulted Dr Ang and concluded that Mdm Lee was in no condition for surgical intervention.

  156. [137] The Defendant testified that he had monitored Mdm Lee’s drainage and abdominal signs at all times, and would have intervened via percutaneous drainage or even surgery had she failed to progress as expected. However, Mdm Lee was from 17 to 19 July too ill for any intervention.

  157. [138] Dr Mak agreed that from 17 to 19 July, Mdm Lee was “not stable or in any fit condition to be transferred out of the [S]ICU for any length of time”. Any consideration of “the possibility of these various interventions, whether surgery or drainage … would have been academic because she was not fit at all for any of these procedures to be done.”

  158. [139] Dr Beckingham observed that up until 18 July, Mdm Lee was “as sick as a dog”. While a laparotomy was theoretically possible, his first choice would have been a percutaneous drainage procedure. Nevertheless, Mdm Lee needed to be stabilised before any such percutaneous drainage could be performed. Given her condition, he would have been cautious, and would have “th[ought] carefully about whether to do a radiological guided drainage” at all. Since Dr Beckingham had observed that any intervention after 18 July would have been “too late” (see [52] above), he did not comment on her management thereafter.

The judgment continues past this page. Read the rest at eLitigation.