Daily Adjustment of Glucocorticoids by Patients With Adrenal Insufficiency

Christof Schöfl; Bernhard Mayr; Nicole Maison; Felix Beuschlein; Gesine Meyer; Klaus Badenhoop; Tina Kienitz; Marcus Quinkler


Clin Endocrinol. 2019;91(2):256-262. 

In This Article


This study evaluates the day-to-day self-management of patients with AI, which provides a wealth of information on the daily actions patients take, especially in situations, which might result in the development of an AC. Furthermore, our study provides first insight in the number of days with discomfort and illness in patients with AI.

Our patients documented discomfort of different degrees in 13.6% of all recorded days. In approximately a third of those days (=4.8% of all recorded days), the patients performed a dose adjustment due to their discomfort. This implies that patients were firstly worried about their health and secondly felt a decreased level of well-being, which was followed by a dose increase. Therefore, we assume that those 4.8% of days might have been sick days. Data from the German Health Ministry showed that female jobholders have a sick day frequency of 7% of working days (230 days; =4.4% of the whole year) and male jobholders of 6.1% of working days (230 days; =3.8% of the whole year).[15] This corresponds well to the 4.8% of possible sick days in the current patient group indicating a similar level of sick leave to that of the general population. In addition, similar to the German Health data, in our cohort we also observed a higher frequency and longer duration of episodes of discomfort in women than in men.

The frequency of dose increase (35% of days with discomfort) seems quite low on first sight; however, discomfort for itself does not make an increase necessary. Probably, patients felt their well-being was not impaired enough to necessarily increase GC dose, or they were too afraid of a dose increase because it might, for example, cause side effects such as osteoporosis or metabolic disturbances. However, if the days with discomfort and no dose increase (65% of days with discomfort) were relevant sick days with significant health impairment, this would have led to a significant increase in adrenal crises. In contrast, we only observed two adrenal crises during the documented period (93.4 years), which is below the reported frequency.[16–18]

During the last decades, the average daily GC dose has gradually been reduced from approximately 30 to 20 mg HC/d. This was supported by studies showing a reduction in GC side effects such as osteoporosis, hypertension and metabolic effects due to daily dose reduction.[19–22] However, critics stated that lower doses of short-acting GC replacement therapy might actually predispose patients to ACs during illness by exposing them to periods of relatively profound hypocortisolaemia.[23] Therefore, when applying lower doses of short-acting GC replacement therapy, it is important to implement a more flexible daily dose regimen with patient's led dose adjustments. In the present study, the average hydrocortisone dose was 23.6 mg/d, and according to the percentage of sick days and adrenal crises, our cohort did not experience any disadvantage of this regimen.

On the other hand, too liberal dose adjustments might lead to overdosing by the patient. Assuming that our ssScore of 1–3 includes only mild symptoms and no imminent risk for AC, several patients doubled or even tripled their dose with this score (Figure 1). This might indicate overdosing by the patient, probably due to lack of knowledge of signs of hypocortisolism and the fear of a possible AC. Therefore, not only neglected dose adjustment, but also inappropriate over-replacement should be addressed in patient's teaching.

Nonetheless, it is unclear why our well-educated (educated locally in their endocrine centres) patients still increased their daily dose only by 55% in situations, which indicate GI infections. These infections are known to be the major predisposing factors leading to AC.[8,16,24] This finding might be driven by anxiety of overdosing and the fear of side effects such as obesity and osteoporosis.[11] Therefore, we suggest that every patient should be trained repeatedly (eg on a 6-month basis) to guarantee better therapy adherence and to achieve better coping of situations with imminent GI infections. The significant variation in behaviour might also be caused by different advice given by the endocrine centres and by the difference the information given is absorbed by the patients during the consultations. Therefore, we believe that the patient's training should be structured and nationwide to avoid inconsistencies across the country. Therefore, by end of 2014 we started to develop a nationwide, structured and certified education programme with teaching teams consisting of an endocrine nurse and an endocrinologist and standardized training materials available to all qualified teaching teams via an Internet-based platform. The structure of the patient education programme consists of small teaching groups of 3–5 patients with accompanying relatives or spouses with a length of 90–120 minutes. The aims of the teaching programme are to encourage and enable patients to adjust their hydrocortisone medication in special situations, to train self-, respectively partner-injection of parenteral hydrocortisone in case of emergency, to create empowerment and to prevent adrenal crises. During the teaching course, general information on AI is given, sign and symptoms of hypocortisolism discussed, steroid adjustment in special and critical situations explained and self-injection of parental hydrocortisone practically trained.

Shortcoming of the presented study is the observation that in some cases patients only documented that GC dose was changed without further details on the extent of that dose adjustment. Secondly, the number of documented days in the diaries varied among patients. Thirdly, we did not further analyse the 20 patients who refused to take part in the study. Therefore, it might be possible that there is a selection bias in our cohort. Furthermore, interpretation of signs and symptoms of each individual included in this study might have led to inherent bias and the sign and symptom score used is unvalidated. Despite these limitations, this study with patient's self-management documentation on a day-to-day basis over a period of one year on average provides valuable insights into the management of chronic AI by patients.

In summary, our study shows that the perception of discomfort does not always seem to result in a GC dose increase in AI patients, especially in situations indicating GI infection. On the other hand also possible overdosing was seen in some patients. This underscores an urgent need for improved training methods. Keeping daily records might be a useful tool for continued and individualized patient education.