All roads lead to Rome

Except for the submarine route leading to Stockholms Skärgård. And so was the story of a kind 80-yo lady with unique attributes presenting to her family physician for a routine check-up. She had a past medical history of asthma treated sporadically with bronchodilators. Upon cardiac auscultation a pan-systolic and diastolic murmur was audible along the left sternal border. Severe pitting edema of the lower extremities was present and in spite of a flat abdomen, the liver edge could be palpated well beyond its normal location. Lungs were clean as a whistle. She was referred for an ultrasound, and this was where I first had the joy of making her acquaintance.

As I walked into the examination room I saw the patient with her back turned towards me. I greeted her but received no response. When I placed my hand on her shoulder asking if she could lay down on the bed, she jumped, produced an ear-to-ear smile and turned red as a rocket. I was stunned at the behavioral constellation. She had quite the severe hearing problem and throughout our dialogue kept flushing red whenever I said something she didn’t catch.

Parasternal long-axis seems all right. How about the four-chamber view? Let’s have a closer look at the tricuspid valve.

Four-chamber view and parasternal long-axis tricuspid valve view. PLAX TV-view can be put into focus by directing the ultrasound waves caudally and towards the right hip. The tricuspid valve appears frozen in time. With an echolucent glaze it produces a severe tricuspid regurgitation. Remember the standard colordoppler setting BART: blue away, red towards.

Parasternal long axis with alternate pulmonary valve view appreciated by leaning the probe in a cranial- and left axillary direction. Same thing here: pulmonary valve is thickened and retracted. The result is a severe pulmonary regurgitation and mild pulmonary stenosis.

The subcostal view through the liver can clearly identify how parts of right ventricular endocardium is fibrotic. The right heart works like an open conduit pumping blood both ways while the left heart is largely unaffected. These are rare findings but almost pathognomonic for carcinoid heart disease. Carcinoid tumors, known as neuroendocrine tumors (NET), are found in the small intestine distal ileum and appendix nine out of 10 times, causing carcinoid heart disease in about half of them. The tumor secretes catecholamines and serotonin. Interestingly these biogenic amines are metabolized and largely inactivated in the liver, lung and brain. Hence, an effected right heart would signify metastasis to the liver. When the left heart is effected one must rule out primary tumor of to the lung, metastasis to the lung, right to left heart shunt or very high tumor activity. The latter was induced iatrogenically back in the 70ies with the drug combo fen-phen (Fenfluramine-Phentermine), used as an appetite suppressants for the treatment of obesity. Our lady had the age to match but I’m not sure how vigorously these drugs were prescribed in Sweden. In addition the left-sided valves looked good. She was referred for a computed tomography.

There seems to be an unusual flame-like contrast-load in the liver. A tumor-mass is present in the lower abdomen and can also be visualized by ultrasound.

Lab-tests for NET were ordered including chromogranin, tachykinin, urinary 5-HIAA, showing very high levels of 5-HIAA. This was a clear-cut case, or so it seemed. When I presented the news of right heart failure and suspected metastatic cancer to my patient she wasn’t alarmed. With age and resulting enlightenment already taken into consideration, I still expected at least a flinch. I kept huffing and puffing and nothing happened. With a jolly old smile on her face, intermittently flashing red, she strolled on happy throughout our conversation. Could it be that she’d become habituated to a constant surge of serotonin? This evokes the question of how her quality of life would be affected if serotonin levels suddenly dropped.

Even though malignant, neuroendocrine tumors tend to have a slow progression. When carcinoid heart disease is present, heart failure generally tends to be the cause of death and not the metastasized cancer. Mainstay of treatment is somatostatin analogue therapy and tumor surgery. Some cases are be eligible for valve replacement in spite of liver metastasis. In the end our patient decided against the valve surgery, but for the removal of the primary tumor. PET-scan and magnetic resonance imaging was performed priorly.

During exploratory laparotomy, the PET-scan finding was confirmed. The tumor was connected to the right ovary. Postoperatively somatostatin analogues were withdrawn and urine-5HIAA stayed normal signifying absent tumor activity! The carcinoid tumor was a primary gonadal tumor and not intestinal! The blood was draining to the right ovarian vein, connecting to the inferior vena cava, bypassing the liver and producing right-sided carcinoid heart disease. Such a case has only been described anecdotally in the literature! Let’s look at the blood flow through the liver.

In the subcostal view we can see the liver, inferior vena cava and severely dilated hepatic veins. The venous blood, identified by the colordoppler, is pushed back and forth through the IVC and hepatic veins due to the failing tricuspid valve. Very large amounts of blood are regurgitated and produce severe liver congestion. This might help us understand how the CT-findings looked so magnificent, when in fact, there was no metastasis.

1.) Bhattacharyya S, Raja SG, Toumpanakis C, Caplin ME, Dreyfus GD, Davar J. Outcomes, risks and complications of cardiac surgery for carcinoid heart disease. Eur J Cardiothorac Surg (2011) 40 (1): 168-172.
2.) Chaowalit N, Connolly HM, Schaff HV, Webb MJ, Pellikka PA.
Carcinoid heart disease associated with primary ovarian carcinoid tumor. Am J Cardiol. 2004 May 15;93(10):1314-5.
3.) Aggeli C, Felekos I, Kazazaki C, Giannopoulos D, Kartalis A, Pitsavos C, Stefanadis C. Echocardiographic imaging of tricuspid and pulmonary valve abnormalities in primary ovarian carcinoid tumor. Cardiovasc Ultrasound. 2010; 8: 37.
4.) Bhattacharyya S, Davar J, Dreyfus G, Caplin ME. Carcinoid Heart Disease. Circulation. 2007; 116: 2860-2865.
5.) Daigeler A, Imoberdorf R, Haller A. Carcinoid tumor with carcinoid heart disease: the rare case of Hedinger syndrome without hepatic metastases. Int J Cardiol. 2003 Dec;92(2-3):295-6.
6.) Simula DV, Edwards WD, Tazelaar HD, Connolly HM, Schaff HV. Surgical pathology of carcinoid heart disease: a study of 139 valves from 75 patients spanning 20 years. Mayo Clin Proc. 2002 Feb;77(2):139-47.
7.) Dumoulein M, Verslype C, van Cutsem E, Meuris B, Herijgers P, Flameng W, Herregods MC. Carcinoid heart disease: case and literature review. Acta Cardiol. 2010 Apr;65(2):261-4.
8.) Lagoudianakis EE, Markogiannakis H, Karantzikos G, Papadima A, Alevizos L, Manouras A. Primary insular carcinoid of the ovary. Eur J Gynaecol Oncol. 2008;29(5):554-5.

A stumble may prevent a fall

A 50-yo man presented to the ED confused complaining of vomiting. Time of onset and frequency couldn’t be determined with certainty, but he noted worsening of symptoms two weeks ago. Admission lab-tests showed S-Na 109mmol/l, CRP 28 and WBC 9,5. Past medical history was significant for laparoscopic appendectomy and cholecystectomy two years previously. He was currently being treated with sertralin, a SSRI, due to depression.

Vital signs were stable except for a sinus tachycardia of 115 BPM. The patient noted he’d tried to stay hydrated drinking water. Serum-osmolality was 233 mosmol/kg (norm 285-295) and urine-osmolality 325 mosmol/kg (norm >800). The kidneys weren’t making a case to concentrate the urine very well and it seemed that he was diluted. This looked to be a clear-cut case of chronic hyponatremia secondary to SIADH, a known side-effect of SSRI. The patient was admitted and a slow IV-drip infusion of sodiumchlorid 0,9% started at the rate 80ml/h with a correction-goal of 8-10mmol /24hours (0,5mmol/hour).

Several hours had passed and I first saw the patient at the ward after he’d had a severe bout of vomiting. Perceived much older than his biological age, there was something out of the ordinary about him. With intermittent facial contractions my thoughts immediately went to tetany and hypocalcaemia. Positive Chvostek (YouTube) and Trousseau (YouTube) signs strenghtened my suspicions. This was most probably a severely dehydrated patient that demanded immediate IV fluid and electrolyte substitution. More about this relation later. His face reminded me of a raisin I’d leave in the box for last; his tongue was shrunk and hiding like a little turtles head; the eyes were sunken as if they’d seen to much. Abdominal peristaltics were reduced! Upon palpation there was slight tenderness, diffuse resistances, but no signs of peritonitis. There had hardly been any diuresis since admission! As the examination was being performed, the patient had another episode of vomitus and became dyspneic. His respiratory rate increased to 40/min, oxygen saturation and BP dropped to 75% and 90/60 respectively. Stabilization was initiated with oxygen and IV bolus of Ringer Acetate. An arterial blood gas was drawn. From nowhere an ultrasound probe appeared in my hand. I tripped onto the patient landing the probe onto his subcostal window.

Subcostal view trying to identify a collapsed inferior vena cava. Even though evident by the physical examination, this just reinforces how severely dehydrated the patient was. What was I to do but apply some more mixed sonographic arts.

Right lateral lung- and abdominal view is significant for lung consolidations. Due to the larger caliber and more vertical orientation of the right mainstem bronchus, there is a high suspicion of aspiration and aspiration pneumonitis/pneumonia.

Ultrasound probe dropped straight onto the abdomen. All abdominal views could identify the same pathologic state. Small bowel loops are dilated beyond 2,5cm and hence there is a high suspicion of ileus. Additional IV crystalloids were administered. A nasogastric tube was placed draining about 750ml of light-brown fluid.


Even though the patient didn’t complain of severe abdominal pain, the computed tomography raised a high suspicion of mechanical small intestinal ileus. The contrast didn’t reach past the distal ileum. In effect the patient was taken to surgery and an exploratory laparotomy performed, identifying an adherence/bride at the distal ileum as the culprit obstruction.


Hypochloremic metabolic alkalosis partly compensated by elevated PCO2. A patient with persistent vomiting may loose in excess of 2 liters of HCl-rich fluid per day. The fluid contains a high level of protons at a concentration of about 100mmol/L. For every H+ secreted a bicarbonate molecule (HCO3-) is generated. In effect the body must rid itself of 200mmol of HCO3- through the kidneys. Initial this works, but as the vomiting persist intravascular volume is drained and glomerular filtration rate is decreased, hence resulting in elevated HCO3- levels and pH. In addition to losing sodium through the GI-tract, maintainance of steady-state relationship between chloride and sodium transmembrane electrochemical potential differences is kept by natriuresis. Chloride is reduced, sodium follows. GFR is reduced, HCO3- and pH is elevated. All is maintained by protracted vomiting! And last but not least, how is the vomiting related to tetany? As pH is elevated, albumin develops a higher affinity for calcium. Hence free calcium-ion levels are lowered.
In conclusion, despite the present medical history being a rare mineral called unobtanium, the case could have been partly solved by taking a good look at the blood-gas. Fortunately, ultrasound was available bedside, reinforcing suspicions and saving yet another day.

1. Small bowel obstruction. Introduction to Bedside Ultrasound: Volume 2 by Matthew Dawson och Mike Mallin (free at Itunes)

Call the plumber!

Life is different on the countryside. The hospital lies just up a hill surrounded by forest. A couple of wards, a small but handy emergency department, and a few operating theaters keep things running. Chief complaints includes “Hit self with axe”, “Tractor accident/MVA” – type of injuries in addition to the usual bread and butter of the ED.

A 63-year old male was brought in by ambulance complaining of left-sided groin pain. The paramedics suggested a short episode of hypotension which had resolved by the time of arrival to the hospital. Apart from Barrett’s esophagus, the patient had a clean bill of health. He was a farmer by occupation working at a field overseeing the countryside.

The pain was described as acute in onset, severe, sustained, and neither alleviated nor exacerbated by motion. No radiation was present, but the pain had partly migrated to the left foot. When asked explicitly, an episode of severe chest pain had been present in the ambulance. Even though milder, it was in fact still present at the moment of examination!

Vital signs: HR 98, BP 120/70, spO2 94% on room air. Respiratory rate 24. Temp 37,1
General: In mild distress, reports moderate pain. GCS 15.
Heart: No murmur, abnormal loud S1, S2.
Lungs: Normal breathing sounds, no rales/rhonchi.
Abd: Soft but tender in left lower quadrant.
Peripheral pulses: R/L radial pulses present, L pedal pulse absent.

With a history of multifocal, severe pain, as in this case chest, groin and left leg, what is one to suspect? What is the likely diagnosis and what would you do next?

Aortic dissection (AD), the haunting diagnosis that keeps us up at night, must be suspected upon the features of severe pain from several body compartments and pulse abnormalities (aka North South symptoms).

AD carries a high mortality which is highest during the first 48h, implying the need for a high index of suspicion and rapid diagnostic pathways. [1] Mortality in type A dissections increases by 1-2% for every hour of delayed intervention.


Pain that is severe and sudden in onset presents in about 80% of AD cases. Pulse abnormalities often sought for, are estimated to be present in only 30% of patients making it an unreliable diagnostic tool. [2]

Cardiac complications often accompany a proximal aortic dissection (Type A by Stanford classification). Evaluation must be done for the presence of severe aortic regurgitation, pericardial hemorrhage, tamponade, and even myocardial infarction caused by coronary involvement. The right coronary artery is more frequently involved in such cases. [3]
Intimal flaps may be present and found throughout all of the aorta and its major branches, and hence produce a variety of symptoms from different body organs. Neurological presentations are not rare and may be present in 15-40% of cases, sometimes even producing spinal ischemia from occlusion of spinal vessels. Any severe pain accompanied by neurological deficits should raise the suspicion of AD. [2]

Under certain circumstances ultrasound may reveal the direct or indirect effects of this pathology. The sensitivity of detecting an AD with transthoracic ultrasound by an experienced sonographer may be as low as 77% and should not be used as a rule-out method, especially not in the bedside setting. [2] Gold standard is contrast-enhanced computed tomography.

D-dimer has been proposed as a rule out-test for AD. However, evidence suggests that sensitivity is too low for a lethal condition such as AD, especially when pretest probability is already high per history and physical examination findings. [4]

Our patient was taken immediately to CT as the treating physician had an high suspicion of AD.




CT aorta showed an extensive type A dissection from the aortic valve down to the left iliac artery with occlusion of the iliacs and subtotal occlusion the superior mesenteric artery. Blood pressure and pulse-rate should be monitored frequently. Generally, a reduction in afterload alone will produce a compensatory chronotropic boost, which could increase wall stress. Therefore it is recommended to use a mixed alpha/beta adrenergic antagonist, such as labetalol (Trandate). [1]

The patient was transported by ambulance to a tertiary care center where surgical repair was performed. Perioperatively, the aortic root was found to be severely affected with intimal rifts near both coronary vessels in addition to severe aortic regurgitation. A small amount of hemopericardium was noted. A biological aortic valve and composite graft of the ascending aorta was placed.

Thanks to J.A.K for contributing to the blog!

1. Nienaber C.A. Management of acute aortic dissection. Lancet. Feb 2015 28;385(9970):800-11. doi: 10.1016/S0140-6736(14)61005-9

2. 2014 ESC Guidelines on the diagnosis and treatment of aortic diseases Eur Heart J 2014;35:2873-2926 – Eur Heart J doi:10.1093/eurheartj/ehu281

3. Manning W.J. (2013). Clinical manifestations and diagnosis of aortic dissection. UpToDate. Retrieved from

4. Sutherland A, et al. D-dimer as the sole screening test for acute aortic dissection: a review of the literature. Ann Emerg Med. Oct 2008;52(4):339-343

Move to the beat

What do you think?



Have a closer look at V4. I. Keep. Dancing. On. My. Own. Could it be?

And one and two, and one and two. Pericardial effusion and electric alterans! Surely the heart seems to have found a beat within the beat. Cardiac depolarization hasn’t changed, just the angle at which the ECG machine records it. But is this an emergency cardiac tamponade? They say that cardiac tamponade is a clinical diagnosis. Let’s look at the tachycardic patient. Presenting with acute onset dyspnea and palpitations. Distended neck veins, hypotension and muffled heart-sound together form Beck’s triad. It is in fact cardiac tamponade! But remember the differential: tension pneumothorax, pulmonary embolism, cardiogenic shock, constrictive pericarditis. Even the astute clinician may feel obliged to reach for the ultrasound probe.

An overanxious subcostal view. As for treatment? In the non-traumatic tamponade, 500-1000mL IV bolus crystalloid would facilitate right heart filling and temporarily improve hemodynamics. Pericardiocentesis next. But the game isn’t always that easy. What if the patient gestalt and vital parameters were not past the threshold of an emergency? Let’s look at a similar patient who is less affected and with a blood pressure of 100/60. Large amounts of pleural fluid are present on the right. There is a high suspicion of malignancy per history.

Is this tamponade? Clinically the patient is not quite there. Recall that pericardial effusion volume is second to pericardial effusion rate in cardiac tamponade. As little as 150ml of pericardial effusion may cause tamponade if very rapid, while as much as 1000ml may not if produced over a long period of time. When the pressure in the pericardial sack exceeds the normal filling pressure of the right ventricle, the result is restricted filling of the RV and decreased cardiac output. When effusion volumes aren’t extensive and clinically we’re in doubt, fret not, Mr. Ultrasound will untangle the knot. As the pericardial pressure rises to the diastolic pressures of the heart, space becomes a rare commodity and the chambers become diastolically dependent on each other. This leads to a cascade of logically predictable events ominously ending in right ventricular diastolic collapse.

Systems seem up and running. Diastolic collapse of RA, LA, and some trampoline bouncing on the RV in systole. How about left ventricular output variation? A systolic BP drop of >10mmHg upon inspiration known as pulsus paradoxus can be estimated more exactly by measuring the mitral valve flow.

There is a variation of about 20% on the mitral ”double-mountain” flow chart (>25% is significant for tamponade). All in all one could refer to this as pre-tamponade. The BP soon dropped a bit more, and a subacute parasternal pericardiocentesis was performed.

Parsternal left view with a linear transducer. Apologies for the picture, but the elbows aren’t the best tools for hitting small buttons. At the 3 cm line we can see a ”black” anechoic horizontal layer representing the pericardial fluid. The moving shadow deeper past that line is part of the right ventricular free wall. The needle can be seen moving diagonally from the upper right torward the center and pericardial effusion. 320ml of dark-red fluid was removed, and patient symptoms and vitals improved significantly.

Until next time, keep on dancing.

P.S. Head on over to, check out Niklas Jonsson’s case on traumatic cardiac tamponade and be sure to read the discussion.

1. Triage strategy for urgent management of cardiac tamponade: a position statement of the European Society of Cardiology Working Group on Myocardial and Pericardial Diseases. Ristić et al. Eur Heart J. 2014 Sep 7 (pdf)
2. Pericardial effusion. Introduction to Bedside Ultrasound: Volume 1 by Matthew Dawson och Mike Mallin (free at Itunes)
3. Pericardiocentesis. Introduction to Bedside Ultrasound: Volume 2 by Matthew Dawson och Mike Mallin (free at Itunes)
4. Ultrasound Guided Pericardiocentesis Microcast! @

Ill weed grows fast

Just when I thought the sun would set behind the mountain, it rose and made yet another run across the valley. Time to step out of the twilight zone and get back on the donkey. Three hours away from the closest hospital and with an ultrasound probe in hand, I made another run for it.

A peculiar case this middle aged man was, bearing a captivating odor of fermented fish. Priorly morbidly obese with OSAS and CPAP night-time, he’d undergone laparoscopic gastric bypass a year previously, lost massive amounts of weight and was now carrying his skin like a half-empty sack of potatoes. His main complaint was tiredness and impaired life-lust. Family and friends had noticed his indifference. For the past months he’d slept most of the days, experienced muscle- and joint aches and some dizziness. I wasn’t immediately alarmed. Might as well have prescribed him antidepressants. I persisted and he answered question after question.

His apathy had almost rubbed off on me when I heard “visual disturbance, night-sweats” and snapped back. Two days ago he had suddenly gone blind on one eye. Five minutes later he regained vision only to be followed by another similar episode. This was a clear-cut case of…
Vitals were fine and he was afebrile. CRP 110. No cutaneous changes. But upon cardiac auscultation a pan-systolic and diastolic murmur was discovered! And wouldn’t you know, he’d had periodontitis for the past 20 years and root canal fixed three months ago. So what was in the Chinese fortune cookie?

Infective endocarditis. A dancing vegetation is present on the aortic valve non-coronary cusp and anterior mitral leaflet. There is an aortic regurgitation. And the crown jewel? Fistula between the aortic root and left atrium. Anatomically the aortic- and mitral annuli are fused at the fibrous trigone predisposing any vegetation to travel between one another sometimes producing an apparent abscess or fistula.

He was immediately admitted to the hospital. Antibiotics were prescribed and the patient planned for subacute surgery due to high risk of embolization. Blood cultures grew Streptococcus mitis, an inhabitant of the oral cavity. The cause of amaurosis fugax proved to be what we expected unfortunately. MR brain showed two small infarcts caused by septic embolization. Neurologic function remained intact however. And like any other success story, christmas didn’t arrive a day too early. The diseased valves were replaced by two intracardiac Philippe Pateks working around the clock keeping his life ticking.

A liver in the lung

65-yo female with COPD presents to the ED severely distressed and dyspneic. Oxygen saturation 70%. Breathing frequency 45/min. Obstructive breath sounds with spread rhonchi. The patient is heavily dependent on accessory respiratory muscles. 120BPM. BP 100/70. Afebrile. COPD with acute exacerbation is suspected. Corticosteroids, bronchodilators and oxygen are administered. A bloodgas shows PCO2 retention and respiratory acidosis. Vitals slightly improved but oxygen saturation is not adequate. Higher levels of oxygen therapy will cause carbon dioxid narcosis. BIPAP is started with progressive improvement. Bedside ultrasound:

Left lateral thoracic- and upper abdominal view plus zoom. Significant for consolidations and hence a high suspicion of severe pneumonia/pneumosepsis. Fluids and antibiotics are administered. Bloodpanel returns showing Lactate 4,5, CRP 200 and WBC 18 with a left shift. A central venous catheter is placed.

The patient is taken to the ICU where a new bedside ultrasound is done. Consolidations are again visualized. In addition the following is seen.

What do you think? Please comment below. All intercostal spaces look the same. The pleura is statically prolapsed intercostally. My guess:the intercostal muscles have been through a marathon of strain. Currently flaccid and off-loaded by a well-deserved non-invasive ventilation. Additionally we see a flattened diaphragm which signifies severe COPD/emphysema.

Three days into treatment inflammatory parameters and oxygen saturation have been improved and NIV discontinued.

Left lateralt thoracic- and upper abdominal view. Interestingly enough the consolidation has changed. What we see now is an air bronchogram/hepatization of the lung which simply means that the lung takes on the appearance of a liver. The bronchi appear white. Causes may vary, but in short it implies lung consolidation or atelectasis. In contrast to chest x-ray, ultrasound can divided air bronchograms into either static with no movement in the bronchi or to dynamic with movement in the bronchi (see youtube). The dynamic form has a very high specificity and positiv predictive value for pneumonia. In our case the air bronchogram is static but the clinical scenario implies an infectious process.

Furthermore the intercostal space and pleura seem to have normalized.

1. The dynamic air bronchogram. A lung ultrasound sign of alveolar consolidation ruling out atelectasis. 2009. Lichtenstein D, Mezière G, Seitz J.
2. Radiopaedia.orgAir bronchogram
3. Introduction to Bedside Ultrasound: Volume 1 by Matthew Dawson och Mike Mallin (free at Itunes)

Ultrasound saves lives! #2

A 55 yo woman lost consciousness at home. When the ambulance arrived she was pale, BP 70/40 with an irregular heart rhythm and ventricular frequency varying between 20 and 40 BPM. The ECG was sent through the “cloud” and the cardiologist on call ordered fluids and atropine. Upon arrival in the ED the patient was very ill, giving the impression of imminent cardiac failure. Diaphoretic, somnolent, BP 100/60 with an unchanged heart rate and frequency. The past medical history was significant for chronic kidney disease, pre-dialysis. Priority number one was to attach the defibrillator plates. Number two – an ECG.

Regular rhythm? P-waves present? QRS width? QTc? ST-segments and T-waves?

It seems to be some kind of a nodal escape rhythm. What about the ultrasound? UnFortunately you’ll have to use your fantasy on this one. I’ll be faster with the trigger finger next time. Imagine a gorgeous ejection fraction, the best one you’ve seen. The problem was that the ventricular rhythm was irregular, frequency low and successively getting lower by the minute. The impressions was that there was a failure of the SA-node, AV-node and other escape foci. Last swan dance.

Any electrolyte disorder that springs to mind?

The blodgas rightfully showed a potassium of 8, pH 7, creatinine 640 (6.8mg/dL), lactate 5.5, Base excess -20, pCO2 10.4, sodium 132. Not completely unexpected. Full measures were taken. Membrane-stabilizing calcium IV without any ECG improvements. Sodium bicarbonate, insulin/glucose and salbutamol nebulization followed. An isoprenalin-drip was started. The result?

Rising potassium and lactate. Dropping heart rate.

Lactic acidosis leads to rising extracellular potassium and vice versa. Like a dead fish turns its belly up, drifting away in loneliness, the mitral valve made a grand final agonal motion whereupon the gracious magic vanished. Seconds felt like minutes. A swirl of spontaneous contrast began making its presence. True ultrasound-witnessed asystole. What do the guidelines say? Compressions/ventilations/Adrenalin?

Based on the ultrasound and the appreciation of the hearts capabilities, the decision was made to try an alternative treatment modality. With continuous ultrasound observation, the external pacing function was started on the defibrillator. Note that part of the pacing-electricity was transmitted to the operator :) (Don’t try this at home kids)

Eureka! Time to pacing was of utmost importance. Any further delays and the race would have been lost. Followup blodgas showed improvement on all values. But why now? Most probably the cardiac output, previously deranged, now managed to deliver the drugs to the systemic circulation. The patient was dialyzed and had a temporary transvenous pacemaker implanted. She was discharged from the ICU after three days without any further complications. Neurological sequela? Nada.

Focus on the snake & miss the scorpion

47-year-old woman seeking due to acute / subacute onset dyspnea.
Past medical history significant for rectal cancer with metastasis to liver, lungs and brain. Previously hospitalized due to severe left sided pleural fluid.
Current vitalsigns: Sat 91% on 2L oxygen, p.90, BP 115/70. The journal shows a habitual BP of 140/70. Clinical examination reveals absent leftsided breath sounds and dull percussion.

Spine-sign, well-defined diaphragm and the absence of curtain-effect indicates a large amount of pleural fluid. A fluid gap is also seen apically. But was this the entire truth? One would expect a gradual onset dyspnea in this setting.

The right ventricle, located adjacent to the liver, seems overloaded basally. Inferior vena cava (IVC) seems plethoric (dilated with poor respiratory variation) signifying increased central venous pressure (CVP). What happens to the right ventricle if preload / returnvolume is increased? Ask the patient to take a deep breath. This generally also improves the subcostal image quality. Alternatively, try passive leg raise.

The basal part of the right ventricle is considerably dilated while the apex is hyperdynamic. Known as McConnell’s sign, it may indicate massive or submassive pulmonary embolism. But the question is if large amounts of unilateralt pleural fluid can induce respiratory variations in right ventricular load. A quick review of the literature comes back empty.
A left-sided pleural chest tube was set up and 1000cc of clear-red liquid drain after which the drainage was stopped. This was followed up by 500cc / 4hrs to avoid complications such as reexpansion-related pulmonary edema. The patient experienced immediate symptomatic relief and was admitted for further treatment. Sat 94%, p.80 and BP 120/70.
After a total of 2.5L fluid was drained, the patient noted a swelling of the lower extremities. Ultrasound showed bilateral DVT in common femoral veins. Pending thoracic CT according to LE protocol, a new cardiac echo was performed.

McConnell’s sign is accentuated by inspiration.
Central right-sided pulmonary embolism. In view of the brain metastasis, a reduced dose low molecular weight heparin was instituted BID.

One may ask if the CT is really necessary in this case. Bilateral DVT was already diagnosed and thus the indication for LMWH present. However, it was still deemed reasonable to perform the CT since the diagnosis was not quite clear and it may affect duration of treatment with anticoagulation.

Easy to get confused by bad TIP(S)

50-year-old man who a week earlier had a transjugular intrahepatic portosystemic shunt (TIPS) installed. Currently admitted due to altered mental status. The TIPS seems to be patent. The liver looks normal in size with a smooth margin. A small fluid gap is seen subcapsularly just below the diaphragm.

The portal vein, a gateway between two capillary beds, is formed by the union of veins from the gastrointestinal tract, spleen and pancreas. Disorders of drainage to, from, or in the liver may lead to portal vein hypertension which may result in varice formation, ascites, splenomegaly, and all of its complications. The most common cause of portal vein hypertension is liver cirrhosis and scartissueformation which result in partial blockage of blood flow passing through the liver. When the portal vein pressure goes up, compensatory shunts are formed from the capillaries connecting the hepatic veins, bypassing the liver first pass metabolism. Elevated systemic nitrogen levels ensue. Hepatic encephalopathy in this setting can occur due to increased nitrogen levels, electrolyte and acid-base disorders, infections, drugs-induced or due to worsening liver failure.

TIPS, a bridge between the portal vein and the hepatic veins, is an alternative therapy refractory portal vein hypertension, ascites and variceal bleeding. However, this increases the immediate postoperative risk of hepatic encephalopathy.

Pure Adrenalin Pure Fun

Vi trillade över en fascinerande studie häromdagen:

Att Adrenalin vid hjärtstopp ökar sannolikheten för ”Return Of Spontaneous Circulation (ROSC)” är välkänt. Huruvida det även resulterar i att fler patienter lämnar sjukhuset via huvudentrén istället för bårhuset är dock inte lika självklart. Franska Dumas och kollegor bestämde sig, förmodligen efter några glas rödvin, för att studera detta. Studien är en retrospektiv registerstudie med alla de brister som kommer med detta men slutsatsen är fascinerande:

”Pre-hospital use of epinephrine was consistently associated with a lower chance of survival, an association that showed a dose effect and persisted despite post-resuscitation interventions”[1]

Hur kan det vara att det läkemedel som är högst prioriterat vid hjärtstopp i själva verket leder till minskad överlevnad?

Vi tar det från början.


Adrenalin syntetiseras i binjurarna genom att ett enzym med det lite krångliga namnet Phenylethanolamine N-methyltransferase och som katalyserar methylering av noradrenalin till adrenalin. Enzymet som vi kan kalla PNMT aktiveras av en rad olika stimuli men i första hand av glukokortikoider. Adrenalin är en oselektiv katekolamin med starka α- och β-stimulerande effekter. Den i sammanhanget önskvärda α-stimulerande effekten leder till perifer vasokonstriktion med ökat diastoliskt och systoliskt blodtryck som följd. Vid hjärtstopp leder detta till ökat blodtryck i aorta vilket leder till ökat blodflöde i coronarkärlen i och med att coronarkärlen perfunderas i diastole. Den mindre önskvärda β -stimulerande effekten leder till ökad hjärtfrekvens och kontraktionskraft men ökad syrgaskonsumtion i myokardiet och ökad risk för myokardischemi.

Flertalet studier har mycket riktigt visat att adrenalin är kopplat till till kardiomyopati, post -arrest myokarddysfunktion, gastointestinal ischemi, lungödem, ARDS, akut njursvikt och laktacidos. Troligen har det även proarytmiska effekter.[2]

Doseringen av adrenalin vid hjärtstopp härstammar från djurstudier och fallrapporter på 60- och 70-talet där det observerades att intrakardiell administration av adrenalin var effektivt vid hjärtstopp. I början på 90-talet genomfördes ett antal försök att bestämma den optimala dosen, höga doser (upp till 15mg!) gavs med varierande resultat dock i samtliga fall utan effekt på överlevnad. American heart association listar totalt 8 studier med fler än 9000 patienter där man inte kunde se någon skillnad i överlevnad med hög jämfört med låg dos adrenalin. [3]

Det tveeggade svärdet!

Nu är det inte bara våra franska vänner som noterat en minskad överlevnad och ett dos-effekt-samband med adrenalin.

Redan 1994 noterade Rivers (!) et al på ett elegant sätt att adrenalin genom hemodynamiska effekter påverkar syrgastransport och syrgasextraktion i perifer vävnad.[4] Måhända ökar möjligheten till ROSC, resonerade man, men till priset av att det riskerar att öka både morbiditet och mortalitet. Man kunde även visa att dessa effekter uppvisar ett dos-respons samband. Och nej, den ackumulerade dosen adrenalin berodde inte på att patienterna befann sig i längre tid utan bärande cirkulation. Liknande resultat sågs i en annan studie där 4 mg jämfört med 1 mg adrenalin var associerat med försämrad neurologisk outcome även efter stratifiering för duration av hjärtstopp och confounders.[5] Att adrenalin har negativa effekter på cerebral microcirkulation är även visat i 2 ytterligare studier.[6],[7]

Äran och hjältarnas land träder in på banan!

Mikael Holmberg och kollegor bestämde sig, förmodligen efter en snaps eller två, att genomföra en registerstudie där man gick igenom samtliga hjärtstopp mellan 1990-1995. 60% av Sveriges ambulanser deltog. Av 14000 inkluderade hjärtstopp påbörjades HLR på knappt 11000. 42,4% av patienterna erhöll Adrenalin och 47,5% intuberades. Den något nedslående slutsatsen blev att både adrenalin och intubering var oberoende riskfaktorer för död.[8] Detta ligger i linje med en mer modern studie som visade att patienter som erhöll basal HLR hade högre överlevnad än de som erhöll avancerad HLR. [9]

Normännen kan mer än att borra olja och åka skidor

Så att adrenalin minskar mortalitet vid hjärtstopp är som vi sett långt ifrån självklart. Några norska, förmodligen skidåkande, kollegor genomförde 2011 en fascinerande studie.[10] Patienter med hjärtstopp utanför sjukhus med misstänkt kardiell orsak randomiserades till att antingen erhålla perifer venkateter (pvk) eller inte. Patienter som erhöll pvk fick sedvanlig behandling med adrenalin och Cordarone (och emellanåt även Atropin). Exklusionskriterier var hjärtstopp bevittnat av ambulanspersonal, hjärtstopp med misstänkt hypoxisk genes eller att resuscitering initierats av för studien externa personer. Studien var power-beräknad för att detektera en förmodad skillnad i mortalitet med 14% i pvk-gruppen jämfört med 7%  hos de som ej gavs pvk. Hypotesen var alltså att enbart sätta en pvk har ogynsamma effekter, dels genom eventuellt negativa effekter av läkemedlen själva, men även genom att det avleder fokus från kompressioner och defibrillering.



Notera att pvk-gruppen i andelen med asystoli/PEA har markant större andel ROSC (29 versus 11%) utan att detta genererar någon minskad mortalitet. Trots 274 patienter överlevde endast 2 sjukhusvården. I övrigt inga statistiskt signifikanta skillnader mellan grupperna, även om det ska sägas att tendensen generellt låg mot viss effekt för pvk-gruppen. (Studien har fått en del kritik för metodologiska brister som det går att läsa om här.[12])

Solen lyser starkare down under!  

Så hur kommer man vidare i detta uppenbarligen ganska snåriga fält? Drömmen vore ju att designa en studie där man dubbelblint randomiserade patienter till adrenalin alternativt placebo prehospitalt. Detta var exakt vad australiensiska Ian Jacobs och kollegor, förmodligen efter några Fosters, bestämde sig för att göra.[11] (Man kan ju tycka att detta borde vara något slags grundkrav innan man introducerar ett läkemedel, men men …)

601 patienter randomiserades till antingen placebo eller 1 mg adrenalin. 90% av hjärstoppen var förmodad kardiell etiologi och 50% erhöll HLR av person i omgivningen.

Hur gick det då?



Vad som är synd (oavsett vem man hejar på i sammanhanget) är att stackars Jacobs och kollegor av något oklara skäl tvingades avbryta sin studie i förtid. Av planerade 5000 patienter blev det inte mycket mer än 500. Men tendensen är liknande de skidåkande normännens – en tendens mot effekt som dock trots relativt stora kohorter inte är signifikant. Notera den låga överlevnaden jämfört med den norska studien.

Så vad ska man tro?

En aktuell review-artikel har gått igenom 9 studier av (minst sagt) varierande kvalitet.[12]
Slutsatsen är ett under av klarsynthet:

“There is currently insufficient evidence to support or reject its administration during resuscitation.”

Så vad ska man göra nästa gång man står där med Adrenalinsprutan i sin hand? Själv skulle jag hälla ut den i vasken. Om jag vågade.

(Eller kanske pröva hemodynamiskt styrd dosering av adrenalin som du kan läsa om här [13] och här [14] eller lyssna på en podcast om här [15])


För en betydligt mer stringent och extensiv genomgång av ämnet rekommenderas Smart EM:s enastående podcast:


[1] Dumas F, Bougouin W, Geri G, et al. Is Epinephrine During Cardiac Arrest Associated With Worse Outcomes in Resuscitated Patients?. J Am Coll Cardiol. 2014;64(22):2360-2367. 
[2] Reardon PM, Magee K. Epinephrine in out-of-hospital cardiac arrest: A critical review. World Journal of Emergency Medicine. 2013;4(2):85-91.
[3] ECC Guidelines. Part 6: Advanced Cardiovascular Life Support Section 6: Pharmacology II: Agents to Optimize Cardiac Output and Blood Pressure
[4] Rivers E, et al. The effect of the total cumulative epinephrine dose administered during human CPR on hemodynamic, oxygen transport, and utilization variables in the postresuscitation period. Chest. 1994;106:1499–1507.)
[5] Behringer W, Kittler H, Sterz F, Domanovits H, Schoerkhuber W, Holzer M, et al. Cumulative Epinephrine Dose during Cardiopulmonary Resuscitation and Neurologic Outcome. Ann Intern Med. 1998;129:450-456.
[6] Ristagno G, Sun S, Tang W, Castillo C, Weil MH. Effects of epinephrine and vasopressin on cerebral microcirculatory flows during and after cardiopulmonary resuscitation. Crit Care Med. 2007;35:2145–2149
[7] Ristagno G, Tang W, Huang L, Fymat A, Chang YT, Sun S, et al. Epinephrine reduces cerebral perfusion during cardiopulmonary resuscitation. Crit Care Med. 2009;37:1408–1415.
[8] Holmberg, Mikael et al. Low chance of survival among patients requiring adrenaline (epinephrine) or intubation after out-of-hospital cardiac arrest in Sweden. Resuscitation , Volume 54 , Issue 1 , 37 – 45
[9]Sanghavi P, Jena AB, Newhouse JP, Zaslavsky AM. Outcomes After Out-of-Hospital Cardiac Arrest Treated by Basic vs Advanced Life Support. JAMA Intern Med. 2015;175(2):196-204.
[10] Olasveengen TM, Sunde K, Brunborg C, et al. Intravenous drug administration during out-of-hospital cardiac arrest: a randomized trial. JAMA. 2009;302(20):2222- 2229.
[11] Jacobs IG, Finn JC, Jelinek GA, Oxer HF, Thompson PL. Effect of adrenaline on survival in out-of-hospital cardiac arrest: A randomised double-blind placebo-controlled trial. Resuscitation. 2011;82(9):1138-1143.
[12] Reardon PM, Magee K. Epinephrine in out-of-hospital cardiac arrest: A critical review. World Journal of Emergency Medicine. 2013;4(2):85-91.
[13] Friess, Stuart H. et al.Hemodynamic directed CPR improves cerebral perfusion pressure and brain tissue oxygenationResuscitation , Volume 85 , Issue 9 , 1298 – 1303
[14] Sutton, Robert M, Stuart H Friess, Maryam Y Naim, et al. Patient-centric blood pressure-targeted cardiopulmonary resuscitation improves survival from cardiac arrest. American journal of respiratory and critical care medicine, no. 11