Case 1(a)

Mrs S was a 34 year-old female who presented to casualty with three episodes of haematemesis over the past week.  She thought that she vomited about 50 mls of frank blood each time.  She had no other past medical history, nor did she have a recent history of any NSAID [non-steroidal anti-inflammatory drug] or alcohol usage.  She had never noted any malaena stools and did not suffer from dyspepsia or heartburn.  She wasn't a smoker, was unemployed, and had no significant family history or travel history.

On examination, she was noted to be cachectic, and exhibited severe pallor.  Her blood pressure was 90/48, with a pulse of 105.  The rest of her vitals were normal, as was the rest of her examination, other than a 3/6 ejection systolic murmer heard best over the aortic and pulmonary areas.  The murmur didn't radiate.  

Question 1: What are the commonest causes of haematesis?

The top few causes are:

• gastric ulcers (25%)
• duodenal ulcers (25%)
• erosive gastritis (20%)
• Mallory-Weis tear (10%), and
• oesophageal varices (5%).  

As noted from the percentages, there are other causes too, however.

An arterial blood gas analysis was done, and the relevant results are as follows:

• pH: 7.36
• pO2 : 80 mmHg (10.6 kPa) [Note: this is normal for the altitude at which it was recorded]
• pCO2: 29 mmHg (3.86 kPa)
• Base excess: -7.5
• Lactate: 4.0
• Haemoglobin: 2.3

Question 2: Comment on the blood gas analysis.

The results indicate a metabolic acidosis, with respiratory compensation (so that the pH is within normal limits).  The most likely cause of the metabolic acidosis is a lactic acidosis, as the lactate is raised. The patient has an extremely severe anaemia.

The patient was dripped, and Ringer's lactate was infused.  Blood was then ordered, and the patient was sent to the acute medical admissions ward.  [Why not to the surgical acute admissions ward?  I've no idea.  There's always the perennial fight between medicine and surgery, but upper GIT bleeds should be surgical until proven otherwise.]  Here, we transfused three units of packed cells, but were careful to pull bloods before the transfusion, so as to get an accurate reflection of things like iron stores, B12, folate and reticulocytes.

Question three: What is the most likely reason for the lactate to be elevated?

There are many reasons for hyperlactataemia, but anaerobic metabolism is likely to be the one here.  A significant upper GIT bleed would deplete the intravascular blood volume, leading to hypovolaemia.  This would cause inadequate tissue perfusion.  Since blood carries oxygen, a lack of blood to a particular tissue means that the tissue also won't get enough oxygen.  As a result of this, the tissue's cells must perform their metabolic functions as best they can by (the vastly more inefficient) anaerobic metabolism.

However, there is another possibility, given that the haemoglobin is so low.  Oxygen dissolves rather poorly in blood, and so we use haemoglobin to bind to it.  This hugely increases the amound of oxygen carried by the blood.  A very low haemoglobin therefore means that the blood's oxygen content is much reduced.  Therefore, even with adequate tissue perfusion, the body's cells still might not be getting enough oxygen for aerobic metabolism.  As before, they will therefore have to switch to anaerobic metabolism.

A gastroscopy was performed soon thereafter, which was... entirely normal.  No ulcers, no erosions, no blood (old or new) whatsoever, anywhere from the the oesophagus to the proximal duodenum.  This was unexpected, but we soon received the patient's blood results, after which things all made sense.  We'll cover that in the second part of this case.