Cushing Syndrome Induced by Serial Occipital Nerve Blocks Containing Corticosteroids


Address all correspondence to Dr. Patrick Lavin, Vanderbilt University Medical Center, 2100 Pierce Avenue, #351, Nashville, TN 37212.


A patient with chronic daily headaches developed overt signs of Cushing syndrome during treatment with serial occipital nerve block injections. Investigation demonstrated an exogenous source of corticosteroids as the cause of the Cushing syndrome in this patient, thus, implicating the corticosteroid component of the occipital nerve blocks. To our knowledge, this is the first report of Cushing syndrome caused by occipital nerve blockade. Caution is warranted in employing even usual therapeutic doses of synthetic corticosteroids, particularly in long-acting or depot preparations.

Occipital nerve blockade is an effective treatment for a number of different kinds of headache and is used by headache specialists, neurologists, and pain management specialists. In addition to occipital neuralgia, it is effective in treating migraine, rebound headache, cluster headache, hemicrania continua, chronic paroxysmal hemicrania, and so-called cervicogenic migraine.1,2

In our experience, occipital nerve blockade results in significant pain relief for variable periods of time and reduces the use of abortive medications. Local anesthetic agents alone may be effective3; however, the addition of depot corticosteroids may prolong the beneficial effect.1


A 39-year-old woman with chronic daily headaches (transformed migraine) attributed to analgesic usage had marked tenderness over both greater occipital nerves. Palpation of the nerves reproduced her headache. She received a series of six bilateral greater occipital nerve blocks over a period of 3 months with temporary benefit each time. Each injection included 1 mL of triamcinolone (40 mg), 1 mL of 2% lidocaine, and 1 mL of 0.5% bupivacaine. In all, she received a total of 480 mg of triamcinolone. Other medications employed during this period included propranolol, hydroxyzine, alprazolam, lorazepam, amitriptyline, riboflavin, sumatriptan, rizatriptan, metoclopramide, prochlorperazine, imipramine, topiramate, metaxalone, diflucan, and acyclovir.

Toward the end of the treatment period, she developed overt signs of Cushing syndrome with intermittent hypertension, severe muscle weakness, fluid retention, centripetal obesity, moon facies, a dorsal fat pad (buffalo hump), and pharyngitis that may have been due to both candida and herpes simplex.

Laboratory evaluation demonstrated suppression of the hypothalamic-pituitary-adrenal axis, indicating an exogenous source of a synthetic steroid not detected by the usual laboratory tests of adrenal function. Her plasma cortisol, adrenocorticotropic hormone, and urinary free cortisol levels were undetectable. Two weeks later, the urinary free cortisol level was still undetectable, and urinary 17-ketosteroids were 2.1 mg/24 hours (reference range, 5.0 to 15.0); however, the thyroid-stimulating hormone level was 1.35 μIU/L (reference range, 0.3 to 5.0), ruling out panhypopituitarism. A week later, a cosyntropin stimulation test with 250 μg was performed. At 1 pm, the baseline cortisol level was less than 1.0 μg/dL (reference range: from 8 am to 9 am, 10.4 to 26.4 μg/dL; at 11 pm, 25% of 8 am to 9 am level), and the aldosterone level was 6.1 ng/dL (reference range: supine, 1 to 16 ng/dL; upright, 4 to 31 ng/dL). At 30 and 60 minutes, the cortisol level rose to 2.9 and 3.6 μg/dL and the aldosterone level to 25.1 and 17.6 ng/dL, respectively. (The normal response of both cortisol and aldosterone is a 3- to 7-fold increase from baseline; the stimulated cortisol should be greater than 5 μg/dL.) The preservation of the aldosterone response to cosyntropin ruled out primary adrenal insufficiency. The cortisol response was blunted but present.


Two months after the last injection with triamcinolone, the patient continued to experience significant nausea. Helicobacter pylori was found in the stomach and treated appropriately, but the nausea continued. A noon serum cortisol level was 1.5 μg/dL, indicating persistent suppression of adrenal function. The nausea was attributed to adrenal insufficiency. Nevertheless, by this time, the patient had been enrolled in a physical therapy program, was losing weight, and was beginning to recover some of her strength, and the physical stigmata of Cushing syndrome were beginning to resolve.


The apparent paradox between the clinical picture of Cushing syndrome and the laboratory picture of adrenal insufficiency was the result of the exogenous corticosteroid, triamcinolone, used in the occipital nerve block injections.4,5 The patient did not receive corticosteroids by any other route. Investigation eliminated Cushing disease, a corticosteroid-producing adrenal tumor, and an ectopic source of corticotrophin such as a neoplasm. None of the other medications she had taken are known to delay the metabolism or elimination of corticosteroids.

This patient demonstrated marked sensitivity to exogenous corticosteroids with rapid development of Cushing syndrome and secondary suppression of adrenal function. She may be a slow metabolizer of corticosteroids.6,7

The systemic administration of exogenous corticosteroids is probably the most common cause of Cushing syndrome. Local administration of corticosteroids, such as paraspinal injections for analgesia,8 intradermal injection for skin disorders,9 intrapericardial injections,10 intraurethral injection for strictures,11 and intra-articular injections12 can also induce Cushing syndrome. However, to our knowledge, this is the first case report of Cushing syndrome caused by serial occipital nerve blockade.

When performing repetitive occipital nerve blockade with corticosteroids, caution is warranted. Patients should be carefully monitored for signs of corticosteroid toxicity.