Sorry doctor, the ABG machine doesn’t believe you…

The most extreme respiratory acidosis ever
Extreme respiratory acidosis

This ABG was taken on 15L/O2 min.

A loose translation:

# = uh oh
! = the patient has probably arrested in the time it took you to analyse this ABG
? = lol, that’s not possible.

It came from the left radial artery of a man in his early 70s who was referred by his GP for seeming drowsy and cold with a saturation of 71%.

He had a PMH of COPD (home oxygen) and CCF.

On arrival, he was too drowsy for a history.

On examination, there was use of accessory muscles although he seemed fatigued. There was diffuse wheeze and bilaterally reduced air entry.

The SpR treated it as a severe exacerbation of COPD.

What are the common causes of an exacerbation of COPD?

Most community-acquired infections are caused by Streptococcus pneumoniae, Haemophilus influenzae and Moraxella catarrhalis. Staph aureus classically causes pneumonia following influenza infection. There are also viral triggers and pollutants.

 What is best medical management?

  • initially controlled oxygen to maintain SaO2 88–92%
    adjust target range to 94–98% if the PaCO2 is normal (unless there is a history of previous NIV or IPPV) and recheck blood gases after 30–60 min
  • nebulised salbutamol 2.5–5 mg
  • nebulised ipratroprium 500 micrograms
  • prednisolone 30 mg
  • antibiotic agent (when indicated).

What are the management options in COPD with hypercapnic failure?

Hypoxic (pO2 <8kPa) Hypercapnic (pCO2 > 6kPa) Acidotic (pH < 7.35) Treatment
No No No Medical management
Yes No No Increase inspired O2
Target sats: 94-98%
Yes Yes No Titrate O2 down if pO2 > 8kPa.
Target sats: 88-92%
Yes Yes Yes NIV or intubation. May need discussion with ITU.

What are the five factors that need to assessed before starting NIV?

  • their pre-morbid state
  • the severity of the physiological disturbance
  • the reversibility of the acute illness
  • the presence of relative contraindications (see Table 1), and
  • where possible, the patient’s wishes.

The ideal COPD/NIV candidate is a well-informed, compliant patient who is able to protect his own airway. He has a reversible COPD exacerbation, and enjoyed good pre-morbid health. His pH is between 7.25 and 7.35 (if pH <7.25 invasive ventilation may be more appropriate), his pCO2 > 6kPa and he is moderately hypoxic (severe hypoxia is an indication for invasive ventilation).

The contraindications are listed below:

  • Life-threatening hypoxaemia
  • Severe co-morbidity
  • Confusion/agitation/severe cognitive impairment
  • Facial burns/trauma/recent facial or upper airway surgery
  • Vomiting
  • Fixed upper airway obstruction
  • Undrained pneumothorax
  • Upper gastrointestinal surgery
  • Inability to protect the airway
  • Copious respiratory secretions
  • Haemodynamically unstable requiring inotropes/pressors (unless in a critical care unit)
  • Patient moribund
  • Bowel obstruction

Having assessed the five criteria above, you should be able to reach a decision about what treatment options are suitable for this patient:

  • requiring immediate intubation and ventilation
  • suitable for NIV and suitable for escalation to intensive care treatment/ intubation and ventilation 
if required
  • suitable for NIV but not suitable for escalation to intensive care treatment/ intubation and ventilation
  • not suitable for NIV but for full active medical management
  • palliative care agreed as most appropriate management

I will find out more about NIV from the respiratory team and write up any clinical pearls they have in the next post.


Cola’s is tightest

It’s the Monday from hell. The Monday you will use as your baseline to compare all future Mondays to, and always conclude that it was never as bad as that Monday. It’s a Monday when the code to the door you normally use to enter the hospital has changed, and you are trapped outside as Shrewsbury briefly forgets that it isn’t Antartica. It’s a Monday when the phlebotomists have, for the first time in the history of phlebotomy, decided to do their round before 8am. It’s a Monday which shouldn’t even be called Monday in order to preserve the mildness of the phrase ‘Monday blues’.

You see a 35 year old lady patient with classic cholecystitis. The three key features mention in the BMJ review article are staring you in the face:

  1. Tenderness in RUQ (Murphy’s sign being helpful in pointing you away from simple biliary colic and towards cholecystitis)
  2. Constant pain in RUQ (at least 12 hours of constant pain according to the same article or at least 6 hours according to emedicine , which distinguishes it from the 1-5 hours of biliary colic)
  3. An inflammatory component e.g. raised CRP or WCC, or temperature

This patient has had RUQ pain constantly for 8 hours and is febrile. She hasn’t eaten for 12 hours, and has been sick once. Her WCC is 14.3, CRP 55 and LFTs are all normal. Her pulse is 90, BP 112/80, RR 18, Sats 100% on air and temperature 37.5.

On examination, she is Murphy’s positive. There is no peritonism.

A chest xray shows no free air under the diaphragm.

An ultrasound abdomen is booked for tomorrow morning.

What should we do now from a medical point of view?

It seems people usually keep them NBM and give IV fluids. They often give paracetamol for the temperature/pain and breakthrough oramorph as needed. They often give antibiotics for the cholecystitis.

Antibiotics means “triple therapy” in this trust i.e. gentamicin, metronidazole and benzylpenicillin. This is a broad surgical prophylaxis against anything and everything. I always wondered why we don’t just give a third-gen cephalosporin and metronidazole, as the cephalosporin would cover most bacterial save MRSA, pseudomonas (unless ceftazidime is used), some intracellular bacteria and anaerobes. Of these, only the anaerobes would need to be covered in this scenario, which would be done by metronidazole. It would avoid the resources wasted on gentamicin levels. When the levels aren’t taken, it can fall to the you on call at 11pm to decide whether or not its worse to skip the only antibiotic providing decent gram negative cover versus causing irreversible ototoxicity and nephrooxicity. Whatever answer you decide will be wrong on tomorrow’s ward round. Especially if it’s Monday.

Part of the reason may be to do with C Diff and some extreme fines – about £450,000 per case of C diff over the allowed amount per trust. This may explain why we rarely use co-amoxiclav for anything, and cephalosporins are a collector’s item. I don’t think I can remember the last time I saw clindamycin.

In any case, the suspected bacteria in superinfection of cholecystitis are actually covered by a second-generation cephalosporin alone. According to BMJ Best Practice:

“Antibiotic therapy should include coverage against microorganisms in the Enterobacteriaceae family (e.g., a second-generation cephalosporin or a combination of a quinolone and metronidazole)”

I was finding it difficult to keep making decisions about gentamicin at 11pm. Naturally in a patient with already poor renal function, or in the elderly,  you may err in favour of not giving the gentamicin. You need to balance it up with the risk of the cholecystitis, and I did not know enough to really assess this. I needed to find out exactly why we give antibiotics in cholecystitis.

What normally happens if you don’t give the patient anything?

Cholecystitis if left untreated would usually self resolves in 5-7 days.

Otherwise it may progress through 5 stages:

  1. Oedematous – 2 to 4 day, gallbladder tissue is basically intact
  2. Necrotising – 3 to 5 days, oedema with areas of haemorrhage and necrosis
  3. Suppurative – 7 to 10 days, WBCs present within the gallbladder wall, with areas of necrosis and suppuration. Intra-wall and pericholecystic abscesses may occur.
  4. Chronic – occurs after repeated episodes of mild attack (mucosal atrophy and fibrosis of the gallbladder wall)
  5. Emphysematous – air appears in the gallbladder wall due to infection with gas-forming anaerobes (usually diabetics)

What do I risk by not giving the patient anything?

If the cholecystitis doesn’t resolve, you risk:

  • Necrosis and eventually perforation of the gallbladder
  • Gallstone ileus
  • Superadded bacterial infection, which can lead to sepsis

Can cholecystitis ever be purely medically managed?

Possibly Grade I, which means “no organ dysfunction and limited disease in the gallbladder; responds to initial medical treatment”.

It is worth emphasising that the ideal management is nearly always going to be surgical.

When do we need surgery?

If Grade I fails to respond to treatment, or the patient is Grade II (no organ dysfunction, but extensive disease in the gallbladder and disease duration more than 72 hours) or III (organ dysfunction), early laproscopic cholecystectomy is preferred. This may be limited by technical difficulty, which may require conversion to open. Alternatively, if the patient is really unwell, a cholecystotomy may be performed, with an interval cholecystectomy at 6-8 weeks.

How does the pathophysiology relate to the principles of medical management?

According to the Tokyo Guidelines:

“In the majority of patients, gallstones are the cause of acute cholecystitis. The process is one of physical obstruction of the gallbladder by a gallstone, at the neck or in the cystic duct. This obstruction results in increased pressure in the gallbladder. There are two factors which determine the progression to acute cholecystitis — the degree of obstruction and the duration of the obstruction. If the obstruction is partial and of short duration the patient experiences biliary colic. If the obstruction is complete and of long duration the patient develops acute cholecystitis. If the patient does not receive early treatment, the disease becomes more serious and complications occur.”

Then, according to the first BMJ article:

“Over 90% of cases of acute cholecystitis result from obstruction of the cystic duct by gall stones or by biliary sludge that has become impacted at the neck of the gall bladder. Obstruction of the cystic duct causes the intraluminal pressure within the gall bladder to increase and, together with cholesterol supersaturated bile, triggers an acute inflammatory response. The trauma caused by the gall stones stimulates the synthesis of prostaglandins I2and E2, which mediate the inflammatory response. Secondary bacterial infection with enteric organisms (most commonly Escherichia coliKlebsiella, and Streptococcus faecalis) occur in about 20% of cases.”

This means that:

1. NSAIDs may be helpful

2. NBM may in itself be helpful. This is because the intraluminal pressure plays a part in driving the inflammatory response. CCK stimulates gallbladder contraction, and CCK release is stimulated by fat containing food. This also explains why the pain can be brought on by fatty foods.

For the initial medical management in general, according to BMJ Best Practice:

“Patients are observed and treated with oral antibiotic drugs or even observed without antibiotics. The pathophysiology of acute cholecystitis is cystic duct obstruction, which causes an acute sterile inflammation. Secondary infection of the gallbladder space by bacteria may follow. Antibiotics are required if infection is suspected on the basis of laboratory and clinical findings. Antibiotic therapy should include coverage against microorganisms in the Enterobacteriaceae family (e.g., a second-generation cephalosporin or a combination of a quinolone and metronidazole); activity against enterococci is not required. 

Nonsteroidal anti-inflammatory drugs (NSAIDs) such as diclofenac or indomethacin are recommended as part of the medical treatment for their analgesic effects and their inhibition of prostaglandin release from the gallbladder wall.”

And antibiotics?

According to emedicine:

“The guidelines of the Infectious Diseases Society of America recommend that antimicrobial therapy be instituted if infection is suspected on the basis of laboratory and clinical findings (white blood cell count >12,500 cells/µL; temperature >38.5°C) as well as radiographic findings (eg, air in the gallbladder or gallbladder wall).

BMJ Best Practice adds on:

“The pathophysiology of acute cholecystitis is cystic duct obstruction, which causes an acute sterile inflammation. Secondary infection of the gallbladder space by bacteria follows. Antibiotics are required if infection is suspected on the basis of laboratory and clinical findings.”

What does this mean for our patient with no gentamicin levels?

I’m not sure. How much of a difference do antibiotics make, if any, given only 20% end up with super-added infection? Risk factors for infection include the duration of the cholecystitis and immunological status of the patient. Without numbers though, it’s hard to weigh things up. Has anyone found evidence of the benefit of antibiotics pre-operatively in cholecystitis?

The sats probe is not God

“Doctor, can you come see this patient please?”

“What’s happened?”

“Her saturations keep dropping. We keep moving the probe but the best we get is 82%.”

“What were her saturations like an hour ago?”

“Oh, you know, around 95, 96 on air.”

“How does she look?”

“She looks distressed”

“OK, be there in a bit. Put high flow oxygen on.”

It’s always the patients on the furthest ward from where you are. There’s this corridor on the way from Day Surgery unit to the rest of the hospital which has three glass walls and the floor. This looks lovely in the day, but it’s Siberia during the night. There is a door at either end of this corridor. You can leave Day Surgery through the first one, but the probability of the distal door being locked is determined by the product of how cold you are and how urgently you need to get through.

When I got there, I saw a slightly breathless 75 year old lady who was asking me  what all the commotion was about. She said she felt a little SOB, but not massively. Her RR was 18. The chest was clear.

Her radial pulse felt inconsistent. Her cap refill was less than two seconds, her BP 108/70 and her pulse 100 and irregular.

An ECG showed atrial fibrillation. This was new for her.

She had been admitted for a Hartmann’s procedure and was day 1 post operatively.

I felt that the saturations did not match up with the clinical picture, so did an ABG which showed a pO2 of 44.3 kPa. This showed that the saturations were misleading.

I later discussed this with the medical registrar. He told me that one of the commonest explanations for fluctuating saturations is atrial fibrillation. He also told me to always check the probe on my own finger if the clinical picture does not match up with the saturations.

We started her on bisprolol 1.25mg OD. She was already warfarinised for a previous PE.

Bottom line: Always go with your clinical impression over the sats machine, and get an ABG if in doubt.

Good links:

The WHO has guidance on pulse oximetry which mentions arrhythmias as a cause of false readings.