Velarde v Guy's & St Thomas NHS Foundation Trust

Alexander Velarde developed a brain injury whilst being treated by the Defendant (“Guys”) in September 2005. He claims that this was the consequence of negligent treatment by Guys.

Alexander was born on 9^th August 2005, at 34 weeks and 5 days (some 5 weeks premature).

On the afternoon of September 17^th 2005 he was admitted to Guys following referral from St Thomas’ suffering from multiple ventral septal defects – holes in the septum between the left and right ventricles of the heart. A standard treatment where, as here, the defects were unlikely to close spontaneously, is to postpone a surgical closure of the holes for some 18 – 24 months until the child has grown and the heart is bigger.

The right ventricle pumps venous blood through the lungs, from where it returns to the left side of the heart. The left ventricle in turn pumps the oxygenated blood through the arterial system, from whence it returns by the veins to the right side of the heart, and to the right ventricle. Where there are holes between left and right ventricles, there is a “left to right shunt” and consequently a reduced flow through the systemic arterial circulation, and increased flow through the lungs. The result is that the flow to the lungs results in excess fluid accumulating there. If this is left unremedied for too long, irreversible changes are likely to set in. To hold the fort in the case of an infant only a few weeks old, pending an operation in some 18 to 24 months’ time, it is common for an operation of pulmonary arterial banding to take place: a band is placed around the pulmonary artery which has the effect of restricting flow to the lungs, reducing the significant left to right flow whilst not abolishing it, thereby improving symptoms and giving the child the opportunity to feed and grow before the necessary repair can more practically be undertaken.

When Alexander was admitted he weighed 3.05kg (18^th September 2005). His lungs contained excess fluid. His liver had also increased in size as a result of fluid leaking into the extra vascular spaces.

The object of treatment prior to the banding procedure (which was to be conducted in Alexander’s case) was to remove this excess fluid: accordingly Alexander was given diuretic drugs Frusamide and Spironolactone; inotropic drugs such as Milirone to support his heart function; and, from 20^th September, Captopril. Captopril is neither a diuretic nor an inotrope. Rather, it works to dilate the blood vessels – more particularly the veins. This has the effect of lowering the resistance of the systemic circulation. This makes it easier for the heart to pump through the systemic circulation, and thus helps to reduce the flow from the left ventricle to the right ventricle when the left contracts.

On 21^st September the intended operation of pulmonary arterial banding occurred. It was a “closed heart procedure” (i.e. not one performed on by-pass). Alexander returned to the Paediatric Intensive Care Unit (PICU) after surgery. He complains about his treatment between his surgery and 28^th September, on which date it became apparent that he had suffered brain injury: peri-ventricular leukomalacia (“PVL”). This PVL was the consequence of a cerebral venous sinus thrombosis (“CVST”): a blockage of the venous drainage from the brain caused by a clot which had propagated. A result of the increasing blockage, and consequent inability to drain venous blood, was a degree of brain death local to the clot: an infarct. This probably occurred in the 12 hours prior to Alexander being seen to fit. The first signs of fit occurred in the night of 27^th/28^th September, so the infarct most probably developed at some time during the 27^th. It later became apparent that the fit was linked to what was recognised eventually as PVL. There is less precision about the time when the clot would have formed, but the evidence was preponderantly to the effect that this would have been some 24 – 48 hours before the infarct – that is, possibly as early as the 25^th. and probably no later than the evening of the 26^th. September.

The Claimant’s case is that the clot was caused at least in part, or contributed to, by Guys’ failure to manage his care in the PICU properly in two separate but linked respects. It is said first that he was given too little fluid for too long such that he suffered a degree of dehydration as a consequence of which clotting of the blood became more likely and occurred. Second, that he was given too much Captopril, too quickly, with the consequence that blood pressure fell to an unacceptably low level, causing sluggish blood flow through the brain both making it easier for a clot to develop or propagate once initiated, and dropping to such a level that the usual systems within the brain itself which compensate for fluctuations in blood pressure (the “auto regulatory system”) did not function. The failure of auto-regulation permitted intra-cranial pressure to rise as arterial blood pressure dropped, with the consequence that blood flowed yet more slowly through the brain. Slow flow prevented the clearance of the clot, or the re-canalisation of the affected parts of the brain, as might have occurred had flow been normal.

What is, on the one hand, “too little” and, on the other, “too much”, and what is on the one hand “over too long a period” and on the other “over too short a period” is necessarily a matter of degree. What is in issue, therefore, is the judgment of the unit and the clinicians concerned. The Claimant has therefore to show that the exercise of judgment was not one which a reasonably careful clinician would make – or, in other words, if a responsible body of clinical opinion might accept what the Defendant’s clinical staff did as proper, proof that most would not do so would not be sufficient to establish a lack of due care (Bolam v Friern Hospital Management Committee [1957] 1 WLR 582) unless it could be shown that to adopt such an approach was incapable of withstanding logical analysis, even though the approach was adopted by some (Bolitho v City and Hackney Health Authority [1998] A.C. 232).

Mr. Maskrey QC’s argument on behalf of the Claimant was that no responsible clinician would do as was done in Guys: it was to take “too aggressive” an approach both to reducing fluid, and to the rate at which Captopril doses were increased; but in any event, he submitted that there was no purpose to be achieved by continuing to keep fluid intake low, nor by raising Captropril levels so quickly, since each came with risks. The first risked dehydration which would make it more likely that clots would form. The second led to reductions in blood pressure, which if not monitored to see how each dose was tolerated before administering the next might cause dangerous hypotension. In Alexander’s case these risks eventuated: the first led to a clot; the second to a reduction in blood pressure below a level at which auto-regulation could compensate and hence to a sluggish blood flow which facilitated the propagation of the clot and development of the infarction. Either (in his argument, both) thus contributed to Alexander’s injury.

The records which chart the fluid balance from day to day, and the blood pressures associated with doses of Captopril, have proved unusually difficult to interpret. A number of recordings of readings relating to particular time periods were begun on one page, but completed on a different page, often without the entries on one page flowing seamlessly from hour to hour and day to day such that the one page naturally followed the other. The records of some readings of mean arterial blood pressure have been squashed into a row not initially designed to accommodate them. This misled both leading counsel until immediately before the start of the trial, despite their experience of dealing with clinical cases, and has led to misinterpretation of the available data by expert witnesses. The difficulty has been aggravated by the use of different records to cover the time when Alexander was in the PICU and that after he was transferred – on what turned out to be the 27^tth. September, though one expert was misled by this too – to the Rothschild ward (in the light of what had up until then appeared to be good progress). An unfortunate consequence of this has been the expression of developed views by the experts on both sides of the argument in the light of that which they understood – sometimes wrongly – to be the data. Though expert witnesses must be expected to moderate their views in the light of different information, and I do not doubt those appearing before me all recognised the need in general terms to do so, I have also to recognise that even for those who have the objectivity of most experts there may be a reluctance to change or moderate a once firmly expressed view. A court has to be alert to this.

It is also the case that some experts – notably Dr. Elliot Shineborne, called by the defendant – did not appreciate the timescale over which the physical causes of Alexander’s injury developed. This was properly to be assessed by reference to the evidence of neuro-radiologists instructed by the respective parties who did not in the event need to be called, such was the measure of their agreement. The injury they saw was, however, a rare one: where a thrombosis occurs in the brain, it more usually does so in the arterial circulation. This was venous. The rarity of the particular insult explains why there was some uncertainty about the precise timescale within which the events which precipitated injury occurred.

It was only just on the eve of trial that Mr. Spencer QC for the Defendants pieced together a reliable account of the fluids, blood pressures, drugs and biochemical markers which are key to understanding the case. It is necessary to set them out in some detail in respect of the two principal complaints – first, fluid management and, second, Captopril dosage.

Restriction of fluids (the aim of fluid management) and reduction of systemic resistance to flow by dilation of the blood vessels (the aim of administering Captopril) both had the purpose of reducing load on the heart. Each also had consequences. Reducing fluid intake would tend to lead to an increase in the concentration of electrolytes, especially sodium, in the blood stream (at least after any excess of fluid outside that contained intravascularly had osmosed from the extravascular spaces back into the bloodstream, which would have the effect of maintaining intravascular volume while this was happening). Any dose of Captopril would have an almost immediate effect of reducing blood pressure in the short term, and have a longer term effect of doing so, though to a lesser extent, over a protracted period. Reduction in pressure had as a consequence an increase in the risk that clots might form.

Alexander was admitted to hospital on 17^th September 2005. That evening, a consultant paediatric cardiologist diagnosed that he was suffering from multiple VSDs (Ventricular Septal Defects). Steps were taken to reduce the amount of fluid in his body both by the administration of diuretics from the start, before the administration of Captopril was commenced, from 19^th. September. This was first at a rate of 0.1mg per kilogram of body weight. There being no significant change in blood pressure noted after an hour, the dose was increased to 0.2 mg per kilogram by 2pm the following morning, and that evening increased to 0.3 mg per kilogram.

Alexander underwent operation on 21^st September, and was re-admitted to PICU shortly after midday. The Cardiologist, Dr Simpson noted that the result appeared to have been good and said “re-start Captopril when orally fed”.

No complaint is made about the administration of Captopril at this stage in Alexander’s treatment nor is it suggested that the rate of increase was too great.

On 24^th September 2005, Captopril was again introduced – at the lowest rate, 0.1mg/kg. The aim was to increase it over time to doses at a rate of 1 mg/kg of body weight. There is no criticism of that rate in itself: Doctor Tometzki, a consultant paediatric cardiologist called for by the Claimant, described it as the “expected full dosage for [Alexander’s age]” (paragraph 7.19 of his report). His criticism was that the dosage should not have been increased with such rapidity after reintroduction. With such rapidity greater caution should, in his view, have been exercised.

It was well recognised generally, and certainly at the time by the Defendants, that the administration of Captopril might lead to a short term loss of blood pressure. Incurring such a risk was inevitable if it were to be administered at all.

The doses of Captopril between its reintroduction on 24^th September and 29^th September were as shown on the chart below:

As I understood the expert evidence, the purpose in increasing the dose in stages was to ensure that each successive dose was tolerated before increasing it further. Since the principal risk was that of a dangerous reduction in blood pressure, the question was whether short term or longer term reduction in blood pressure following one dose contra-indicated the next. This is essentially a question of clinical judgment. There is no guidance set out in any text book or formulary as to the appropriate rate of increase. In Dr. Tometzki’s unit the increase would, he said, be more cautious than it was here. I accept, however, Mr Spencer’s submission that it could not be said to be a breach of duty in itself to move from a dose of 0.1mg/kg to 1.00mg/kg over a space of a little over 2 days so long as appropriate care was taken to assess the clinical effects of the increasing dose. Both cardiologists indicated as much in their answer to question 21 in their report of their meeting prior to trial. Dr Shineborne said “incremental increases were performed whilst undertaking appropriate BP checks to ensure the fall in BP was not prolonged…”. Dr. Tometzki said that “given the significant falls in BP as demonstrated on 27^th. September 2005” he maintained that it was simply too aggressive and would not be normal practice in most cardiac intensive care units (my emphasis). Both these views show that clinicians must be sensitive to the effects on the child before effecting any increase.

In essence, the issue for the court’s enquiry was whether the clinical condition of Alexander at the time, and in particular the demonstrated effect of each dose when administered, both considered over the short term and over the longer term, in respect of blood pressure, was such that to administer the next increased dose (or for that matter any dose at all) would fall below the standard to be expected of the reasonably careful and competent medical practitioner.

In order to check on the effects in Alexander’s case, mean arterial blood pressure readings were recorded both prior to the administration of Captopril, and at three or four 15 minute intervals thereafter. Care was taken to check. However, that is only half the story: there must also be care in the interpretation of the results shown by those checks.

Dr. Davis, an intensivist called for the Claimant, told me that a mean blood pressure below 40mm.Hg. indicated hypotension. He would accept that the normal range of mean arterial pressures for a child of Alexander’s age fell from between 40 and 45 up to 60mm.Hg.

The 15 minute measurements demonstrated the following. On 25^th September at 14.15 Captopril was administered at a rate of 0.5mg/kg. Immediately prior to administration, the mean arterial blood pressure was 45. Both of the next two readings (14.30, 14.45) showed it had dropped to 42, but it recovered to 45 by 15.00. The next dose (0.7mg/kg) was administered at 22.35. Blood pressure immediately prior to administration was 67. 15 minutes later it dropped to 65 and then to 52 and then at 23.05 was at 49. Immediately prior to the next dose the mean arterial blood pressure was 64; the 15 minute readings thereafter were 74, 73, and then 87. These are high, and in no way could indicate that Captopril had dangerously lowered blood pressure. On 26^th. the first full dose was given. This was at 14.00. Immediately prior to that mean arterial blood pressure was 51. It dropped to 36, then recovered to 40 then 44. At 20.15 Alexander received a similar dose: immediately before he did so the mean arterial blood pressure was 51. After administration it dropped to 40, to 35, then rose to 44. On 27^th September at 06.45 a further full dose was given. The blood pressure beforehand was 56. It dropped to 31 at 07.00 then recovered to 52 at 07.15. At 14.15 the blood pressure before administration of a further full dose was 54. 15 minutes later (probably) it was 51. At 22.00 on 27^th September the blood pressure prior to administration was 68. The reading immediately after was 66.

It was during the night of 27^th September that Alexander’s recorded blood pressures fluctuated, and his condition became unstable. The next dose was therefore omitted.

At 15.30 on 28^th September the mean blood pressure prior to administration of Captopril was 52. It then dropped to 47 (after 15 minutes) and 31 (after 30 minutes) before recovering 5 minutes later to 40. A last dose was given on that day at 22.00, before which the mean blood pressure was 44.

In his expert report, Dr. Tometzki said at paragraph 7.17 that as the dose of Captopril escalated there was a concomitant decline in Alexander’s blood pressure measurements. He repeated this at paragraph 7.20: “despite a gradual fall in blood pressures prior to transfer to the ward the Captopril dose was increased rapidly” and at paragraph 7.21: “I believe that the Captopril dosage should not have been increased with such rapidity given declining blood pressures.” He had therefore identified from these readings that there had been a gradual fall in blood pressure readings, and that this trend contra-indicated further doses (at least of the full amount).

The expression of this opinion was influenced by that which the evidence (as was agreed by the parties) showed, by the time he gave oral evidence, was his misreading of the documentation in a number of respects.

Towards the end of his evidence, Mr Spencer suggested that far from the trend being one of downward blood pressure, as the doctor had supposed, it was one which showed a gentle increase. Dr. Tometzki indicated that he preferred to view the data graphically. In order to allow him to do so, such that he could address this point, a graph was then prepared by the Defendant, and became page 167D in Trial Bundle File 1. A line of best fit is shown on the graph, demonstrating the trend of the data. The chart shows readings from 26^th September (14.15) to 27^th September (23.00). The trend is one of a gradual upward increase from between 45 and 50 to between 55 and 60.

In response, Dr. Tometzki drew up his own table, but added data for the 28^th and 29^th of September to that for the period from 26^th. The line of best fit then dropped from somewhere just over 50 to a point between 40 and 45, with a downward trend.

These two charts amply demonstrate that an interpretation of trend may depend on the starting point and the length of time over which readings are taken. For the purposes of assessing whether the Defendant had been in breach of its duty of care, however, in a manner that might be causative, the readings which are relevant are those which address the allegations put in the particulars of claim. The claim makes two allegations: at paragraph 31(iv) it complains that the Defendant “rapidly and unnecessarily increased the dose of Captopril between 24 and 26 September.” Secondly, it complains that the Defendant “failed to reduce the Captopril between 24 and 28 September”.

When Mr Spencer cross-examined in respect of each dose of Captopril given between the 24^th. and the 27^th. as to whether it was contra-indicated, it was demonstrated that in respect of each administration of Captopril there were no contra-indications on the information available to the clinician at the time. The period particularly relevant, causatively, is that between 26^th and 27^th. A clinician assessing whether it was appropriate to give the next dose (by then a full dose) would ask at the time whether the blood pressures were showing a trend of dropping further with each full dose received, or not. If not, there would be no contra-indication to the dose. Viewed on the 26^th. and 27^th., in the light of the information then available, as Mr Spencer’s graph covering the relevant time demonstrates, a careful clinician would have concluded that, if anything, there was a gentle upward trend in blood pressure, fully justifying the continued administration of Captopril, without which it would be likely to rise further.

In the morning of 28^th, after a difficult night for Alexander the clinicians chose not to administer Captopril: they responded to the swings in blood pressure recorded during the previous night by leaving out what would otherwise have been the dose of Captopril to be administered that morning.

Having considered the evidence as a whole I conclude that the rate at which Captopril was increased was quicker than it would have been in many units. The reference by Dr. Tometzki in his answer to question 21 at the expert meeting suggests that nonetheless there are some units, even if a minority, in which such a rate would be adopted. Unless it can be argued that there was no clinical purpose in administering Captopril at all, or at such a full dose, the administration of each successive dose of Captopril up to the full dose did not demonstrate a fall below the appropriate standard. Nor was there any increase after 26^th at 14.15. Prior to the events of the night of 27^th in particular, I am satisfied that no clinician in treating Alexander did so negligently by administering Captopril in the dose in which it was given. The allegation at paragraph 31(iv) of the claim falls away. Nor could it be said that clinicians were administering Captopril as a matter dictated by the policy of the unit, without giving any thought to the condition of their patient: the omission of the dose in the morning of the 28^th. is testament to this.

Though I am cautious in accepting the evidence of either cardiologist this is for different reasons. Dr. Tometzki overlooked the fact that there had been gradual increases in dose since Captopril was re-started on 24^th till reaching the full level on 26^th; and seemed to have thought that the doses were still being increased when Alexander was “transitioned to the ward” (though this took place well over a day after the dose had reached its maximum). In his further written remarks (supplied at my invitation) after finishing his evidence save for this matter, he gave a considered response to the graph showing the trend of blood pressures which Mr Spencer’s team presented in the course of his evidence. This was in response to a passage of questioning in which he had been taken to each individual reading before, and that immediately after, the doses of Captopril at full level since 14.15 on 26^th (see above for the details), in the course of which he began to talk about trend being that which was important. In what he said in writing he insisted that the trend, after full dose was reached, was for the blood pressure to drop. The period which he had been asked about in particular in evidence was in respect of the period before the signs of an infarct started to show – that is, before the night of 27^th. I would have expected his further observations to have considered the position of the clinician at the time, and not one acting with the benefit of readings yet to be taken. I do not think that he was prepared to accommodate what had during the trial become an agreed understanding of that which the records showed, nor that the relevant period for the purposes for determining whether clinicians should have continued (prior to 28^th) to have administered Captopril depended upon their view from their position at the time as to what the trend might be. I do not think it helped me that in response he should produce a table which, although accurate in terms of the data, did not support his thesis to the effect that a clinician, caring for Alexander prior to the night of 27^th would have realised that pressures were dropping, and therefore Captopril should be reduced.

Dr. Tometzki clung too fast to his opinions first expressed rather than to that which an impartial observer of the data would have concluded. For all these reasons, I have reservations about accepting the entirety of his evidence.

Dr. Shineborne, unfortunately, sought to argue a case. This was principally in respect of fluid management. Asked a question for the joint experts meeting, which read:

“Alexander’s birth weight was 2.06 kg; his weight on 16 September 2005 was 3.1 kg. Given Alexander’s cardiac condition, is it probable that some of this weight was due to excessive fluid retention? If so, is it possible to state, on the balance of probabilities what percentage if this is within your area of expertise”

Dr. Shineborne responded, in common with Dr. Tometzki, that it was approximately 5%.

He was present during the evidence of Dr. Habibi, also called by the Defendants. The view of Dr. Habibi was that the amount of fluid retention was most likely to be nearer 10% than 5% (from the weight before admission said to be 3.1 kg, measured as 3.05kg on the ward, down to 2.8kg. is a 10% drop) since he saw no clear signs of clinical dehydration, and thought the effect of fluid restriction and administering diuretics would in the first place be to draw fluid from the extravascular space before it began to do so to any great extent from the bloodstream. He would however defer to the cardiologists. Dr. Shineborne wished to reconsider his own evidence in this light. Just before he was due to give evidence, following on from that of Dr. Tometzki, he volunteered a departure from his earlier agreement with Dr. Tometzki to express this, though it had been understood (from earlier discussions between the court and counsel during the case, which Dr. Shineborne was in court to hear) that late evidence would most probably not be received. Though it was then agreed by him, through Mr. Spencer, that he was content to say that he could not be certain as to the extent of fluid retention, and he was permitted to give this evidence, as soon as he was in the witness box he began to advance that which he had earlier said he would not. I regret that I could not exclude his behaviour as being an attempt to manipulate court procedure so that he could advance a view which favoured the party calling him: if the view had been his true expert view it is difficult to see why he would have agreed through Mr. Spencer that he was uncertain as to the extent of fluid retention.

At an earlier stage, he had sought to analyse the extent of fluid retention by hypothesising that unmeasured Cow & Gate feeds had been given, so that reliance could not be placed on the apparent difference between recorded fluid input and output: part of the fluid input, supposedly closely monitored, had not been measured. There was no evidential basis for this suggestion.

Both these events demonstrated a willingness of Dr. Shineborne to attach himself to arguments which tended to favour the Defendants rather than give me a dispassionate view as to that which the evidence showed.

In conclusion, he was an unsatisfactory witness, and Mr. Spencer in closing did not seek to argue the contrary. By contrast, Dr. Tometzki did not express views based on supposition rather than evidence, even if I was bound to treat his evidence too with the reserve I have already mentioned.

This is not to say that I reject all that Dr. Shineborne had to say. In particular, he (in common with some of the other experts) expressed the view repeatedly that the individual child required to be treated as, indeed, individual: that treatment needed to be addressed to the child’s particular circumstances. I accept this. The implication of my accepting this for the present case is twofold: first that I accept Mr Maskrey’s point that protocols which detail treatment generally to be given to children within a certain class cannot be applied slavishly to all children, irrespective of factors individual to their cases which would suggest that for them aspects of the protocol might not be appropriate; but second that I also accept that whether an approach to treatment can be described as “too aggressive” or “too quick”, by comparing the general approach in one unit with that in another, does not answer the question whether it is “too aggressive” or “too quick” in the case of any given individual child, such as Alexander. I have already answered the questions that arise in this case in respect of Captopril: I now turn to the question of the restriction of fluids.

I am satisfied on the evidence that there was good reason in general to restrict fluids for those children who suffered from VSDs whilst awaiting later surgical repair. Without doing so, the continuing function of the heart and lungs could permanently be compromised. This does not in itself answer the criticism the Claimant makes, that this restriction was too great for too long, but it is a necessary starting point: there was nothing inherently wrong in restricting fluid intake.

To maintain normal fluid balance for a child of Alexander’s age 4-5ml. per hour per kilogram would be required. From the date of the operation until extubation on the 23^rd. September fluid input was deliberately restricted to 2 ml. per kilogram per hour. Fluid was being administered both intravenously and orally until then, and orally thereafter. There is no explicit criticism of this restriction up to the point where extubation occurred and oral administration of fluids started: it is standard treatment where there is such an operation, although as Mr Spencer pointed out in his closing submissions the Claimant’s stance was somewhat equivocal as to whether there should have been a relaxation of the fluid restriction on the 22^nd. However, the rate was not relaxed to 3 ml. per kilogram per hour on 23^rd. That did not occur until 25^th. September. It continued at the same time as diuretics were being administered. Fluid restriction from at least the 23^rd. is subject to clear criticism.

The fluid balance day by day, taken from the charts (and as set out at page 344 of the trial bundle), showed an overall fluid reduction on the 22^nd. of 110 ml.; and on 23^rd. of 141ml. Thereafter the fluid balance over each following day began to run into positive territory: it was plus 6ml. on the 24^th., plus 84ml on the 25^th., and then into positive three figure readings on 26^th. and 27^th.

However, whether or not a child is dehydrated is not best seen by looking at the results of the balance of one day’s input and output in isolation. Cumulative fluid loss is more informative. In this case, that involves looking back to the pre-operative period. Over the four days prior to the day of operation, Alexander lost 106 ml. By the start of the 24^th. this had become a loss of 239ml. Loss of water does not necessarily equate to dehydration: if there is fluid overload, there is by definition excess fluid within the body, and losing that excess is not to cause dehydration in any meaningful sense.

Dr. Davis (an intensivist, called by the Claimant) told me, and I accept, that the adjective “significant” used widely in the pleaded claim to describe the extent of dehydration was effectively meaningless: he adopted the taxonomy of dehydration as being “mild” (below 5% loss of water), “moderate” (5 – 10%), and severe (over 10%).

Whether there is dehydration such as may tend towards adverse health effects can be assessed by reference to three principal matters in the case of a child of Alexander’s age: the percentage degree of weight loss (weight loss here being a surrogate for loss of fluid from the body); biochemical markers: principally serum sodium, urea and creatinine levels (electrolytes) and the haematocrit (the percentage of red blood cells in the circulating volume); and the clinical presentation of the child. In addition, Dr. Habibi told me that for a child who had just been through an operation of pulmonary artery banding in order to address the effects of a VSD, the absence of pulmonary decompensation would suggest there had been no significant dehydration. The output of urine is another indicator – if it is well maintained, this is suggestive of proper hydration. However, more than one expert told me that one of the effects of the drug regime following an operation such as that Alexander had is to “energise” the kidneys, so that there may not be such a marked (or indeed, any) drop off in urine output even in the presence of clinically significant dehydration, and for this reason I have placed less weight on this factor than the others of which I was told. In so far as it went, however, it was reassuring.

As to weight loss, it is very difficult to know precisely how much of Alexander’s pre-operative weight consisted of excess fluid. Some of his body weight would have done so: it was the very reason why his lung and heart function was compromised, requiring the operation which followed. His weight reduced to 2.805 kg pre-operatively, by the 25^th. was at 2.68 kg, and by 26^th. was 2.63kg. Taking Alexander’s weight at admission as being 3.1kg, this is a reduction approaching 10% pre-operatively, followed by a further reduction of a little less than 5% (from a baseline of 2.805 kg) in the immediate post-operative period, taken to 25^th. September, or a little over 5% taken to 26^th.

As I was told more than once in evidence a reduction in weight due to fluid loss cannot be assumed to represent a loss only of extravascular fluid (i.e. that which can be termed “excess”): it is always likely also to involve a loss of intravascular fluid to some extent. Although fluid within the circulation is replaced by osmosis into the veins and capillaries from fluid retained in extravascular spaces, this does not necessarily occur immediately fluid is lost from the intravascular volume being, as it were, an immediate replacement of every millilitre lost from the bloodstream by a millilitre extracted from the excess fluid in the extravascular space. The experts accepted that a degree of dehydration intravascularly is necessary in order to reduce the amount of fluid retained extravascularly, and therefore some dehydration is inevitable if the therapeutic aim is to be achieved.

Though it was agreed in evidence that Alexander undoubtedly had excess fluid, and that a proportion of the reduction in fluids shown during the period to which the measurements of fluid balance related consisted of a reduction in or removal of this excess fluid, the extent of this proportion was not agreed. A millilitre of water is equal in weight to a gram (give or take minor fluctuations dependent on the temperature of the water): if the entirety of the weight loss between admission and post-operative treatment were due to loss of the surplus, extra-vascular, fluid alone, he would have lost some 430 ml. of fluid during this period, but none from the circulating volume within the blood vessels. By contrast, if all had come from the intravascular space and none from the extravascular, that loss would come entirely from the intravascular space, such a volume would be a significant proportion of the circulating volume, and Alexander would be severely dehydrated with all the attendant risks that brings.

The expert difference as to the proportion of fluid loss which could be attributed to a reduction in excess fluid was guided on the one hand by a view of the weight which Alexander could be expected to have reached at his age after birth, if unaffected by any heart condition; and on the other by examining the biochemical “markers” to see what, if any, degree of dehydration they showed: in particular, blood sodium, urea and creatinine concentrations, and the haematocrit. The weight to be expected of someone born at 35 weeks gestation as was Alexander, with the birth weight he had, was generally accepted to be around 3 to 3.1 kg given his age at admission. Since his weight on admission was 3.1 kg this gives little scope for any of it to be accounted for by excess fluid. On the other hand the biochemical markers in his blood showed some evidence of dehydration but (summarising that which I accept after hearing all the evidence and adopting Dr. Davis’ scale of mild, moderate or severe) none of them showed evidence of severe dehydration: in the main, the readings at worst hovered around the border between mild and moderate.

Dr. Shineborne and Dr. Tometzki were asked jointly to consider a question at their experts’ meeting, which read:

“Alexander’s birth weight was 2.06kg; his weight on 16 September 2005 was 3.1kg. Given Alexander’s cardiac condition, is it probable that some of this weight was due to excessive fluid retention? If so it is possible to state, on the balance of probabilities, what percentage if this is within your area of expertise?”

They agreed that the fall in weight equated to approximately 5% (Dr. Shineborne) and that the percentage of fluid retention was approximately 5% (Dr. Tometzki), although when the latter was cross-examined he said he considered that most of the loss (he estimated around 75%), when weight dropped from 3.1 to 2.8kg, was extravascular: this, of course amounts to 7.5% of the body weight being excess fluid rather than 5%. The intensivists (Dr. Habibi and Dr. Davis) agreed in joint meeting that some of the weight Alexander had gained since birth was due to excessive fluid retention, but thought that the percentage was not strictly within their expertise so would defer to the cardiologists. When giving evidence, Dr. Habibi commented that at his expert meeting with Dr. Davis they had not understood how the cardiologists had come to derive the percentage they did, because “…we agreed that we didn’t know how much fluid overload there would be, and in my opinion it was significant”. In fact, he disagreed with the cardiologists’ view. This was because he could not reconcile the large cumulative negative fluid balance with the electrolytes and Alexander’s clinical appearance: the problem was that, given those electrolytes, “...nothing seemed to have happened that was bad to this child. Clinically he seemed to be warm and well perfused, based on the notes…” . If he were to begin with the hypothesis that the extent of fluid retention was no more than 5%, then he accepted it would undermine a conclusion that there was no significant dehydration: but because he considered that the biochemical and clinical indicators (especially the latter) were inconsistent with any more than moderate dehydration at most, and further that there no signs of cardiovascular decompensation, he concluded that the estimate was no more than a guess, and in fact an under-estimate. In short, he was persuaded that more than 5% (or 6%, taking the difference between 2.8 and 2.63kg) of the fluid loss came from what had been retained water.

Turning to the biochemical markers, the normal range for an infant of Alexander’s age, following birth at 35 weeks gestation were:

The sodium concentration in Alexander’s blood fell between 144/145 on 22^nd. September; increased to 150 on 23^rd., 151 on 24^th, then dropped back to 149 on 25^th. and 146 on 26^th., 141 (27^th.) and 140 (28^th.). There was one reading, taken by blood gas sampling, on the 24^th. which went as high as 157, but the experts in their evidence were inclined generally to treat this as an outlier, and appeared to me content to accept that the “real” reading for that day was the reading of 151 reached by laboratory analysis.

Urea concentrations were between 7.1/7/7 on 22^nd. then 7.7 (23^rd) 7.5 (24^th.) 5.6 (25^th.) 6.0 (26^th) 4.6 (27^th.) and 3.9 (28^th.)

Creatinine measurements were 59 (22^nd.) 67 (23^rd.) 67 (24^th.) 63 (25^th.) 66 (26^th.) 62 (27^th.) and 53 (28^th.).

Between 22^nd. and 28^th. the haemoglobin concentration was never out of normal range.

The individual measurements outside normal range were thus, in respect of sodium, those on 23^rd. (150) 24^th. (151) 25^th. (149) and 26^th. (146); in respect of urea, only that on the 23^rd. (7.7); though for creatinine, all the measurements for the critical dates were in excess of the normal range, and in respect of haemoglobin all were within it.

No expert placed any significant reliance on the creatinine readings. Nor did it strike me that any placed heavy emphasis on haemoglobin. They concentrated on sodium and urea.

Dr. Habibi thought that a serum sodium of greater than 160 mmol/l would be indicative of “significant dehydration”, whereas Dr. Davis thought that a reasonable body of clinicians “would at least take notice when the serum sodium is above the normal range, whilst a value of 150 mmol/l or more would generate concerns regarding dehydration”. Dr. Habibi commented that at 150/151 the level had stayed high “...but in clinical terms we are not too concerned about that, because we know that is the desired effect. That is causing the osmotic gradient to suck the water out. Of course there comes a point when you have to stop doing that”. He did not think that the level had reached a danger point, nor did he understand Dr. Davis to take that view, though he accepted that a clinician should not wait for danger signs to emerge before taking action.

My conclusion from the evidence of the intensivists is that both considered that the levels of serum sodium were high, but that neither suggested that it would be necessary to take immediate action to reduce those particular levels, and each recognised that rises in serum sodium and urea were inevitable because the therapy involved a degree of dehydration. In support of the thesis that the levels of serum sodium were not alarming in the present case, Mr. Spencer pointed to two studies which were referred to by Professor Fenella Kirkham (a paediatric neurologist, called by the Claimant), those by Oddie and Forlan, in the former of which a study had been conducted indicating that an entry point for consideration of the effects of excess sodium began at levels of 150 mmol/l and not below, and in the latter of which hypernatraemia was defined as beginning at 150 mmol/l.

As for urea, Dr. Davis thought that a reading in excess of 10 mmol/l would be a marker of significant dehydration; Dr. Habibi opted for 12 mmol/l. The readings came nowhere near either, the maximum being 7.7. This is of added significance given that Dr. Davis considered that urea is potentially a more specific marker than others of severe dehydration.

As to clinical presentation, the experts began by accepting guidelines propounded by NICE (National Institute for Clinical Excellence) in 2015. None of the markers of severe dehydration which NICE identify – an appearance of being unwell and deteriorating; irritability and lethargy; sunken eyes; tachycardia; tachypnoea; and reduced skin turgor – were present in Alexander’s case. By contrast, the evidence from the clinical notes suggested a child who was doing well, appeared well-perfused and warm.

It is against this evidence that I have to consider the claimant’s case. In his pleadings he asserts that the rise in serum sodium concentration to 151 mmol/l was an indication of severe dehydration. Severe dehydration is not supported by the evidence. I reject this extreme.

I conclude that a degree of dehydration was inevitable given the chosen treatment. It was at times moderate, adopting Dr. Davis’ threefold classification, though this must be seen in context. The weight loss was equivocal as to the extent of dehydration, since to draw significance from it in favour of the claim a court must be satisfied as to the extent to which it was a loss of excess fluid or of intravascular fluid. There is very limited evidence it was a significant loss of the latter, thus suggesting to me that the extent of retained fluid was consistent rather with the instincts of Dr. Habibi than with the agreement once reached between Dr. Shineborne and Dr. Tometzki, reached without any express basis other than assumption by those cardiologists, albeit that it was reached between two experts - though their expertise has to be tempered by the fact that neither was particularly convincing in evidence for the different reasons I have given, and Dr. Shineborne later resiled from it. Of the principal biochemical markers, serum sodium showed evidence that there was a moderate degree of dehydration between the 23^rd. and 26^th. September, though the trend by then was a return to within normal range, consistent with a lessening of the degree of dehydration. The pattern shown by the concentrations of urea was similar, as was that of creatinine, but in neither case was there any strong support for there being any marked indication of dehydration of clinical significance even for a brief period. The clinical effect of such limited dehydration as there was was demonstrated by the apparent condition of Alexander, who showed no indications at all of severe dehydration, and seemed well-perfused and warm.

I was struck by the repeated refrain of a number of the experts (Dr. Shineborne, whose evidence on this I do accept, and Dr. Davis in particular) in common with the evidence of Professor Murdoch, the consultant responsible for Alexander’s care, who spoke of the need to consider the individual child, his reaction to treatment and apparent well-being rather than protocols, or whether readings of electrolytes fell precisely within range. If I had needed to do so, for this reason I would have weighed the evidence of clinical presentation most strongly, though in the event I have not needed to do so to reach my conclusion

I conclude that the level of dehydration was moderate at most, to an extent which caused no particular concern to the treating clinicians, nor which should have done: though the individual readings for serum sodium taken between 23^rd. and 26^th. were high, and the fluid balance ran to a negative balance of 239 ml., an holistic approach is to be taken (Dr. Davies recognised this where he agreed that serum sodium, concentration “should be reviewed in conjunction with both clinical indicators…and other laboratory values to confirm significant dehydration”). On this approach, I cannot be satisfied that Alexander’s condition was such that to fail to give a bolus of fluid, or to end fluid restriction, fell below the response reasonably to be expected of a competent clinician. Alexander did not appear to be so dehydrated so as to call for immediate action to relieve his condition.

That does not, however, answer the claim in its entirety, nor deal with that part of the defence which argues that even if there had been a drop below standard in the extent to which fluid intake was restricted such a drop had no causative impact.

As to the claim, Mr. Maskrey argues that the evidence shows that a cause of Alexander’s injury was relative dehydration. This was contributed to by the extent of fluid restriction following operation. To restrict fluid intake to around or just below 50% of a maintenance carries with it some risk of injury. Unless there is some acceptable reason to restrict fluid intake, to do so at all is thus to take an unnecessary risk with a patient’s health. To take an unnecessary risk is a breach of duty. Since the breach consists of causing a degree of dehydration, it has inevitably caused or contributed to the injury.

This argument depends upon there being no good reason for restricting fluid intake beyond 23^rd. September.

The evidence as to the purpose of a regime of fluid restriction in general terms and until 22^nd. September at least was uncontroversial. Although there were hints of it in argument, I do not think that, so far as the evidence was concerned, there was any sustained criticism of fluid restriction until after the 23^rd., when extubation occurred and oral feeds began again. Beyond that date, Dr. Davis thought that whereas the immediate level of fluid restriction advised in 2005 for those children who had undergone a “closed” heart procedure was 75% of maintenance, in Alexander’s case it was 50%, akin to that adopted where surgery had been “open” (i.e. conducted on bypass), even though in his case surgery was closed. To continue it for 4 days post-operatively at 50% was to apply a restriction to an extent which went beyond even that which applied where there had been open heart surgery. Dr. Tometzki thought the level of restriction was more usually related to post-operative management after by-pass procedures. By contrast, Dr. Habibi considered that the approach adopted in the Defendant hospital had been in use for some time, and was not significantly outside the normal range of practice in other ICU units in the UK. Dr. Shineborne provided no useful comment, since he proceeded from a view that there had been greater hydration than there was, for which there was no sufficient evidence.

Professor Murdoch explained in his witness statement that maintaining fluid intake at 2ml/kg/hr following cardiac surgery (whether open or closed) was unit policy in 2005 and continues to be so. He relied in part upon guidelines produced by Frank Shann, the twelfth edition of which was produced in 2003 and was current at the relevant time, which he said supported this. A negative fluid balance was deliberately maintained. He recognised that some units, such as that of the Defendant, had very restrictive regimes, others had liberal regimes: “The unit at Guy’s and St. Thomas’s is restrictive and is restrictive for a purpose, because we know that fluid balance cumulation leads to longer length of stay and higher mortality, both in the paediatric and the adult literature, and in the adult literature excess extravascular water and increase in that makes those patients higher risk…so… the reason we keep [fluid restriction] tight is we don’t want these to get more fluid inadvertently going back into the lungs, which makes it more difficult for them successfully to come off the ventilator.”

Mr. Maskrey argues that a fair reading of the statements of Professor Murdoch and the cardiologist Dr Miller who was engaged in the treatment of Alexander suggested that the Defendants’ case was that it had a policy, and applied it, without any or sufficient consideration of the circumstances of the claimant, the nature of his operation, or the balance of risks and benefits. There was nothing to show why the restriction was maintained for the length of time it was. Moreover, it was inappropriate to apply the same approach to closed heart operations as would be applied to operations on by-pass, yet this was precisely what was done. Professor Murdoch seemed to agree at one stage that a distinction should be made (as a textbook by Elliott & Delius suggested) but then to seek to disagree with the textbook view.

This was compounded by the fact that it was not until he began to give evidence that it became apparent that he had far more to say than was to be anticipated from his witness statement, and some of this was new to the Claimant’s team.

Mr Maskrey urged that there was no clinical need to think that the Claimant would become overloaded with fluid following his operation because the operation appeared to be successful – the band had been well-placed, the liver (which together with the lungs would be the organ most affected by excess fluid) had reduced in size, the heart was not enlarged, there was no generalised swelling, his respiratory rate following extubation was normal, and he seemed to progress well in air. There was no X-ray evidence of pulmonary plethora; and Captopril would reduce the risk of fluid overload due to increased pulmonary flow.

Though neither counsel made a specific point of it, I have also borne in mind that the 23^rd. was a Friday; that Professor Murdoch though nominally responsible for overseeing Alexander’s care would be absent for the weekend until the morning of the 26^th., and I have to be alert to the possibility that there might be a temptation at a weekend to leave the parameters of drug and fluid intake untroubled by any alteration.

In my view, Professor Murdoch’s reasoning for restricting fluids as was done related to risks that might eventuate, rather than responding to any particular change in presentation. There is no fault in this in principle: to the contrary, proper care must anticipate risk, and head it off so far as possible. In the case of an expert clinician, that appreciation of risk is born out of experience. I accept his evidence that he was concerned to balance the risk that Alexander would have to be re-intubated against the risks of fluid deprivation. That re-intubation was a real risk is demonstrated by something Dr. Shineborne said which, despite my reservations about his evidence as a whole, I nonetheless accept – that positive pressure ventilation (in other words, intubation) can drive fluid out of the alveoli; if it is discontinued there is then a tendency for the lungs to retain fluid again. There is thus a period straight after extubation where it is necessary to see how the child is progressing before increasing the rate of fluid intake adopted before then.

When Professor Murdoch told me that he had seen many children who had had similar types of procedures to Alexander, whose fluids had been increased too quickly and who had as a result to go back on to the ventilator, with adverse consequences, it was plain to me he believed this passionately. He had had experience of patients who had undergone pulmonary artery banding who had “misbehaved and died on you”; there was a serious risk of mortality and morbidity.

Mr. Maskrey’s points, summarised at paragraph 78 above, essentially show that the risks to which Professor Murdoch referred had not, or not yet, materialised. However, that is (a) to take something of a retrospective view; (b) cannot show there was no risk that they might yet have materialised had no precautions been taken; (c) invites focus on any exercise of judgment by the clinician responsible at the relevant time as to how best to balance the risks of abandoning fluid restriction against the risks of continuing it.

I accept therefore that there was a proper aim sought to be achieved by fluid restriction. I accept also that in some units it may continue for as long as, and to such a degree as, it did at Guys. This evidence was given to me by Dr. Habibi. Though he has been of consultant status for over 25 years, has been convenor of the Intensive Care Specialty group of the Royal College of Paediatrics and Child Health, and is currently director of Paediatric Intensive Care at St. Mary’s NHS Trust Mr. Maskrey pointed out that he has not worked consistently in one of the few centres in the UK which conduct heart surgery on neonates and infants. He had less experience than others of paediatric cardiac cases. This is so: but he is not without experience, and I accept his evidence that he is aware of the regimes which different centres apply because of the work he does with CATS (Children’s acute transport services), and the links which his unit has in particular with the Brompton as a centre for paediatric heart surgery.

The aim was to keep the risk of re-intubation to a minimum. Though Alexander had had a successful operation, he was still an ill child. There remained a significant left to right ventricular flow of blood, since the holes in the septum of his heart remained to be closed at a later date. The arterial band would restrict some flow, and increasingly so as Alexander grew, but it had to allow for that growth and so would have only some effect on flow to the lungs. On admission, Alexander had had a significant fluid overload – the extent of which may be gauged by the fact that despite the negative fluid balance mentioned above he showed few signs of dehydration, and none of serious dehydration. There was plainly a chance it might recur, despite a drug regime involving Frusemide, a diuretic, and Captopril, augmenting peripheral flow.

I do not accept that the policy in place was applied unthinkingly to Alexander, as the Claimant suggests: for instance, having reached a large negative fluid balance, Frusemide was disapplied on 25^th. September.

Accepting, as I do, that treatment has to focus on the individual patient, and the risks (though understood by experience of many patients) are those relevant to that patient, any judgment has to balance the possibility that those risks might materialise, coupled with the severity of any consequence if they do, against the risks of continuing to restrict fluid intake. Here, it is clear that Alexander was closely monitored for any signs that the balance needed to be redrawn. An example that the clinicians were sensitive to this is that in the light of some observations not only was Frusemide, but also (where measurements of blood pressure were concerned), doses of Captopril, were withdrawn. I have already concluded that Alexander was at most moderately dehydrated. I accept that individual readings were not such as to justify alarm or to require a change of direction; and conclude that taking a proper holistic view Alexander showed no signs of trouble which would be apparent to a treating clinician until the night of the 27^th.. A blockage of the venous drainage causing an infarct is very rare. On the other hand there was a real possibility that he might need reintubation if prophylactic measures were not taken.

I am consequently unable on the evidence to hold that to draw the balance where it was drawn was negligent.

In summary, though the regime operated at Guy’s and St. Thomas’s was strict, and probably stricter than that in most other units, the clinical treatment of Alexander so far as fluid restriction was concerned did not fall below the standard to be expected so as to place the hospital in breach of its duty to him.

I have considered the allegations in respect of Captopril and those in respect of fluid restriction separately, and rejected each. I have also considered whether the two should be taken in combination, and whether if so this would lead to a different conclusion. I have no sufficient evidence to suppose that it would.

In the light of these conclusions, by reason of which the claim fails, it is strictly unnecessary to consider the issues of causation which arose. Had I needed to do so, I would have concluded that the evidence established that the mechanism of injury was as set out at paragraph 7 above. The formation of the clot might be attributable to a number of causes, but what has also to be explained is the propagation of the clot, and its failure to clear, which was responsible for the infarct at least in part if not in whole. The only credible explanation offered in evidence was that the moderate dehydration led to the formation of the clot, and cerebral perfusion pressures were insufficient to clear the block to the venous drainage system. This was rare, unexpected, and tragic. The fluid regime had been relaxed, though not entirely, by the 25^th. which was the first date on which the clot would have formed, and from that date onward there were positive fluid balances (small though they were), the biochemical indicator of some dehydration – serum sodium – was on a trend of returning to normal range; urea levels were dropping; and creatinine stable. Blood pressures on 25^th. and 26^th. gave no indication of hypotension, and the trend at that stage appeared to be upward rather than downward. A clinician could not reasonably be expected to foresee that a clot to the brain might occur, and almost certainly would not have foreseen the actual injury which befell Alexander.

In short, though I would have found the factors I have set out to have caused the injury, the injury would not have been reasonably foreseeable as a consequence of those factors to any clinician treating Alexander during the critical periods of 25^th. and 26^th. September.

I have been driven to reach the conclusions I have despite what might be seen as unfortunately late developments in the Defendant’s presentation of their case: If any submissions arise in respect of costs in this regard, I will consider it in a separate ruling: if the parties wish it, they may agree directions for this subject to the approval of the court.