Iatrogenic adrenal insufficiency related to corticosteroid eye drops: a case report

Lesson

Intra-ocular corticosteroids have the potential to cause adrenal insufficiency when used long term. Patients and clinicians should be counselled on the risks of long-term steroid use and consider appropriate monitoring.

Corticosteroids are widely used in the management of a variety of medical conditions, from autoimmune and inflammatory, to malignancies and after organ transplantation. The main mechanism of action of corticosteroids, making them so widely used, is through their anti-inflammatory and immunosuppressive properties. ¹

Though beneficial in numerous ways, it is widely researched and documented that the chronic use of exogenous glucocorticoids has the potential to lead to the suppression of the hypothalamic-pituitary-adrenal (HPA) axis due to negative feedback, with no substantive correlation with dose of steroid or duration of treatment. ² Though topical administration of corticosteroids is deemed to be safer than the oral or parenteral routes, there is still a risk of adrenal insufficiency (AI) with prolonged use due to systemic absorption. ³ This case report discusses an incidence of adrenal insufficiency following the prolonged use of corticosteroid eye drops.

Case

A young boy had been diagnosed at birth with bilateral Peter's anomaly - a condition characterised by central, paracentral, or complete corneal opacities, congenital glaucoma, iridocorneal adhesions and cataracts. ⁴ He had been treated with glaucoma tube implant surgery, bilateral penetrating keratoplasties (full thickness corneal transplants), and dexamethasone 0.1% eye drops twice daily for prevention of rejection.

By the age of 9, he presented with cushingoid features and symptoms of possible adrenal insufficiency. As a possible precipitant for his symptoms, his daily dexamethasone eye drops were therefore held for 1 week. During this time, his parents were educated on the possibility of developing adrenal crisis and sent home with an emergency hydrocortisone supply. The following week a short synacthen test was arranged which demonstrated a suppressed corticosteroid response of serum cortisol <10 at 0, 30 and 60 min. Additionally, his ACTH level was <5 (ref 7.2–63.3 ng/L) and demonstrated negative adrenal antibodies. Subsequent assessments of the HPA axis have demonstrated complete adrenal suppression.

Due to the need to continue dexamethasone eye drops for his chronic ophthalmic condition, he is concurrently treated with hydrocortisone replacement therapy to manage his adrenal insufficiency.

Discussion

Secondary adrenal insufficiency is a widely proven yet under-recognised complication of most forms of long-term glucocorticoid therapy. Adrenal suppression can often present asymptomatically or with non-specific symptoms, and a high clinical suspicion is needed to make the appropriate diagnoses. Undiagnosed AI can have serious consequences, and is associated with significant morbidity, including poor response to illness, hypoglycaemic seizure, coma and even death following adrenal crisis.

The development of adrenal insufficiency following intra-ocular steroid treatment is rare and unusual. The Society for Endocrinology guidance suggests an indicative threshold of 1000 micrograms per day, and they reason that this dose limit would be very unlikely to be reached with available steroid eye drops, and therefore HPA axis suppression would be unlikely. ⁵ The current research available on this topic provides mixed results, with no conclusive association between ophthalmic corticosteroid use and adrenal suppression.

A study analysing the incidence of HPA axis suppression in infants with the use of steroid eye drops after surgery found that they were at risk of receiving high dose steroids per weight by topical ocular administration. ⁶ Another similar study analysing incidence of AI in infants treated with steroid eye drops following cataract surgery found that two thirds of the infants showed evidence of adrenal suppression, with there being a correlation between cumulative daily dose of steroid and the test result. ⁷

On the contrary, a study by Sandhu et al. sought to establish whether HPA axis suppression is possible following long-term topical ophthalmic corticosteroid use in patients who have undergone penetrating keratoplasty (PKP) suggests there is no evidence that long-term low dose use of corticosteroid eye drops to prevent graft rejection causes adrenal suppression in adults. ⁸ The study acknowledges that many conditions require higher strengths of corticosteroids and therefore those patients may be more at risk of AI, especially if used in combination with systemic steroids (Table 1).

Table 1.

Doses of exogenous steroid preparations above which patients can be at risk of HPA axis suppression. ⁵

	Medicine	Dose
Long-term oral glucocorticoids	Prednisolone	5 mg per day or more
	Methylprednisolone	4 mg per day or more
	Dexamethasone	500mcg per day or more
	Hydrocortisone	15 mg per day or more
Inhaled glucocorticoids	Beclometasone (Clenil, Easihaler, Soprabec)	>1000mcg per day
	Beclometasone (Qvar, Kelhale or Fostair)	>500mcg per day
	Budesonide	>1000mcg per day
	Fluticasone furoate (Trelegy and Relvar)	>200mcg per day
Nasal glucocorticoids		>1000mcg per day
Steroid eye drops		>1000mcg per day
Topical glucocorticoid creams and ointments	Potent or very potent topical glucocorticoids	>200 g per week

Many conditions require the use of corticosteroids and hence their use in medical treatment is often unavoidable. However, measures can be put in place to minimise the adverse effects of their use. When commencing patients on exogenous corticosteroids, potential risk factors that could potentiate AI should be identified, such as physical stressors, dose and duration of steroid use, and warning of sudden cessation of treatment. Particularly in infants and children frequent monitoring for adverse effects is advised, with follow up including growth charts, BMI tracking, and cortisol tests. ⁹

Clinicians should aim to achieve the lowest effective dose for treatment and regularly review the need for continuation of treatment. There is some evidence that opting for alternate day steroid prescribing could result in lower risk of AI. Similarly, using multiple doses throughout the day or nightly doses may result in greater adrenal suppression as this counters the body's natural cortisol rhythm.

Most importantly, patient and family education are key. They should be educated regarding the risks of long-term steroid use, to be aware of signs and symptoms of adrenal suppression and actions if AI is suspected. They should be reassured that though these risks exist, this should not be a barrier to continuation of treatment, and that appropriate medication adherence and regular follow-up are advised to monitor for any adverse effects. Patients should carry their steroid treatment card in case of emergency, be educated on stress dosing and when and how to administer intramuscular hydrocortisone in the instance of severe illness or adrenal crisis.

The European Society for Endocrinology and Endocrine Society have outlined recommendations for patients on long-term glucocorticoid therapy, reinforcing the importance of patient education, advice for the approach to steroid-tapering, and when to refer to endocrine specialists. They also highlight the current guidance on diagnosing with morning serum cortisol levels, and treatment options of adrenal crisis in patients with steroid induced AI. ¹⁰ These recommendations are further summarised in Table 2.

Table 2.

Summary of recommendations for patients on chronic glucocorticoid therapy based on advice from the European Society of Endocrinology and endocrine society joint clinical guideline. ¹⁰

General recommendations	Detailed recommendations
Recommendations for initiation of glucocorticoid therapy for non-endocrine conditions	1. Patients on standard or tapering glucocorticoids for non-endocrine conditions do not need specialist endocrine evaluation
	2. Clinicians who initiate steroid treatment should educate patients about important aspects of glucocorticoid therapy
	3. Patients on glucocorticoid therapy should have access to current information about endocrine aspects of their therapy.
Recommendations regarding tapering for non-endocrine conditions, steroid-induced AI and steroid withdrawal syndrome	1. Tapering is not necessary in patients on <3–4 week of glucocorticoid therapy, regardless of dose
	2. Taper for patients on longer term glucocorticoid therapy should be attempted when underlying disease is controlled and steroids are no longer required. Taper continued until physiological dose reached
	3. Consider glucocorticoid withdrawal syndrome can occur during taper. If severe, the dose can be increased, and duration of taper increased
	4. Routine testing for AI in patients on supraphysiologic doses of steroid is not recommended
	5. Patients taking long-acting glucocorticoids should be switched to short-acting glucocorticoids as soon as can be tolerated
	6. Patients on physiologic daily dose equivalent and planning to discontinue therapy should: –– Gradually taper steroid dose with monitoring for symptoms of AI, or, –– Be tested with morning serum cortisol.
	7. Morning serum cortisol should be used as confirmation of recovery of HPA axis
	8. Patients tapering or stopping glucocorticoid therapy should not have routine dynamic tests for diagnosing AI
	9. Highlight possible glucocorticoid-induced AI in more at-risk patients
	10. Patients who have received glucocorticoid therapy and presenting with features of exogenous Cushing syndrome are assumed to have steroid-induced AI
	11. Patients who aim to discontinue glucocorticoids and do not have recovery of HPA axis in one year on physiologic daily dose equivalent should be referred to an endocrine specialist.
	12. Fludrocortisone is not recommended in patients with glucocorticoid-induced AI.
Recommendations for adrenal crisis in patients with glucocorticoid-induced AI	1. Patients with current/ recent glucocorticoid use (who did not undergo biochemical testing to rule out glucocorticoid-induced AI) should have stress dose coverage when exposed to stress. 1a. Oral glucocorticoids should be used in cases of minor stress when there is no haemodynamic instability or diarrhoea or vomiting. 1b. Parenteral glucocorticoids should be used in cases of moderate to major stress, procedures under general/ regional anaesthesia, procedures requiring decreased oral intake, signs of haemodynamic instability or prolonged diarrhoea or vomiting
	2. In patients presenting with the above, diagnoses of adrenal crisis should be considered regardless of steroid type, mode of administration or dose. Patients with suspected adrenal crisis should be treated with parenteral glucocorticoids and fluid resuscitation

Conclusion

Secondary adrenal insufficiency induced by intra-ocular corticosteroid eye drops, although rare, is an important cause of AI that should be assessed and investigated appropriately. Clinicians and patients should be educated on the potential adverse effects of long-term steroid treatment, how to reduce risks of this and when to seek prompt medical attention. The recommendations in this review are applicable to a wide range of specialties and AI is a pertinent and clinically relevant condition to consider in the ever-growing landscape of exogenous steroid treatment, regardless of therapy route. Though there is a risk of adrenal suppression with the use of corticosteroids, for many conditions this constitutes necessary treatment, and should not be a hindrance to their continued use.