Electronic Letters to:

Article: Roderick F.A. de Lind van Wijngaarden, Barto J. Otten, Dederieke A.M. Festen, Koen F.M. Joosten, Frank H. de Jong, Fred C.G.J. Sweep, and Anita C.S. Hokken-Koelega High prevalence of central adrenal insufficiency in patients with Prader-Willi syndrome J Clin Endocrinol Metab 2008; 0: jc.2007-2294v2 [Abstract]

eLetters: Submit a response to this article

Electronic letters published:

Interpreting central adrenal insufficiency in PWS

Roderick F.A. de Lind van Wijngaarden, Barto J. Otten and Anita C.S. Hokken-Koelega   (29 August 2008)

Interpreting central adrenal insufficiency in PWS

Aaron L. Carrel, David B. Allen, MD   (22 June 2008)

Interpreting central adrenal insufficiency in PWS 29 August 2008
Roderick F.A. de Lind van Wijngaarden, Clinical Research Fellow Dutch Growth Research Foundation and Erasmus University Medical Center / Sophia Children's Hospital, Barto J. Otten and Anita C.S. Hokken-Koelega Send letter to journal: Re: Interpreting central adrenal insufficiency in PWS r.delindvanwijngaarden{at}erasmusmc.nl Roderick F.A. de Lind van Wijngaarden, et al.	We thank Carrel and Allen for their comments. The E-letter contributes to the debate about the possible causes of death in patients with Prader-Willi syndrome (PWS) and the interpretation of our data. The diagnostic accuracy of the metyrapone test (MT) is questioned by a recent report (1). In this article, the insulin tolerance test (ITT) was used as a reference test. It has often been published that the MT diagnosed more patients with adrenal insufficiency than the ITT, especially those with central adrenal insufficiency (CAI) (2-4). There are two possible explanations for this finding. Either the ITT is truly the golden standard and the MT has a higher false positive rate (1), or the MT is more sensitive than the alleged golden standard. The latter has frequently been suggested in the literature (2-4). Giordano et al. (1) found a high false positive rate when the ACTH cut-off level during a MT is set at 33 pmol/l. This conclusion is directly dependent on the cortisol cut-off level during the ITT (500 or 550 nmol/l). If this study had used 550 nmol/l, 13% more patients would have been diagnosed, indicating that the difference between the ITT and MT may be much smaller than suggested. Moreover, it is not clear when the ITT was performed. It is well known that CAI may develop over time (3). If the ITT was performed some time before the MT, a higher number of patients diagnosed with the MT may reflect development of CAI in the period between tests. These results should therefore not be considered false positives. When suggesting 18 pmol/l as an ACTH cut-off level, Carrel and Allen refer to Steiner et al (4). However, this study showed that a cut-off level of 33 pmol/l should be used to diagnose patients with partial CAI, as in PWS. The suggested cut-off level (18 pmol/l (1)) reduced sensitivity to 71% and about one-third of diagnosis would be missed. Slightly higher cortisol levels after the MT in patients with CAI have previously been reported (4), but could be interpreted as a reduced cortisol inhibition and consequent ACTH response. Morning cortisol levels in all our patients were low. There was no significant difference between the two groups in the decline of cortisol levels from 2330h to 0400h, in the maximally suppressed cortisol levels (0400h), nor in the rise of cortisol levels thereafter. Thus, cortisol production was equally inhibited by metyrapone in both groups. A significant difference in ACTH levels was already present at 0400 h and lasted throughout the MT. This clear distinction between children with CAI and those without was depicted in Figure 1. In our study, the prevalence of CAI in patients with PWS was 60%. Although we do not agree with a cut-off level of 18 pmol/l, the prevalence would still be ~30%. We are now faced with a problem of incorporating this knowledge into clinical care. We agree with Carrel and Allen that ACTH tests may not be an option for diagnosing patients with partial adrenal insufficiency. The MT is not always a practical instrument and we hope that future research will find a test that will equally distinguish between patients with and without CAI. Awaiting these results, because of the high prevalence of CAI in PWS, we believe that patients should be treated with a stress-dose of cortisol during illness, unless proven otherwise by a MT. References 1. Giordano R, Picu A, Bonelli L, Balbo M, Berardelli R, Marinazzo E, Corneli G, Ghigo E, Arvat E 2008 Hypothalamus-pituitary-adrenal axis evaluation in patients with hypothalamo-pituitary disorders: comparison of different provocative tests. Clin Endocrinol (Oxf) 68:935-941 2. Feek CM, Bevan JS, Ratcliffe JG, Gray CE, Blundell G 1981 The short metyrapone test: comparison of the plasma ACTH response to metyrapone with the cortisol response to insulin-induced hypoglycaemia in patients with pituitary disease. Clin Endocrinol (Oxf) 15:75-80 3. Oelkers W 1996 Adrenal insufficiency. N Engl J Med 335:1206-1212 4. Steiner H, Bahr V, Exner P, Oelkers PW 1994 Pituitary function tests: comparison of ACTH and 11-deoxy-cortisol responses in the metyrapone test and with the insulin hypoglycemia test. Exp Clin Endocrinol 102:33-38
Interpreting central adrenal insufficiency in PWS 22 June 2008
Aaron L. Carrel, MD University of Wisconsin Children's Hospital, David B. Allen, MD Send letter to journal: Re: Interpreting central adrenal insufficiency in PWS alcarrel{at}wisc.edu Aaron L. Carrel, et al.	We applaud de Lind van Wijngaarden et al. (1) for addressing the possible contribution of central adrenal insufficiency (CAI) to increased mortality in children with Prader-Willi Syndrome (PWS). The authors report a 60% occurrence of CAI (ACTH response <33 pmol/liter after single-dose metyrapone). Given other known hypothalamic-pituitary (HP) disturbances associated with PWS, this is certainly a plausible finding. However, given its potentially profound clinical care implications, some points warrant additional scrutiny and clarification. First, diagnostic accuracy of the metyrapone tests is strongly influenced by cut-offs for both ACTH and 11-deoxycortisol (11-DOC). In patients with known HP disorders, a post-metyrapone ACTH cut-off of ~18 pmol/liter showed sensitivity and specificity that minimized misclassification of subjects determined to be ACTH-deficient or normal by insulin tolerance testing (ITT) (2). Raising the threshold to 33 pmol/liter increased false positive rates without improving sensitivity. An 11-DOC cut-off of 144 nmol/liter showed the highest sensitivity and specificity (2) considerably less than the 200 nmol/liter level referred to by de Lind van Wijngaarden et al. as indicative of adrenal atrophy. Thus, cut-offs used may have resulted in false-positive identification of CAI in some children. Second, examination of 0730 11-DOC and cortisol values post-metyrapone reveals both a higher mean level of cortisol and rise from 0400 levels in those identified as CAI. Assuming equal inhibitory effect of metyrapone on cortisol synthesis in all patients, the “more normal” AM cortisol in those labeled with CAI seems counterintuitive. The possibility that some subjects showed a reduced response to metyrapone inhibition (and therefore reduced stimulus for endogenous ACTH release) should be considered. Third, Figure 1 creates the impression that subjects are clearly divided in CAI and non-CAI groups. We suspect, rather, that this appearance is an artifact of segregating the data of all subjects above a pre-determined threshold (in this case, ACTH levels >33pmol/liter) from those below, and then reporting the medians of these two groups. Clinical accuracy might be enhanced by including individual data points at each time, likely revealing the more continuous data variation. These criticisms notwithstanding, we concur that CAI should be considered a possible contributing factor to some cases of sudden death in children with PWS. The difficult question is how to incorporate awareness of this possibility into clinical recommendations that are useful without being overly burdensome. Routine testing with metyrapone is problematic, given its limited availability and the need for overnight hospitalization. Outpatient low-dose (1 mcg) ACTH stimulation testing, with a suitable cut-off of 480 nmol/liter, is an alternative that approximates the accuracy of metyrapone testing in patients with known HP dysfunction (2). In either case, it is uncertain and best, doubtful at worst, that normal results of such testing would fully assuage the concerns regarding varying degrees of CAI. Consequently, a consequence of this report and new practical reality may be that, while more is learned about CAI in children with PWS, it will be incumbent upon care providers to determine and provide reasonable guidelines for administration of glucocorticoid for stress or illness. References 1. de Lind van Wijngaarden RF, Otten BJ, Festen DA, Joosten KF, de Jong FH, Sweep FC, Hokken-Koelega AC. 2008 High Prevalence of Central Adrenal Insufficiency in Patients with Prader-Willi Syndrome. J Clin Endocrinol Metab 93:1649-1654 2. Giordana R, Picu A, Bonelli L, Balbo M, Berardelli R, Marinazzo E, Cornell G, Ghigo E, Arvat E. 2008 Hypothalamus-pituitary-adrenal axis evaluation in patients with hypothalamic-pituitary disorders: comparison of different provocative tests. Clin Endocrinol 68:935-941