Bilateral adrenal haemorrhage secondary to rivaroxaban in a patient with antiphospholipid syndrome

Abstract

A 46-year-old man with antiphospholipid syndrome (APS) and previous pulmonary embolism on anticoagulation with rivaroxaban was brought in to the hospital after a syncopal episode. He was found to be hypotensive and tachycardic and later admitted to the intensive care unit. Clinical presentation and laboratory findings were consistent with adrenal insufficiency. MRI revealed bilateral adrenal haemorrhage and he received appropriate steroid replacement therapy. Symptoms slowly subsided and anticoagulation regimen was changed to warfarin. Adrenal haemorrhage was likely caused by APS and rivaroxaban, which brings into question whether novel oral anticoagulants are safe in this patient population.

Background

Adrenal haemorrhage (AH) is a less common cause of adrenal insufficiency (AI) however it is considered a life-threatening emergency so early recognition and prompt steroid replacement therapy is imperative.¹

Multiple aetiologies can precipitate an AH including trauma, sepsis, antiphospholipid syndrome (APS), heparin-induced thrombocytopenia, disseminated intravascular coagulation, among others. APS is a less understood causality of AH and the precise mechanism is not fully understood. A possible explanations include adrenal vein thrombosis with disruption of adrenal gland outflow that ultimately lead to haemorrhagic infarction. When there is thrombosis of the single vein blood drainage of the adrenal gland, further infarction is possible. In patients with APS where thrombosis is a major complication, choosing the appropriate anticoagulation regimen might be challenging. The evolving landscape of pharmacological anticoagulation is continuing to produce newer agents whose effectiveness and safety profiles are only beginning to be fully understood.^1–3

The following case presents an individual with APS who developed bilateral adrenal haemorrhage (BAH) while on rivaroxaban.

Case presentation

A 46-year-old man with APS and previous pulmonary embolism presented to the emergency department (ED) after a syncopal episode. Syncope was preceded by palpitations and episode was witnessed by bystanders who denied seeing any seizure-like activity or bowel or bladder incontinence. Prior to presentation, patient had been reported decreased oral intake, fatigue, abdominal pain, nausea and weight loss for about 1 month. On further questioning, the patient reported he had been on warfarin for 6 years (indicated for his APS and history of pulmonary embolism) until he was recently switched to rivaroxaban.

In the ED he was found to be hypotensive with a blood pressure of 72/45 mm Hg and tachycardic with a heart rate of 103 beats/min. He was started on intravenous fluids; however, he was persistently tachycardic and his blood pressure remained low. He was ultimately started on vasopressor support with phenylephrine and admitted to the intensive care unit (ICU). Laboratory findings revealed hyponatremia (sodium 121 mEq/L), hyperkalemia (K 5.4 mEq/L) and elevated creatinine at 1.8 mg/dL. CT of the brain was negative for any acute findings of haemorrhage or infarcts. Due to concerns for AI in the setting of APS and anticoagulation, a random cortisol was ordered and found to be low at 0.8 µg/dL. He was immediately started on hydrocortisone 50 mg every 6 hours. CT of the abdomen revealed early findings of BAH that was later confirmed with an MRI showing increased T1 signal in region of bilateral adrenal glands (figure 1).

Figure 1.

MRI showing increased T1 signal in region of bilateral adrenal glands.

The patient was started on anticoagulation with heparin drip 24 hours after initial presentation as he was also at a high risk for thrombotic complications. After steroids were started, his blood pressure responded well and phenylephrine was stopped. After vasopressor was discontinued, the patient was started on midodrine as blood pressure was still borderline low. The endocrinology service was consulted and as per their recommendations, the patient was also started on fludrocortisone 0.1 mg daily for mineralocorticoid support in the setting of primary AI. Anticoagulation with warfarin was started and heparin drip continued until international normalised ratio (INR) became therapeutic.

Outcome and follow-up

Patient was transferred from the ICU to the floor and by which time his symptoms and hypotension had subsided. Hydrocortisone was slowly tapered and he was safely discharged home on physiological doses of hydrocortisone 15 mg in the morning and 10 mg at night as well as fludrocortisone 0.1 mg daily. He was discharged with an alert bracelet, dexamethasone emergency kit and instructed about sick day rules. He continues to do well and currently follows with haematology and endocrinology as an outpatient.

Discussion

AI secondary to BAH is a very rare complication from APS and it has been reported in 0.4% of APS cases. Haemorrhagic infarction might be the first manifestation of this disease and it carries an estimated mortality rate of 15%. The presentation of AH includes lightheadedness, hypoglycaemic, hyperkalemia, hyponatremia, as well as non-specific symptoms such as nausea, vomiting and abdominal pain. CT imaging is usually sufficient for diagnosis where adrenal haematomas along with soft-tissue stranding around the adrenal gland are seen. In some patients, CT is non-diagnostic and MRI is indicated.^{4 5}

To the best of our knowledge, there is only one additional reported case of BAH secondary to novel oral anticoagulants (NOACs) in a patient with APS in the literature showing this association.⁶ Potential causes of haemorrhagic infarction in these patient population include supra and subtherapeutic INR in patients taking warfarin. The pathophysiology of how NOACs lead to thrombosis and infarction has not been elucidated.⁷

In the literature, there are also cases reported about BAH secondary to rivaroxaban suggesting that NOACs have a potential for this complication even in patients without underlying APS. Some case reports have also shown increased bleeding risk with concurrent use of NOACs with other drugs such as amiodarone, fluconazole, rifampin and phenytoin. Our patient was not taking any medications to our knowledge that are known to lead to this drug interaction.⁵

Similarities with the case reported by Alidoost et al are the presence of AH and the use of the same anticoagulant: rivaroxaban. Differences are that our patient had antiphospholipid syndrome while the reported case by Alidoost et al did not have antiphospholipid syndrome. Our case is also different because our patient had been taking rivaroxaban for few months; the patient reported case by Alidoost et al had just started taking it (2 weeks only). The case by Alidoost et al was more subacute as it started on the left adrenal initially and later affected the right adrenal. The presentation was less acute as only one adrenal was affected first and vitals were stable initially, AI developed later (when both adrenals were affected).⁵

Similarities with the case reported by Sanford et al are the use of anticoagulants of the same class: factor Xa inhibitors and that AH occurred several months after starting anticoagulation. Differences are: the drug was apixaban not rivaroxaban. The case by Sanford et al did not develop AI, as this patient did not develop BAH, only one adrenal was affected, in our case both adrenals were affected at the same time. The case by Sanford et al also had retinal haemorrhage and the AH was recurrent (twice after switching warfarin to apixaban).⁶

The management of APS patients presents unique challenges as existing literature is scant on establishing an ideal dosing regimen. Even with uncertainty, NOACs are increasingly being used.⁸ Current data suggest that treatment failure remains possible with NOACs and that warfarin is superior. A recently published randomised trial evaluating rivaroxaban versus vitamin K antagonist therapy in APS revealed recurrent thrombosis in patients with APS when switched from warfarin to NOACs.⁹ These findings are consistent with our case report on how this patient had no complications while on warfarin and developed BAH when switched to rivaroxaban.

NOACs are relatively new agents in the management of acute and chronic deep vein thrombosis (DVT) and pulmonary embolism (PE) with little literature studying its effects in patients with APS, therefore, warfarin seems to currently be the safest option.

Lack of sufficient data regarding the efficacy of these medications including ideal dosing regimens may result in higher rates of adverse effects, such as bleeding. Further studies and the development of specific protocols are needed prior to the implementation of NOACs as the first line anticoagulation option for this patient population.

Patient’s perspective.

I am new to the club. I did not know it was so exclusive. Nor that you do not choose to be in it, but rather one is chosen to be part of it.

Six weeks into my diagnosis, I am relieved that I finally know what ailed me and totally transformed my personality and way of life for 7 treacherous weeks. I first presented with acute abdominal pain; the pain was so profound that I thought I was going to give birth to a Gremlin. After being in the hospital for 4 days and having run every single imaginable test for the kidneys, gall bladder, liver, upper GI, my ‘hypothetical’ diagnosis was duodenitis. I did not completely understand it, but I was glad to come home.

The ensuing following 7 weeks saw me gradually grow weaker, more fatigued while being simultaneously hungry and nauseous. My energetic positive personality faded into one of frustration, irritability and moodiness. Standing up for long periods of time was non-existent and I developed a pain in the mid-section of my back. I worked as I could, when I could. After multiple visits to my PCP and gastro—their enhanced hypothesis was that I had Helicobacter pylori—a bacteria in my stomach.

That all changed on Saturday, October 25. I had a major crisis. The next 11 min would change my life.

While waiting outside a supermarket for my girlfriend to pay, I started having hallucinations. It felt like thousands of javelins were being thrown at me simultaneously. It then felt as though my eyes started spinning and rolling backwards like a slot machine. They called 911 rescue and I was rushed to the hospital, where they initially found that my blood pressure was 62/42 and my sodium was 121.

I spent 7 days in total in the hospital. I did not realise how precarious my medical condition was until on day 5 the Internal Medicine Department wrote up my case, presented it to me, and asked for my permission to publish it. Haematology and endocrinology were interested as well and gladly I provided them with my consent. I had Addison’s, and my crisis occurred when both of my adrenal glands haemorrhaged simultaneously.

I am blessed to have an amazing medical team and an National Adrenal Diseases Foundation (NADF) support group that is helping me transition to my new reality. I am a sponge to acclimating to the nuances of my condition, my disease and as a marketing professional to find ways to raise awareness and consciousness about the disease and how it goes horribly mis-diagnosed.

So, as I prepare to enter a new decade, I am optimistic and brimming with gusto. I am also considering changing my middle name to ‘Resilient’ because that is my plan for the next 40+ years I have left on Mother Earth. One day at a time, one pill at a time, I too, will conquer this.

Learning points.

• Adrenal haemorrhage is a life-threating emergency, early recognition is key to improve survival.

• Novel oral anticoagulants efficacy including ideal dosing regimen in antiphospholipid syndrome has not been well studied and therefore, must be used with caution.

• Warfarin is currently the anticoagulant of choice for patients with antiphospholipid syndrome (APLS).