NUH Liable for Causing Kidney Donor’s Death
SIM JUNHUI (Juris Sub-Editor)
A loving wife donates a kidney to her sick husband. What could have ended as a tribute to true love and the gift of organ donation ended badly when, in a cruel twist of fate, he survived but she died. This was the first donor death following kidney surgery in Singapore.
It precipitated the case of Surender Singh v. Li Man Kay  SGHC 168 [Surender Singh], an interesting case because of the court’s application, or lack thereof, of Dr Khoo James v. Gunapathy d/o Muniandy,  2 Sing. L.R. 414 [Gunapathy].
Summary of facts
Diagnosed with kidney problems in 1996, the first Plaintiff continued to deteriorate. Eventually in 2004, the Deceased, his wife, offered to donate a kidney to him. To this end, she underwent a Left Hand Assisted Laparoscopic Donor Nephrectomy (“HALDN”) at National University Hospital to remove her left kidney for transplantation into the first Plaintiff.
The two Defendant-doctors—Li and Consigliere—performed the operation. They used several “hem-o-lock” clips (“clips”) to prevent the various kidney-related blood vessels they dissected from bleeding. After an uneventful operation, the Deceased was sent first to the recovery room and then to the general ward for monitoring.
In the general ward, she was to be monitored hourly. Her parameters were first recorded at 1430 hours. However, the next entry was recorded at 1600 hours. Visitors arriving between 1600 and 1615 hours found the Deceased to be unresponsive, cold to the touch and very pale. Her sister-in-law alerted nurses who then informed doctors. The Code Blue Team was activated, but failed to save the Deceased who passed away thereafter.
The autopsy revealed nine clips in total in the Deceased. Specifically, four clips were attached to soft tissue at the periphery of her left renal artery, dissected during the operation and now “wide open”. The Plaintiffs’ medical expert (“Nicholson”) opined that the doctors had applied four clips to the artery, increasing blood pressure which displaced the clips and caused fatal blood-loss. The doctors insisted they had only attached two clips, an accepted medical practice which would not have increased blood pressure unusually.
The court’s decision
The court considered, inter alia, whether the doctors applied two or four clips, and consequently, whether they had fallen below the reasonable standard of care in conducting the HALDN. In addition, the court considered whether NUH had acted negligently by not complying with standard medical practice regarding post-operative monitoring.
The court held that the applicable Singapore law on medical negligence is Gunapathy, which approved the Bolam test. Courts must not find doctors negligent (provided a respectable body of medical opinion, logically held, supports their actions. To qualify as ‘responsible’, the body of medical opinion must pass the threshold test of logic, i.e., the expert must have considered “the comparative risks and benefits relating to the matter”, and arrived at a “defensible conclusion”. To do so, medical opinion must be “internally consistent on its face”, and “not fly in the face of proven extrinsic facts”.
Ultimately, the court held that the doctors followed accepted medical practice in using two clips and were not negligent in conducting the HALDN. However, the court found NUH was liable for causing the Deceased’s death by not complying with standard medical practice regarding post-operative monitoring, which was hourly.
Two parts of the judgment are interesting to consider and will be dealt with in turn.
On whether the doctors attached two or four clips to the artery
Holding that the doctors attached two clips, the court cited, inter alia, the high unlikelihood of attaching four clips in the 1.5-minute warm ischemic time recorded in the Operation Report. “Warm ischemic time” begins when the artery is clamped to interrupt blood circulation and prepare for dissection, and ends with the commencement of cold perfusion, when the extracted kidney is flushed with an iced solution to prepare it for transplantation.
Li testified that he required 45 seconds to apply two clips to the artery, and another 45 seconds to do likewise for the vein, amounting to 1.5 minutes for the warm ischemic time. This accorded with the 1.5 minutes recorded in the Operation Report. Moreover, the Defendants’ medical experts—Howard and Cheng—opined that applying four clips in 1.5 minutes, as the Plaintiffs alleged, was “very improbable”.
However, Nicholson testified that he would take 3 to 4 minutes to apply four clips to the artery and vein. When referred to the 1.5-minute warm ischemic time in the Operation Report, he contended that his own warm ischemic time lasted between 1 and 7 minutes, and that he could achieve a 1-minute warm ischemic time applying five clips. He also observed that, to him, “warm ischemic time” ended with the cold perfusion’s conclusion, not commencement, conceding nonetheless that medical literature did not support his definition.
The court disregarded Nicholson’s evidence because it “did not comport with the requirements of expert evidence laid down in …Gunapathy’s case.” The court probably found that his opinion “[flew] in the face of proven extrinsic facts”—i.e., medical literature defining “warm ischemic time”.
It is submitted that this is questionable. Gunapathy absolves doctors of liability for their acts where a respectable body of medical opinion approves of them. Implicitly, the acts in question would already have been found. The question is whether the doctor’s acts, in circumstances arising from known facts, are sound and supportable. Here, the question remained one of fact.
In other words, the question in Gunapathy is: “Given these known facts, should the doctor have done this, or that? Can liability ensue because he did this, instead of that?” Here, the question remained: “What did he do in the first place?” Gunapathy should not be applied to determine facts that are yet to be determined.
In any case, Nicholson’s evidence was not substantively problematic. To disregard it just because he defined terms differently is perplexing.
Notwithstanding the court’s reasoning here, however, other far more convincing reasons the court gave supported its finding that the doctors used two clips on the artery. It is submitted that the finding of fact is correct.
On whether NUH complied with standard medical practice regarding post-operative monitoring
The court found that standard medical practice, even in NUH, was “hourly monitoring”, and NUH admitted a one-and-a-half hour lapse from 1430 to 1600 hours.
However, NUH alleged that strictly construing “hourly monitoring”, i.e., that the Deceased should have been observed at 1530 hours, ran contrary to “established medical practice”. Their experts testified that “hourly monitoring” was only a guide and did not require reviews exactly every hour. Howard noted that a 30-minute buffer was “entirely acceptable” given the Deceased’s post-operative condition. Cheng remarked that nurses could take 30 minutes to check the parameters of six patients in a ward-room.
Despite agreeing that hourly monitoring on-the-dot was practically impossible, the court held NUH liable in negligence. The court opined that both Cheng and Howard premised their testimonies on the ward actually having six patients on the same monitoring schedules. Without evidence to prove this, the court dismissed their opinions.
It is submitted that Gunapathy should perhaps have been applied here. NUH’s assertion was presumably that interpreting “hourly monitoring” loosely as a guide (which could include an-hour-and-a-half monitoring) was established medical practice. Such was especially the case in special circumstances like those Cheng and Howard cited. No such special circumstances existed here.
However, the court failed to discuss whether a respectable body of medical opinion supported NUH’s assertion as to established medical practice. Either the court decided, on its own accord without reference to medical opinion, that Cheng and Howard were wrong, or it implicitly based its judgment on Gunapathy. Either way is problematic. Judges stepping into the shoes of doctors is precisely what Gunapathy sought to prevent. On the other hand, much may be said about expressly justifying a decision.
Implications of this case
Medical negligence law has come a long way from its inception in England and reception locally in Gunapathy. The Bolam test has its disadvantages. An example of which may be the excessive reverence for the medical profession: “We often enough tell doctors not to play god; it seems only fair that, similarly, judges and lawyers should not play at being doctors.” If taken to the extreme, this statement may sanction judicial indifference to medical behaviour, whether reasonable or otherwise. As evidence of the shortcomings of the Bolam test, Australia, Canada and Malaysia have all rejected it.
Nonetheless, before the Court of Appeal revises its decision on the matter, if it ever does, Gunapathy remains the applicable law locally. Indeed, there may actually be little in principle that is offensive about the Bolam test. When judged in his professional capacity, a skilled professional should be held to the standard of his profession and peers. The phrase quoted from Gunapathy above has often been cited disapprovingly. However, it is submitted that the learned former Chief Justice surely must only have intended it as poetic flourish, and not for it to be construed as strict law.
Whatever the case, Gunapathy must be applied accurately and appropriately. Here, its application in deciding the number of clips used was unnecessary. It was, however, necessary to apply it to consider NUH’s assertions on established medical practice. The failure to do so may have ultimately led to a finding of negligence liability on the part of NUH.