Abstract
Triamcinolone is a long-acting glucocorticoid medication that can be responsible for transient suppression of the hypothalamic–pituitary–adrenal (HPA) axis. This physiologic alteration may persist for weeks after repeated or even single localized injection of this agent. However, when this glucocorticoid agent is given to patients receiving the HIV protease inhibitor (PI) ritonavir (RTV), inhibition of their shared cytochrome P450 3A4 degradation pathway leads to an increased bioavailability of triamcinolone, with subsequent heightening and prolongation of the glucocorticoid serum levels. In those instances, iatrogenic Cushing syndrome may ensue. The authors encountered such an event in an HIV-infected patient on chronic treatment with an antiretroviral regimen containing RTV. The patient’s clinical presentation and laboratory investigations confirmed a diagnosis of Cushing syndrome and secondary adrenal insufficiency. This was believed to have occurred in close association following cervical vertebral column facet joint injections with triamcinolone acetonide for cephalagia deemed related to cervical spine disease. The discontinuation of the RTV-boosted PI therapy alone, promoting the clearance of the elevated triamcinolone serum levels and restoration of HPA homeostasis, proved successful in this patient. For this case, the authors review the published English medical literature relating to this uncommon phenomenon.
Introduction
Ritonavir (RTV) is a frequently prescribed protease inhibitor (PI) medication administered to HIV-infected patients as part of combination antiretroviral therapy (cART). In and of itself, RTV possesses potent antiviral activity. However, it currently is included in low doses in cART solely for its effect of pharmacologically “boosting” other PIs. The RTV serves to synergistically prolong the terminal half-life, and thus augment the area under the concentration curve (AUC), of the parent PI component of the patient’s cART regimen. Mechanistically, this occurs via the action of competitive inhibition of the cytochrome P450 isoenzyme CYP3A4 that serves to metabolize almost all HIV PIs. 1 Coincidently, glucocorticoid medications, such as triamcinolone, are also metabolized by the same CYP3A4 enzyme system. Therefore, deliberate inhibition of CYP3A4 by RTV inadvertently increases the bioavailability and half-life of the prescribed glucocorticoid agents, potentially resulting in unanticipated and deleterious alterations in the individual’s hormonal physiology. The authors describe a case of a 34-year-old female who presented with rapid onset of Cushing syndrome after undergoing facet joint injection treatment with triamcinolone acetonide while taking cART that included RTV. For this case, the authors review the published literature of similar reported cases of iatrogenic Cushing syndrome in HIV-positive patients receiving RTV.
Case Report
A 34-year-old HIV-infected female presented to the hospital, complaining of labile mood swings, fatigue, muscular weakness, menorrhagia, facial swelling, and abdominal distention over a period of 3 weeks. Her long-standing cART regimen included daily tenofovir (TDF) 300 mg, emtricitabine 200 mg, atazanavir (ATV) 300 mg, and RTV 100 mg. She had an undetectable HIV RNA (<20 copies/mL) level and a CD4 count of 307 cells/mm3. Her physical examination was notable for normal vital signs, moon facies, hirsutism, central obesity with sparing of the limbs, and significant proximal muscle weakness. One month prior to her admission, the patient was evaluated at an outside hospital for severe headaches. She received empiric meningitis treatment with ceftriaxone and a single dose of 4 mg of dexamethasone. Subsequently, an alternative diagnosis of cephalalgia due to cervical spine disease was made, and she underwent 2 sets of spinal facet joint injections 1 week apart, for a total of 120 mg triacmcinolone acetonide, with subsequent symptomatic relief.
Given her clinical presentation and her history of glucocorticoid administration, a diagnosis of Cushing syndrome was suspected. Measured serum cortisol levels were diminished; 1.28 (3-13 ug/dL) and 0.46 μg/dL (5-31 ug/dL) in the evening and morning, respectively. Adrenocorticotropic hormone (ACTH) was noted to be low at 2.0 pg/mL (7.2-63.3 pg/mL). A cosyntropin stimulation test was performed. Serum cortisol levels measured at 30 and 60 minutes following the intravenous administration of 0.25 mg of cosyntropin were 5.95 and 8.87 g/dL, respectively (Table 1). The results were consistent with iatrogenic Cushing syndrome, with concomitant secondary adrenal insufficiency. In an attempt to relieve the patient’s clinical symptoms of fatigue and muscle weakness, replacement therapy with oral hydrocortisone (25 mg
Cortisol and ACTH Levels during Patient Follow-Up.
Abbreviation: ACTH, adrenocorticotropic hormone.
a A cosyntropin stimulation test was performed on day 2 of hospital stay. Cosyntropin of 0.25 mg was given intravenously and the cortisol level was measured after 30 and 60 minutes.
Following her hospital stay, the patient’s clinical course steadily improved. At the 1-week clinic visit, having already received a 10-day course of hydrocortisone, the patient’s hydrocortisone was discontinued upon conclusion that her clinical presentation was not related to diminished glucocorticoid levels. At this time, her muscle weakness and fatigue had improved. The pharmacologic effect of RTV on triamcinolone acetonide resulted in an exaggerated prolonged elevation of this glucocorticoid, which was postulated to be the etiology of her clinical presentation. Random serum cortisol levels obtained at 1 week and 4 months after her initial presentation were found to be 2 and 16.7 µg/dL (>10 µg/dL), respectively. Her ACTH level at 4 months post initial presentation was found to be 19.2 pg/mL, which was within the normal limits (7.2-63.3 pg/mL). Four months after her initial presentation, the patient states that she feels close to her baseline; the patient’s facial swelling and abdominal distention had also resolved.
Discussion
Iatrogenic Cushing syndrome with hypothalamic–pituitary–adrenal (HPA) axis suppression from systemic corticosteroid use can develop rapidly, within 5 days, when high doses of corticosteroids are given to a patient. 2 The normal physiological release of glucocorticoid occurs via both circadian and high-frequency pulsations, with these stimulations determining the release of glucocorticoid receptor–mediated transcripts. This phenomenon has been referred to as gene pulsations. Persistence of glucocorticoid with the loss of circadian and high-frequency pulsations, as observed in exogenous corticosteroid use or in Cushing disease, leads to altered gene expression and cushingoid presentation. 3 In this patient, an iatrogenic Cushing syndrome likely developed secondary to increased bioavailability and prolonged exposure to triamcinolone acetonide owing to its interaction with RTV by competitive inhibition of its metabolism via the CYP3A4 isoenzyme.
A search of the literature for articles in English resulted in 13 case reports of iatrogenic Cushing syndrome associated with the concomitant use of RTV with triamcinolone acetonide. 4 –10 Including this patient, the total number of described instances of iatrogenic Cushing syndrome felt to be clearly associated with the concomitant use of RTV with triamcinolone acetonide is 11 (Table 2). To compare the efficacy of the different treatment options, the authors used the normalization of HPA, defined as a return of serum cortisol levels to normal physiological values and/or a decrease in serum level of triamcinolone acetonide to an undetectable level (<0.1 µg/dL), as an objective measurement of time to resolution.
Summary of Interaction between Ritonavir and Triamcinolone.
Abbreviations: AVN, avascular necrosis; d/c, discontinue; DVT, deep vein thrombosis; F, Female; HC, hydrocortisone; HHS, hyperosmolar hyperglycemic state; M, Male; RTV, ritonavir.
aWeeks are recorded as 7 days; months are recorded as 30 days.
Of the 11 patients, 4 underwent no intervention, 3 were provided with tapering hydrocortisone alone, 2 patients were discontinued from RTV alone, and 2 patients were provided with hydrocortisone and discontinued from RTV. The onset of the cushingoid presentation occurs rapidly and within the first 2 weeks after the last administration of triamcinolone acetonide. Patients who received higher or repeated doses of triamcinolone acetonide also had increased morbidity, including the development of DVTs, hyperosmolar hyperglycemic state, and avascular necrosis.
The discontinuation of RTV as part of the boosted PI regimen has demonstrated to have a significant effect on triamcinolone acetonide metabolism/clearance and subsequent recovery of the patient’s glucocorticoid homeostasis. 8 Treatment with tapering doses of hydrocortisone did not seem to affect the HPA axis recovery time but has been reported to offer some symptomatic relief. 7,10
The timing of symptomatic relief varied greatly among the patients, and physical manifestations of Cushing syndrome may persist long after HPA axis normalization. Of the various approaches to management, the patients who discontinued RTV therapy experienced a more rapid recovery of their HPA axis from the time of the last triamcinolone acetonide administration (range of 3-4 months) 5,7,9 compared with the patients who did not discontinue their RTV therapy (range of 2.5-6 months). 4,6,8,10
Iatrogenic Cushing syndrome has also been described with RTV and other glucocorticoids including fluticasone and dexamethasone. A literature review by Foisy et al 11 examined the use of inhaled and intranasal fluticasone in 25 HIV-infected patients receiving cART inclusive of RTV. In these individuals, therapeutic interventions consisted of the use of alternative non-PI-based cART therapy, use of alternative glucocorticoid therapy with less pharmacokinetic interaction with RTV, reduction of fluticasone dose, and/or discontinuation of the glucocorticoid. Although no specific intervention was proved to be clearly superior, Foisy et al 11 recommended continuation of hydrocortisone replacement therapy, while morning serum cortisol levels remained at less than 3 µg/dL.
A report by Molloy et al 12 examined a single case of iatrogenic Cushing syndrome believed to be related to RTV use in an HIV-infected individual while receiving concomitant chronic dexamethasone ophthalmic drops. The syndrome fully resolved with discontinuation of the glucocorticoid ophthalmic drop. These cases help to further exemplify that even some topical or inhaled glucocorticoids, when given along with RTV, can cause iatrogenic Cushing syndrome.
Conclusion
Diagnosis of iatrogenic Cushing syndrome in HIV-infected patients can be difficult, as the clinical presentation may mimic other HIV-associated diseases, such as opportunistic infections, malignancy, lipodystrophy, AIDS, or failed cART. Concomitant antimicrobial agents including macrolides and azoles are potent CYP3A4 enzyme inhibitors that may lengthen or worsen the patient’s symptoms. Alternative cART medications with potent CYP3A4 enzyme inhibitor activity, such as other PIs or the newly approved boosting agent cobicistat, may also prolong or worsen the patient’s clinical course. A detailed medication history is important in making the appropriate diagnosis, as patients will often fail to report corticosteroid use, especially if administered for a brief period of time and believed to be separate from their HIV treatment. A cosyntropin stimulation test may be suggestive of an iatrogenic source of corticosteroid. In addition, if suspected, synthetic corticosteroids can be detected from serum to confirm a diagnosis of iatrogenic Cushing syndrome.
Treatment should depend on the severity of the symptoms. Along with symptomatic treatment in severe cases, all corticosteroid should be discontinued when possible. The RTV should be removed from the cART regimen. Moreover, a thorough review of the patient’s medications is essential to eliminate any other potent CYP3A4 enzyme inhibitor medications that may compete with the metabolism of the synthetic corticosteroids.
Improvement in and eventual resolution of the signs and symptoms of excess corticosteroids should be determined by close follow-up clinical observations. In those individuals who symptomatically improve, repeat cortisol assays and cosyntropin stimulation tests are unnecessary to determine HPA axis recovery. In those situations where corticosteroid medications are still required, formal testing demonstrating full normalization of the HPA axis is recommended.
As HIV-infected patients are now living longer, many are likely to develop unrelated illnesses and conditions that require glucocorticoid medications for treatment. It is important that clinicians be made aware of the potential pharmacologic interactions of the components of their cART regimen (not limited to RTV) before prescribing these agents, in order to avoid any unexpected adverse events.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.