Adrenal Insufficiency: A Common Finding in Adults with Sickle Cell Disease (SCD)

Inflammation is believed to play a major role during vaso-occlusive crises (VOC) in SCD patients (pts) and brief courses of high dose steroid therapy have been shown to shorten the duration of these events. However, rapid relapse after steroid withdrawal has limited this therapeutic approach. Even less well understood are the chronic pain syndromes experienced by many adult SCD pts. It is of interest then that chronic adrenal insufficiency may cause musculoskeletal pain; that impairment of the pituitary-adrenal axis has been described in association with opioid use as well as in some pts with chronic pain syndromes (e.g. fibromyalgia) or psychiatric disorders (e.g. anxiety and post-traumatic stress); and that hypocortisolism has been linked to increased pain sensitivity in some settings. Thus, a retrospective review was performed of measures of pituitary-adrenal function determined during the course of inpatient treatment of severe VOC in 12 adult SCD pts during a 2 month period from March thru April 2008. Pts were 22–54 yrs old and 7 were men. The SS genotype was present in 10 pts and 2 pts were S-β-thalassemia. Adrenal insufficiency was present in 9 pts (75%). Eight pts had serum cortisol values < 7μg/dl (including 7 pts with values ≤ 4 μg/dl) (normal = 7–25 μg/dl) and 1 pt had a low 24 hour urine value for free cortisol of 2.8 μg/24 hrs (normal = 4–50 μg/dl). Moreover, factors usually associated with increased cortisol secretion were present in 6 of these 9 pts (67%) including infection in 2 pts; cocaine use in 2 pts; both infection and cocaine use in 1 pt; and inflammation with an increased high-sensitivity C-reactive protein (hsCRP) value of 24.3 ng/ml (normal = 0.1–4.9 ng/ml) in 1 pt. Low serum cortisol values were accompanied by low serum total CO2 values suggestive of metabolic acidosis in 4 pts (40%) and by increased absolute eosinophil counts (>500/mm3)in 6 pts (67%) including 3 of the 6 pts without evidence of cocaine use (50%). Serum ACTH was measured in 4 of these pts and was normal in all of them with values of 6–15 pg/ml (including 2 pts with values of 15 pg/ml) (normal = 5–27 pg/ml). Administration of 250 μg of cosyntropin intravenously resulted in increases in serum cortisol >20 μg/dl in 5 of the 7 pts tested (71%). In the 3 pts with “normal” basal serum cortisols, values were all in the low-normal range (7, 8, and 9 μg/dl) despite the presence of active infection in 2 of them and cocaine use in the third. Serum total CO2 was low in 2 of these 3 pts and eosinophilia was present in 1 of them. Moreover, one of these 3 pts had a low serum cortisol value of 4 μg/dl on a second admission 4 months later despite an active infection with an hsCRP value of 24.8 ng/ml and evidence of cocaine use. In conclusion:

1. hypocortisolism is a common finding in adult SCD pts during severe VOC despite the frequent presence of factors usually associated with stimulation of the hypothalamic-pituitary-adrenal axis (i.e. acute pain, infection, inflammation and cocaine use); and

2. the presence of metabolic acidosis and eosinophilia suggest that hypocortisolism in this setting may be functionally significant.

Further studies are needed to define the mechanism and metabolic significance of hypocortisolism and to establish its role in both pain sensitivity and the chronic pain syndromes in adult SCD pts.