A woman was seen in ED with bleeding in early pregnancy.
Her ßHCG had fallen from 95,000 to 50,000 over 2 weeks.
On the basis of this she was told she had miscarried.
The next day she had an ultrasound which showed a viable pregnancy.
It is normal for ßHCG to fall in early to mid pregnancy.
In our shop if a woman presents with threatened miscarriage we will rule out a life threatening miscarriage or ectopic pregnancy:
No foul-smelling discharge or fever to suggest a septic miscarriage
Pain not localised to one side or the other
Not bleeding lots – eg bleeding able to be controlled by hourly changes of pads
Patient lives within say 30 minutes of hospital
If she passes all of the above we will usually discharge with advice to return if she is gushing blood, feeling faint or has severe pain. Give paracetamol and a weak to moderate opioid for pain (not a NSAID in early pregnancy)
She will be asked to return in the morning to Early Pregnancy Clinic (EPC)
Our EPC and ultrasound doesn’t run on the weekends so often one of the ED senior docs will do a bedside ultrasound to try to confirm an intrauterine pregnancy. If we can’t we might get the woman to return to ED or EPC in 2 days for repeat ßHCG and repeat scan.
One day you may find yourself working somewhere with a defibrillator that uses paddles instead of adhesive pads, because the health system of that country cannot afford the cost of single-use defibrillator pads.
You may also find that your team takes an awfully long time to first shock because they are mucking around trying to read the rhythm through the monitor or defibrillator leads, rather than through the paddles.
So key points to using an old school defib:
You need to familiarise yourself with the machine(s) in use in your setting.
All defibrillators should default to reading through the paddles or pads. If not you will need to change the lead to “paddles” or “pads” on the defibrillator
Some defibrillators have a single button (the yellow button in the photo above) which charges the defibrillator. Some defibrillators are charged by depressing the same two buttons used to deliver the shock. Alternatively you can ask someone else to push the “charge” button on the defibrillator machine.
Put gel on the paddles.
Put the paddles on the patient’s chest.
Charge the defibrillator as soon as the paddles are on the chest.
While charging read the rhythm (via the paddles) on the defibrillator screen.
Stop CPR briefly while you read the rhythm.
If it is a shockable rhythm, deliver the shock as soon as charging is complete.
Recommence CPR as soon as the shock is delivered.
Here is a video of some Fijian doctors resuscitating our old friend Annie with an old-school defibrillator with paddles.
By getting them to change to reading the rhythm through the paddles rather than via the leads, time to defibrillation was reduced from about 2.5 minutes to 40 seconds.
At 0300 the next morning some of the doctors were able to put this into practice and got ROSC (return of spontaneous circulation) on the second shock.
A 30-year-old indigenous Fijian male presented to a Fijian emergency department with several days of productive cough and lethargy. He had normal vital signs and had decreased air entry and dullness to percussion at the right base.
This is his CXR
I thought this was a right lower lobe pneumonia and would have treated his as such.
One of the good local docs, Vivek, put an ultrasound on the patient’s chest and found the real cause.
He then went on to CT. This is a video of the CT captured by phone (this is known as “mobile PACS” (picture archiving and communication system): there are no computers in that ED for viewing images, clinicians go to radiology and take images on their phones and take them back to ED to share with the team).
This is a massive liver abscess with a small pleural effusion. The productive cough was a red herring. Liver abscess in Fiji are usually seen in young adult males and are believed to be due to amoebic infection from contaminated water used to prepare kava. Kava is prepared from a local plant appears to have anaesthetic properties causing tingling of the lips and tongue and sedation. It tastes like mud and probably should be tried once to keep your hosts happy.
A couple of times a week in this ED young men were diagnosed with liver abscesses. They usually presented due to jaundice (this patient was not jaundiced) +/- mild right upper quadrant pain +/- fever.
Patients are treated with antibiotics eg oral metronidazole. Ultrasound is used to see if the abscess is fluctuant (ie changes shape with pressure). If it is fluctuant the abscess is drained percutaneously. If the abscess is not fluctuant the patient is treated with antibiotics for a few days until the abscess becomes fluctuant.
UpToDate says aspiration or drainage of amoebic liver abscesses is not usually required, but drainage of pleural effusions is recommended. I don’t know why the practice in Fiji differs from this. Maybe someone from Fiji can let us know in the comments. Perhaps they have some with bacterial superinfection. UpToDate recommends subsequently treating with paromomycin to eliminate intraluminal (within the gut) cysts. UpToDate also reminds us that oral metronidazole has excellent bioavailability and there is little to be gained by IV treatment if the patient is able to take oral meds – and oral metronidazole is much cheaper.
So keep liver pathology in mind with a patient with right lower zone dullness / opacity, especially if they have recently travelled to a Pacific Island.
Approach to the CXR
When viewing a right lower zone opacity our radiologist recommends considering:
above the diaphragm eg pneumonia, pleural effusion, subpulmonary effusion
the diaphragm itself eg eventration, phrenic nerve palsy
below the diaphragm: eg hepatomegaly, liver abscess, subprenic abscess.
This CXR does not particularly suggest something subdiaphragmatic is going on, our radiologist had his money on a subpulmonary pleural effusion.
Always try to find the cause of agitation and talk the patient down if possible.
Call for security or police early. Better to over-call than under-call.
Sometimes it is good just to step back and wait. Sometimes you need to intervene early before things escalate further.
You have a responsibility to restrain / sedate unwell or injured patients who may be a risk to themselves or others.
Ketamine is a great sedative but it is not generally accepted for behaviour control yet, especially by unsupervised junior doctors. If you are in a small hospital with no seniors on site – phone a senior for advice, if time allows. Once you have given someone ketamine you are committed to finding the cause of the agitation and to keeping them sedated, usually overnight, until the precipitant has worn off. You don’t want them waking up from a bad ketamine trip on your shift.
Haloperidol is a good sedative and doesn’t cause airway or respiratory depression. This is particularly useful when the patient has a large quantity of airway depressant on board eg alcohol ie most aggressive people in Australasian EDs
Benzodiazepines are good sedatives but may cause loss of airway control and respiratory depression especially for patients with lots of alcohol onboard. Benzodiazepines are the drug of choice for alcohol withdrawal and for agitation due to stimulants such as methamphetamine.
Often we use haloperidol and midazolam – probably without good reason
If you need to take a patient down get as many people to help as possible – preferably 5 + one ready with sedative medication. Failing that a rugby tackle may work.
Don’t restrain patients face down – they may suffocate.
If you have given the patient lots of sedative put them in the recovery position and monitor their oxygen saturation continuously. They will need one-on-one nursing until you are happy the patient can safely maintain their own airway.
If using physical restraints (eg leather wrist and ankle straps) use chemical sedation as well to ensure the patient isn’t straining against the restraints (which could make them more agitated and may rarely lead to rhabdomyolysis). In many hospitals use of physical restraint needs to be recorded on particular forms and obs documented regularly.
It is acceptable to leave physical restraints on a patient overnight if you consider them to be high risk.
RSI is not good for first line behaviour control. Sedate the person then, if needed to a “delayed sequence intubation” or DSI
Sometimes we have to restrain and sedate the crap out of aggressive patients.
In some hospitals we have little or no security assistance and doctors will be expected to take control of an aggressive patient.
The question is often asked: can we sedate or restrain a patient against there will if they don’t have a psychiatric problem? The answer is usually yes.
If you believe the patient may have a medical problem that’s putting themselves or others at risk, be it intoxication, head injury, sepsis, hypoxia or psychiatric disorder, you have a responsibility, also called a “duty of care”, to do what is needed to keep the patient and/or others safe.
If they are bad rather than sick or mad, you need to get police or security to deal with them. If in doubt assume the patient is unwell.
A gang member in his 50s, 160kg of lard and muscle, was in ICU on BiPAP for an exacerabation of COPD. I was called into help at 6am. He was in the ED waiting room when I arrived. We tried reasoning with him to no avail. He was mildly agitated and sick of being in hospital. The police sent one 60kg unarmed officer to help.
Another gang member arrived to collect the and we elected to allow him to leave – for our safety and so as not to inflame the situation. We asked the associate to try and convince the patient to return.
The patient and associate returned 20 minutes later as the patient had become more short of breath. He tolerated BiPAP and an IV line and we transferred him to ICU.
He was seen by the duty anaesthetist who wanted to place an arterial line but quickly backed off when the patient became stroppy.
He was given a small prophylactic dose of 3mg haloperidol. A short time later he became agitated again, ripped off his BiPAP and was thrashing around and being quite scary. His associate tried to calm him down. We kept our distance and waited.
About 30 minutes later the patient was asleep or unconscious. I snuck into his room and gave him a massive dose of IV ketamine. I would normally give a patient 1mg/kg of IV ketamine to completely flatten them. I gave this 400mg of ketamine. I wanted to be sure he was dissociated quickly and completely. He quickly woke up and looked like a sterotypical zombie: arms stretched out in front of him staring into space and looking confused. We were able to get the BiPAP on him. 5 minutes later he was still moving. I gave him another 200mg of IV ketamine and started a ketamine infusion at 200mg an hour, 5 mg IV haloperidol. He settled briefly. 5 minutes later he was getting agitated again and I gave him 5mg of midazolam. Then he finally stopped moving and tolerated the BiPAP
We converted him slowly to fentanyl for sedation. Fentanyl and/or propofol are probably the best medium term sedatives once you have the patient under control. There is less chance the patient will wake up delerious than with ketamine or even benzodiazepines.
The patient needed to be heavily sedated for behaviour control and probably needed to be intubated as he was going to need ventilatory support for days. The patient was in ICU so I needed to run this past an anaesthetist. I spoke to the anaesthetist who had seen the patient earlier who suggested we needed to wake the patient up and have that discussion with the patient. I invited him to have that discussion with the patient and went home to bed.
When I arrived for my shift a few hours later the patient was intubated.
I saw the pateint a few days later. He recognised me, thanked me for caring for him. He had no recollection of a bad ketamine trip.
Bottom line: some patients need massive doses of anaesthetics and sedatives to control their agitation so they can be safely managed. For patients like this get senior help early.
Later that day a man in his late 20s was brought in by ambulance and police, drunk after putting his arm through a window. His father said the patient had only used alcohol. The patient was handcuffed, agitated and abusive and bleeding quite a bit from a forearm laceration. A CAT tourniquet was applied. The patient was given 10mg IM haloperidol and 5mg IM midazolam with little effect. 10 minutes later he was given another 10mg IM haloperidol and settled. He was kept on oximetry without supplemental oxygen. He was restrained with leather wrist and ankle restraints. IV access was obtained and further boluses of 10mg haloperidol or 2.5 midazolam were given as required. A nasophyaryngeal airway was inserted. His wound was loosely sutured then dressed to control the bleeding. He was kept sedated and restrained overnight and then taken to theatre for definitive management the next day.
Case two could have been controlled with ketamine but I think it would have been overkill. Case one needed to be immobilised completely, very quickly. Partial sedation may have left him still able to inflict serious harm to staff. I wanted an agent that worked quickly, totally immobilised him but wouldn’t stop him breathing. Ketamine was ideal.
A chap in his late 30s was in ICU with a delirium of unknown cause: toxicology screen, CT, LP, bloods were normal. Formal toxicology and a brain MRI was pending. He heard someone talking about the psychiatric unit and thought he was going to be sent there and went ballistic. He hit 3 staff, made many holes in the walls and smashed a window with a drip pole. Luckily the was a visitor in the unit his 50s who had been a rugby player in his younger days. He did what was reported as a stunning tackle on the patient.
A bit like this one by one of my daughter’s friends:
I arrived a minute later. The patient was being restrained by the visitor and a couple a male nurse. The patient still had an IV line in place. He was given 10mg of IV haloperidol, 5mg of IV midazolam which settled him well. He was bound with leather restraints and charted PRN haloperidol and midazolam and a security guard was assigned to watch him.
A 93-year-old was in an orthopaedic ward post hip replacement. He was delirious and whacking everything and everybody in site with his crutches.
We formed a 5 person team. A sixth person was standing by with 3mg of IM haloperidol. I was at the point of the chevron shaped formation, armed with a pillow to take the blows from the crutches. The person on my left and right were assigned to an upper limb each, the people behind them were assigned to a lower limb each. We heroically over powered the old gentleman and lowered him onto his bed. The people on upper limbs grab a shoulder and a wrist. The people on lower limbs hold a knee and ankle. People are best held down on their backs as they have less purchase. Avoid face down due to risk of suffocation. If the patient is given lots of chemical sedatives it is best to have the patient in the recovery position if possible. The lead person quickly changes position and restrains the head to prevent biting.
We then did a thorough assessment to see if there was a cause for his agitation: hypoxia, fever, fluid overload inadequate analgesia, excess analgesia, full bladder. Often it is just post operative delirium.
Meningococcal infections are something we don’t see often but we need to know about. There was an outbreak in Suva, Fiji while I was there.
They may present with meningococcal sepsis, meningitis or both. They may or may not have a rash. They may or may not have neck stiffness. They may or may not look sick.
When I was a young doc I almost sent a child home because who had a fever and purpura, but had no neck stiffness and was walking around looking fine. Luckily the paediatrician I consulted decided to come in and see the child himself.
Papura can be easy to miss, especially in dark-skinned people: looks at the palms and soles for pupura and look at the conjunctivae and in the mouth, especially on the soft palate, for petechiae.
I treated one of the Suva patients myself. She was a 22-year-old student from Vanuatu who had mild fever and myalgias the day before. On the day of presentation she was found confused and hot and brought to hospital. Her temperature was 38.5, HR 130, no palpable pulse, no BP, no sats reading, her neck was rigid and she had widespread purpura – but this was missed by a good local doctor. She was alert but confused. She did not vocalise or responding to questions.
She was given IV penicillin while we waited for ceftriaxone to arrive from pharmacy (consultant order required!). Cefotaxime is the emperic therapy in some centres. Neonates required different antibiotics (eg amoxycillin and cefotaxime) to cover maternal vaginal flora and those over 50 usually have ampicillin or amoxycillin added to cover Listeria. Where there is a high incidence of peniciilin resistant penumococcus vancomycin is usually added to the emperic antibitoics.
After 2L of IV fluid the local doc ultrasounded her: her IVC was not changing with respiration but her cardiac contractility was reduced on bedside echo. She was started on an adrenaline (1mg in 1L of saline, running freely) while a noradrenaline infusion was prepared. The noradrenaline quickly went up to 20mcg/minutes and the adrenaline wasn’t slowed for several hours.
She was also given intravenous steroids. Hydrocortisone 50mg. Starship Children’s Hospital recommends Dexamethasone 0.15 mg / kg 6 hourly for 2 days for children with presumed bacterial meningitis.
After about 4.5L of fluid she became tachypnoic and agitated – I thought we were going to have to ventilate – but there was no ventilator in ICU and no ventilator in ED and no ketamine. Fluids were slowed right down and luckily she settled.
A central line was placed.
Within 3 hours of arrival she was sitting up talking and saying thank you.
She was transferred to ICU.
She was transferred out of ICU approximately 24 hours after her arrival in hospital with no sequelae.
Last I heard there were 3 other meningococcal cases in the 6 bed ICU.
Typically this is a disease of the young – preschoolers and young adults, but my mother-in-law had it in her 60s. She was found confused at home, and had no fever, no rash.
In New Zealand we do blood cultures and meningococcal PCR to confirm the diagnosis. LP in this setting is controversial. There is little utility to doing a lumbar puncture and there are concerns that the patient could “cone” if they have raised ICP from the meningitis. Starship Children’s Hospital says that an LP is contraindicated in a child with meningitis who is “sick or has a rapidly evolving rash” or “haemodynamically unstable” or “GCS < 13″.
CSF can be sterilised within an hour of giving IV antibiotics, so by the time your patient has had their CT to rule out raised ICP it is not that useful to do a LP. Administration of antibiotics should not be delayed till after the LP.