PEM Adventures Chapter 3 – Toxic Trouble

It’s time for another PEM adventure. Join us on another journey (with an inbuilt time travel machine) as we manage Francesca, a teen who dreams of being a pop star…

Meet Francesca, a 15-year-old girl who dreams of being a pop star.  She is making ripples in the world of teen music videos and has a HUGE audition tomorrow for a music video. But for the last 24 hours, she’s been feeling a little shaky and pretty nauseous. Putting it down to nerves, her mother (also her agent) continued packing for the big trip. But Francesca vomited, had two episodes of diarrhoea and then spiked a fever, and her mother knew she needed to get her fixed. Fast.

It’s the middle of a run-of-the-mill shift for you. You’ve just fished a bead from a child’s ear, reduced a decently angulated forearm fracture and admitted a child with pyelonephritis. When you look at Francesca, though, you know she’s sick. She is agitated, clammy, flushed, and febrile at 39.3°C. She’s tachycardic at 130, with a bounding pulse, blood pressure of 128/74 and normal heart sounds. She is tachypnoeic at 24, with sats of 98% in air and a clear chest. Her abdomen is tender in the epigastric region, with no guarding or rigidity. Her GCS is 15 with no focal neurology. Her triage weight is 44kg.

You grab the sepsis trolley, insert the cannula, and send some bloods: FBC, coagulation, CRP, renal and liver function, and blood cultures. You run venous gas, and this is what you see…

That lactate is horrific. You hastily prescribe a 20ml/kg bolus of 0.9% saline and a broad-spectrum third-generation cephalosporin. But what’s your next step?

Give paracetamol and treat that fever

You prescribe some paracetamol. Easy enough. And then you go back and think about what to do next.

Close the tab and think about some more choices, or move on to the next section.

Give more saline – she’s shocked

You give another 20ml/kg of 0.9% saline and reassess.

Her heart rate drops a couple of beats per minute, but it bounces up again. So you give her more saline. But her heart rate goes up a bit higher. And higher again. She begins to hyperventilate. Heart sinking, you repeat her gas. Her pH has dropped, her lactate has climbed, and her potassium looks horribly high. This was NOT supposed to happen. The saline has done exactly the opposite of what you’d like it to do. How could that be? 

You wish you could go back in time and make that choice again. Luckily for you, that’s exactly what the inbuilt time travel machine is for.

Close the tab and review the choices again. If you have no further action in mind, move on to the next section.

Give Hartmann’s or Plasmalyte – she’s shocked

You give another 20ml/kg, but instead of reaching for the saline this time, you go for Plasmalyte (or Hartmann’s, if that’s your fluid of choice). You reassess. Her heart rate drops a couple of beats per minute, but it bounces up again. Her pulses remain bounding, BP holds, and JVP isn’t raised with no rales in her chest, so you give another bolus and reassess. The same thing happens: a minuscule response but nothing substantial. You don’t make things worse, but you can’t seem to make things better, either. Why isn’t fluid bringing down Francesca’s heart rate?

Close the tab and review the choices again. If you have no further action in mind, move on to the next section.

Get an ECG. This could be an arrhythmia!

Her ECG shows a sinus tachycardia. She’s in sinus rhythm with a p wave before each QRS and a normal p wave axis; her QRS axis is normal, and intervals are normal, too. There are no voltage criteria for ventricular hypertrophy and you can’t spot any subtle ST changes or delta waves. You use a handy ECG proforma to double-check, but apart from the tachycardia, it all looks fine. So you go back to Francesca and think about what to do next.

Close the tab and review the choices again. If you have no further action in mind, move on to the next section.

Get an abdominal US. She does have that abdo pain after all

You have recently been on a POCUS course and want to try out your ultrasound skills. You ultrasound her abdomen. It looks normal. The eminent professor of ultrasonography wanders by. You ask him to double-check your findings. He agrees the ultrasound is normal. 

Close the tab and review the choices again. If you have no further action in mind, move on to the next section.

While pondering what to do, you receive a phone call from your bank. It’s noisy in ED, so you pop into the corridor. The bank tells you they’ve just realised they owe you a couple of hundred pounds (*replace pounds with Euros, Dollars, Australian Dollars or any other local currency). That’s great news! Smiling, you type out a quick text to your best friend. “Epic windfall. Celebrate later in Chinatown?” You can almost smell the chow mein. Your stomach rumbles. It’s time for some lunch. You let Francesca’s nurse know you’re going for a break.

As you finish your sandwich, Francesca’s nurse comes rushing to find you. What with the phone call, the texting and the lunchtime queue in the canteen, it’s been almost three-quarters of an hour since you last reviewed Francesca. She is far more agitated. The monitor is alarming. Her temperature is now 40.2°C, and she’s very sweaty. She looks a little blue. With a heart rate of 159, BP of 108/72, respiratory rate of 32 and O₂ saturations of 93% in air, things are not looking good.

Her mother wails, “Will she be better for her audition?!”

You repeat her gas. It’s not good—her lactate is now 9.3, and her pH is down to 7.03.

Her nurse hands you an ECG. Scanning it, you spy peaked T waves, wide QRS complexes and a prolonged PR.

Hang on! What was the potassium on that gas?! You snatch up her gas – her potassium’s 7.2! How are you going to bring that potassium down? What will you prescribe first?

Phew! Francesca’s ECG rhythm is improving. Crisis averted. Or is it?

By now, Francesca is so agitated that it’s becoming impossible to keep her in bed. “I have to rule out an intracranial infection”, you think to yourself. She needs a CT.

Her nurse begs you to give her a sedative. This makes you a little anxious (pun intended). You know that sedation in a sick child can be lethal. So, how will you manage her agitation?

You want Francesca out of the ED—this is too stressful!

Thankfully, the PICU has a bed. You compassionately explain to Francesca’s mum that the PICU team will work very hard to treat Francesca, but she’s very, very sick. Her mum starts crying, “She’s such a beautiful girl! She was going to be famous! She’s worked so hard to lose weight for her audition!”

Internal alarm bells start ringing.

“Hang on! How has she lost weight?”

Eyes wide, you ask her mother, “Has she been taking something?!?”

Just as you garble this, Francesca’s dad arrives. He’s found a bottle of pills in Francesca’s room. The label says DNP. They were next to her exercise bike.

You ask the switchboard to put you through to the national toxicology advice line. The toxicologist who answers the phone tells you that DNP, short for dinitrophenol, is a diet pill that’s illegal in most countries but quite freely available over the internet. It’s a fat burner because DNP shortens mitochondrial ATP production by uncoupling oxidative phosphorylation. Because ATP can’t be produced, the metabolic rate increases and energy is released as heat. People who take it literally burn fat. But even a single pill can lead to uncontrolled hyperpyrexia, and its toxic effects are increased with exercise.

They tell you that Francesca’s bloods must be monitored closely. Her liver function will deteriorate as her liver literally cooks from within, she will become hypoglycaemic as her glycogen stores are consumed, and she’ll become hyperkalaemic. Monitor her methaemoglobin, and if it reaches 30% or if there are signs of tissue hypoxia, give her methylene blue.

They give you a long list of treatments, including…

Cold intravenous fluids…

…ice packs…

…gastric and bladder cold fluid lavage with peritoneal cooling if you can…

…and Dantrolene…

…and if that fails… then a cooling heat-exchange central line… or ECMO if you’re stuck.

That temperature has to come down.

You thank toxicology, replace the handset, and think to yourself, “Now where will I find Dantrolene?”

But while you’re pondering this, things go from bad to worse. Francesca’s temperature continues to climb. She’s now 41°C. She’s boiling. Sweat drips onto the sheets. She starts to have a generalised tonic-clonic seizure. You give her a dose of IV. Lorazepam, but she continues to seize.

What will you give next?

Things can’t get any worse, right? Wrong. She is making a funny snoring noise. You’re worried about her airway. You fast bleep the anaesthetist. Finally something’s going right, he’s just walking past, and he’s in resus in less time than you can say “dinitrophenol.” He’s up to speed quickly and agrees she needs a tube. Your RSI cocktail of choice is ketamine (1-2mg/kg), fentanyl (1mcg/kg) and rocuronium (1-2mg/kg). It’s the least cardio-unstable combination of drugs, and you don’t want to make things worse. But, luckily for you, the  anaesthetist is a clever guy and says, “Let’s avoid fentanyl since she’s hyperpyrexic as fentanyl’s serotonergic – we don’t want to raise her body temperature any higher than it is already.”

The resus nurse mishears his instruction and almost makes a fatal mistake. Spying a syringe labelled suxamethonium, the anaesthetist (who you decide is your new best friend) calmly says, “No suxamethonium. Her potassium is high. She’ll arrest with sux.”

He intubates successfully using midazolam, propofol and rocuronium.  She’s easy to ventilate.

Finally Francesca’s ready for PICU. With cold fluids, ice packs and Dantrolene her temperature comes down to 37.9°C. You hand her over with clear instructions to avoid…

…serotonergic drugs (put away that fentanyl)

… or drugs that prolong QRS (don’t even think about phenytoin if she fits again)

…and to set up ECMO if her temperature climbs again.

Eighteen months later, you watch Francesca perform live in Eurovision. She receives “Douze points!” from every country, setting the record for the highest ever Eurovision score. She campaigns for better awareness of body image in girls and is vocal about the dangers of diet pills.

But let’s hop back in that time travel machine one last time and see what you learned from her case…

You find this review article about DNP.

Grundlingh J, Dargan PI, El-Zanfaly M, Wood DM. 2,4-dinitrophenol (DNP): a weight loss agent with significant acute toxicity and risk of death. J Med Toxicol. 2011;7(3):205-212

Fascinatingly, along with all the clinical management advice you received from your friendly toxicologist, it also tells you a bit about the history of DNP. You’re intrigued to read that the first death from DNP was over 100 years ago, in 1918, secondary to occupational exposure to DNP powder. 

It was used in France to manufacture munitions during the First World War. In 1933, it was discovered that human consumption led to significant weight loss. It became very popular as a weight-loss drug, but within five years, it was recognised as extremely dangerous and was labelled as “not for human consumption” by the FDA in 1938. Anecdotally, it was prescribed to Russian soldiers during World War II to keep them warm.

It all went wrong in the 1980s (didn’t it all?). An American doctor prescribed DNP tablets to thousands of patients through his private weight loss clinic. In 1986, he was convicted for drug law violations, fined, and prohibited from dispensing DNP to patients. But this didn’t stop him. He was eventually jailed for fraud in 2008. But DNP is still out there and sadly widely available on the internet…

So, what has Francesca’s case taught us (aside from reminding us how very cool the Eurovision Song Contest is)?

1. Infection isn’t the only cause of fever

Keep your differentials open. You only need to Google ‘differentials fever + tachycardia’ and the first thing that pops up is a 2013 article titled, ‘Intoxications Associated With Agitation, Tachycardia, Hypertension, and Fever: Differential Diagnosis, Evaluation, and Management.’ (True as of 1st November 2020).

Toxicological agents include drugs that cause:

• Serotonin Syndrome: some antidepressants including SSRIs, SNRIs and lithium, anticonvulsants such as valproate, analgesics such as fentanyl, antiemetics such as ondansetron and street drugs such as cocaine, ecstasy, methamphetamine and LSD.

• Neuroleptic Malignant Syndrome: ‘typical’ antipsychotics such as haloperidol, newer ‘atypicals’ such as risperidone and olanzepine, antiemetics such as metoclopramide and promethazine.

• Malignant Hyperthermia: an inherited skeletal muscle disorder triggered by inhaled anaesthetics, succinylcholine, heat or exercise.

• Sympathomimetics: cocaine, ketamine, ecstasy, amphetamines, synthetic cannabinoids.

Toxicology isn’t where it ends though. In our COVID world we’ll be used to including inflammatory syndromes like PIMS-TS to our list of differentials, but don’t forget other inflammatory syndromes including inflammatory bowel disease and rheumatological; oncological presentations; intracranial causes (bleed, tumour, basically anything that damages the hypothalamus can dysregulate temperature control); endocrine causes like thyroid storm, adrenal crisis… and the list goes on.

2. Engage your toxicology colleagues early

Even if you don’t think the primary cause is toxicological, as soon as it could be, pick up the phone and call your regional/national toxicological service. Sedatives, anticonvulsants, anaesthetic induction cocktails… there are many ways things can go wrong. Ask a friend for advice before prescribing drugs in a potentially unstable situation.

3. Familiarise yourself with the management of acute behavioural disturbance

Acute behavioural disturbance can be a very challenging situation to manage. RCEM, the Royal College of Emergency Medicine in the UK, has a short guideline explaining the pros and cons of the different drugs of chemical restraint.

Although not specifically tailored to paediatric presentations, the explanation of the drug side effects is a useful guide to frame your management. From a paediatric perspective, NICE (The National Institute of Health and Care Excellence, UK) have a pathway specific for children.

If behavioural techniques don’t work and you need to move onto a pharmacological approach, NICE only advocates the use of IM lorazepam. The Royal Children’s Hospital in Melbourne’s ‘Acute Behavioural Disturbance: Acute Management’ CPG has an escalation ladder from behavioural management, to oral, then IM / IV medications, clearly stating antipsychotics should only be given to children who have previously taken antipsychotics or who have a normal ECG. Please read it in conjunction with the RCEM guideline to understand the risks of each drug.

4. Think about the approaches to managing fever

We love a bit of paracetamol or ibuprofen to bring down a fever. But do you know how they work? Although paracetamol’s been used for over 100 years, we’re still not entirely sure how it works. Its antipyretic actions are thought to be due to inhibition of prostaglandin synthesis, resulting in a reset of the temperature centre in the hypothalamus. Non-steroidals, such as ibuprofen, also inhibit prostaglandin production, although via a different cyclooxygenase (COX) pathway (all sounding vaguely familiar?).

However, fever caused by toxins is not caused by prostaglandin or COX inhibition and needs a different approach to resolve.

Start with non-pharmacological measures. Fans, ice packs in the groins and axillae, ice baths and internal techniques such as gastric and bladder cold fluid lavage, or, more invasively, Intravascular Heat Exchange Catheters (the ICY Catheter). The ICY catheter is placed in the inferior vena cava via the femoral vein, acting as an extracorporeal cooling device. Cold saline circulates through the catheter, which is closed, so it does not infuse saline into the bloodstream, instead returning the now-warmed saline out of the body.

The patient’s core temperature is measured via a thermometer in the bladder. An automated feedback loop between the thermometer and the ICY Catheter ensures that the treating clinician can adjust the patient’s temperature to a target range. Add benzodiazepines to prevent shivering and for sedation to help the child or young person tolerate these techniques.

There’s extremely high mortality in severe hyperthermia – if these measures don’t work, then RSI with muscle paralysis (but avoiding suxamethonium), with benzodiazepine infusions.

And reach for the antipyretic drugs. Dantrolene is frequently used in the management of anaesthetic-induced malignant hyperthermia and neuroleptic malignant syndrome. It works as a postsynaptic muscle relaxant, inhibiting calcium ion release and, therefore, decreasing the amount of excitation-contraction coupling from muscle cells.

It’s usually found in theatres to keep it ready to be handed to the patient for the treatment of malignant hyperthermia. But, theatre is often far from the ED, and unless you know it’s there, it can take a while to hunt it down in the hospital – don’t let this delay you from using it emergently in the ED.

Although the use of Dantrolene in DNP toxicity is currently under debate with only a few case reports citing its efficacy in DNP toxicity, its use is still recommended to bring down temperatures above 39-40°C by Toxbase (the UK National Poisons Information Service) because of the high lethality of DNP.

Other options include Cyproheptadine, a first-generation antihistamine with additional anticholinergic properties and antagonist to serotonin, used in the treatment of serotonergic-driven hyperpyrexia (Serotonin Syndrome). To date, there are no case reports of cyproheptadine being used in DNP toxicity.

And don’t forget to monitor CK and renal function.

5. Consider your resuscitation fluid

You may have heard the phrase ‘(ab)normal saline’ before. Sure, one bolus with 0.9% saline is probably fine, but we should reach early for a balanced crystalloid like Hartmann’s or Plasmalyte, probably from the outset.

Francesca has pure metabolic acidosis and is trying to compensate by dropping her PaCO2. (Ab) Normal saline is 0.9% NaCl—that’s one chloride ion for every sodium ion. Chloride binds with hydrogen to form HCl, hydrochloric acid. Giving Francesca more acid in the form of chloride will plunge her pH lower. This will cause her to hyperventilate to compensate further, which will tire her out faster.

And then Francesca becomes hyperkalaemic. Worsening Francesca’s acidosis by giving more saline will only serve to make the hyperkalaemia worse for a number of reasons, the simplest one being that acidosis drives intracellular potassium to the extracellular (intravascular) space. ‘Why is that?’ you might wonder.

Remember, we use alkaline sodium bicarbonate to treat hyperkalaemia by driving potassium into the intracellular space. Giving acidic sodium chloride does the opposite: the hydrogen potassium pump exchanges extracellular hydrogen for intracellular potassium, pushing potassium out of the cell into the intravascular space. Giving acid, makes hyperkalaemia worse. Have a look at this Paediatric FOAM post, ‘Hartmanns in hyperkalaemia: Is that (O)K?’, for a more detailed account as to why we shouldn’t use saline in hyperkalaemic patients.

6. Have a strategy for your emergency treatment of hyperkalaemia

The treatment of life-threatening hyperkalaemia has three facets. All three are important but there is physiological and clinical  merit in doing these in order:

1) Membrane stabilisation

2) Shifting K+ into the cells

3) Reducing total body K+

The first two are the quick-fix solutions for the ED. The last solution involves potassium diuresis and haemodialysis or haemofiltration and will traditionally be dealt with in the renal unit or PICU – we will expand on these in a separate blog.

IV Calcium Gluconate

Calcium is vital for stabilising the myocardium. In life-threatening hyperkalemia, the primary concern is avoiding a lethal arrhythmia, so giving calcium first is a priority.

Initial dose: Assuming we have peripheral access the dose is 0.1-0.3 ml/kg IV calcium gluconate 10%  over 10 minutes, diluted fivefold to 20mg/ml. Aim for an ionised calcium >1.15  and repeat if required, remembering that a one-off dose will usually last between 30 minutes to an hour.

In the case of persistent arrhythmias or a particularly resistant hypocalcaemic state, further doses of calcium may be indicated, or an infusion can be considered (0.2ml/kg/hr of calcium gluconate 10% diluted as above).

Bicarbonate

It is important to understand that bicarbonate will only work in hyperkalaemia if the patient is in an acidotic state. In this context not all bicarbonate solutions have been created equal.  8.4% bicarbonate is very hypertonic, and a number of RCTs suggest that, if given neat, it will not work in reducing serum potassium levels in hyperkalemic patients. This is due to solvent drag; the hypertonic fluid drags potassium ions to the extracellular compartment due to an osmotic shift. This essentially neutralises the effect a neutral or alkali pH has in the direction of movement of the K+ ions, making the overall net shift minimal.

On the other hand, isotonic bicarbonate works in patients in an acidotic hyperkalemic state. Although isotonic bicarbonate isn’t commercially available in most UK-based hospitals, it can be made by diluting each millilitre of 8.4% sodium bicarbonate with 4.6 ml of sterile water for injection or 5% dextrose. 

A 1.5% solution of sodium bicarbonate is approximately isotonic. Isotonic bicarbonate can rapidly improve hyperkalemia if the patient is acidotic: a) by shifting potassium into the intracellular compartment, b) by increasing potassium diuresis due to alkalosis and c) due to a dilutional effect.  1mmol/kg of isotonic bicarbonate can be given to alkalinise the pH and cause a K+ shift.

Insulin

Insulin shifts potassium into cells by stimulating the activity of the Na^{+—H+ channel on cell membranes. This, in turn, promotes the entry of sodium into cells, which leads to the activation of the Na+—K+ ATPase, causing an influx of potassium. The decline in serum potassium levels by insulin is dose-dependent}. Due care must be taken to avoid hypoglycaemia, especially in infants and children with nephropathies. 

The doses of IV insulin are as follows:

Neonates: 0.3 – 0.6 units/kg/hour

Children > 1 month: 0.05 – 0.2 units/kg/hour 

Run with glucose 0.5 – 1 g/kg/hour (5-10 ml/kg of glucose 10% via peripheral administration)

Salbutamol

Salbutamol causes a small shift of potassium into cells but a high dose is needed for an adequate effect, around 10-20mg on average. This equates to 4 to 8 back-to-back nebulised doses, depending on the patient’s age. Salbutamol use comes with a caution, however; it can both worsen a pre-existing acidosis by driving up lactate (essentially having a neutral effect on potassium clearance) and will also cause tachycardia, and in patients prone to arrhythmias, it can cause SVT or even VF. It should not be a first-line treatment, and certainly not before the membrane has been stabilised with calcium or the pH has been less acidic.

7. And DNP?

DNP toxicity is a well-reported presentation to the ED, including a case report of a fatality in a teenage girl using it as a weight-loss drug. 

Features usually occur within 4 hours, with agitation, flushing, hyperthermia and diaphoresis. As with Francesca, there may be abdominal pain, vomiting and diarrhoea. There may be yellow discolouration to the skin and urine, which can be confused with jaundice, and rash and desquamation can be a feature, (mis)leading you down the path of toxic shock. The deterioration can be remarkably rapid with grossly elevated temperatures, heart rates and respiratory rates.

And the investigations? Metabolic acidosis secondary to raised lactate, methaemoglobinaemia, hyperkalaemia, hypocalcaemia, and hyperglycaemia (at least until glycogen stores become depleted, when the blood sugar will drop).

Read the letter to the editor in response to this case report, two case reports from the States, and a further report from London and decide whether you’ll be reaching for Dantrolene to treat DNP toxicity.

But let’s finish with a cautionary tale. Dantrolene can be hepatotoxic so monitor liver function closely. This case report describes a child who developed hepatitis after dantrolene at a pretty low dose.

We would LOVE your feedback about these DFTB PEM adventures, so if you can spare a minute, please complete our survey at www.tiny.cc/DFTBpemadventure or use your smartphone to let the QR code take you straight there. We timed ourselves completing it, and it took less than a minute. Thank you.

A huge thank you to Dr Laura Hunter, EM and Toxicology consultant at Guy’s and St Thomas’ NHS Foundation Trust in London, UK. Laura has a wicked sense of humor and an encyclopedic knowledge of all things toxicological. Thank you, Laura.

We are absolutely delighted to announce that our friend Costas Kanaris has joined the PEM adventures team, bringing with him his wisdom of critical care and general brilliance.

References