BIOMEDICINE PHARMACOIHERAPY
ELSEVIER
Biomedicine & Pharmacotherapy58 (2004) $99-S 106 www.elsevier.com/locate/biopha
Circadian hypo- and hyper-amplitude-tension (CHAT) associated with putative pheochromocytoma and 12-hourly phenoxybenzamine treatment L.D. Sabath a, Germaine Corn61issen b,*, Franz Halberg b,*, Alexander Stoynev c, Ognian Ikonomov d, Kuniaki Otsuka e oDepartment of Medicine, University of Minnesota, Minneapolis, MN 55455, USA bHalberg Chronobiology Center, University of Minnesota, Minneapolis, MN55455, USA CTransport Medical Institute, Sofia, Bulgaria dwayne State University School of Medicine, Detroit, MI, USA eTokyo Women's Medical University, Daini Hospital, Tokyo, Japan
Abstract A 26-year-old white woman had an ablated sino-atrial node and ventricular pacemaker as an unusual feature of a pheochromocytomacompatible history. Her status quo included, on three occasions, elevated 24-hour urinary epinephrine and metanephrine excretion. She monitored her blood pressure (BP) and heart rate (HR) at 15- to 60-minute intervals over several days, with interruptions, before and after the institution of 10 mg phenoxybenzamine per os every 12 hours (between 7:30 and 8:00 and between 19:30 and 20:00), with continued monitoring over several months. Her data were summarized for consecutive 3-day intervals by sphygmochron. Circadian parameters and original data are compared with gender- and age-specified reference values, yielding also non-parametric endpoints, such as the percentage time elevation, the extent of excess, and the timing of excess, that all can be acceptable for some days but unacceptable for other days. In her broader time structure, or chronome, cosinor analyses revealed a prominent and statistically significant circadian rhythm in BP and HR before and during the 12-hourly therapy. The 12-hour component of BP was more prominent during therapy than prior to it. A statistically significant decreasing trend occurred before therapy, and recurred during treatment. Chronomically interpreted monitoring revealed: 1) the persistence of a statistically significant circadian rhythm during 12-hourly phenoxybenzamine treatment; 2) days-long changes in BP MESOR, the duration of which could not be previously determined based on spotchecks; 3) changes in the circadian amplitude of BP, which can be either very small or very large, compatible with the diagnosis of intermittent circadian hyper-amplitude-tension (CHAT); and 4) a very wide range of BP and HR values, so that occasional (casual) measurements fail to convey the dynamics that may underlie this infrequently found clinical condition of an elevated catecholamine excretion compatible with a pheochromocytoma. All findings support the need for long-term monitoring of BP and HR that may account for controversy in earlier publications. © 2004 Elsevier SAS. All rights reserved. Keywords." Pheochromocytoma;Blood pressure; Heart rate; Circadian rhythm; CHAT; Chronobiology;Phenoxybenzamine;Alpha-receptorblockade; Alphaadrenergic blockade; Circadian override of 'blockade'
1. Introduction S y m p a t h o m i m e t i c amine-secreting p h e o c h r o m o c y t o m a is notoriously difficult to diagnose. In some estimates, the *Corresponding author E-mail address: corne001 @umn.edu and/or
[email protected] (L.D. Sabath) © 2004 Elsevier SAS. Tous droits r6servds.
m o s t c o m m o n setting in w h i c h d i a g n o s i s is m a d e is at autopsy [1,2]. A m o n g the numerous clinical manifestations of excessive catecholamine secretion is an elevated BP. Virtually all of the manifestations of p h e o c h r o m o c y t o m a are episodic in the sense of intermittent, presumably because any time patterns of secretions from the tumor have not been m a p p e d for rhythms with multiple frequencies and even
S100
L D. Sabath et aI. / Biomedicine & Pharmacotherapy 58 (2004) $99-S106
broader time structures, or chronomes [3,4]. For the same reason, among others [5,6], the value of a random serum measurement of free catecholamines as a means of diagnosis is generally discouraged, except for plasma free metanephrines [7], whereas measurement of free catecholamines, and their metabolites, in 24-hour urine collections, is encouraged and is the current standard means of making the diagnosis. Obviously, the 24-hour urine collection 'summarizes' the events of the preceding 24 hours, whereas a single determination of catecholamines in blood gives only information about a single point in time, the way a single BP measurement does. In the present report, we demonstrate, with around-the-clock BP measurements, by use of a CO 2 pressure-driven automatic device [8], the persistence in the presence of high catecholamine excretion before and during treatment of a circadian rhythm in BP, which was a topic of debate when it was based on days, rather than week-long monitoring spans [8-11]. We then describe, in a BP monitoring of 2 months, the association of excessive catecholamine excretion with intermittent circadian hyper-amplitude-tension (CHAT) [12,13], a BP overswinging which carries a very high risk of stroke, higher than an elevated BP itself [14], irrespective of whether the chronome-adjusted average BP - M E S O R , M (midline-estimating statistic of rhythm) is or is not within acceptable gender- and age-matched peergroup-derived limits.
2. Case report A 26-year-old white woman experienced the first symptoms eventually related retrospectively to her putative pheochromocytoma at 21 years of age. Episodic sensations of palpitations were eventually followed by episodes of fainting, which led to cardiological and electrophysiological evaluations. On March 15, 1991, a radiofrequency treatment was performed with the intent to ablate the atrial focus responsible for her tachycardia. During the ensuing week, the patient continued to have recurrent (daily) tachycardia with ventricular rates up to 190 bpm, the use of beta-adrenergic blocking agents, calcium antagonists, and type 1C antiarrhythmic agents notwithstanding. On April 11, 1991, she underwent an AV nodal-His bundle ablation using radiofrequency energy. This procedure required the placement of an AV sequential dual chamber rate-responsive permanent pacemaker. On July 17, 1992, the patient returned to the operating room to have her pacing system revised due to chronic infection. A single chamber ventficular rate-responsive (VVIR) pacemaker was then implanted. The patient remains pacemaker-dependent with an idioventricular escape rhythm of 40. When the possibility of a pheochromocytoma was eventually considered a random serum determination showed epinephrine values that were acceptable in the context of being within their reference range and minimally elevated
norepinephrine concentration. Two different subsequent 24hour urine collections firmly established extremely high epinephrine and elevated metanephrine concentrations. Norepinephrine and normetanephrine concentrations were acceptable. Dopamine, vanillylmandelic acid and homovanillic acid were all within their reference range. It was concluded that the patient had a putative pheochromocytoma of the type secreting epinephrine. During this time of establishing a tentative diagnosis, i.e. April 16-19, 1993, while conducting studies to localize the tumor, the patient's BP and HR were monitored and analyzed by parametric and nonparametric methods described elsewhere [15]. It was of interest to see whether a decrease in overall BP operating average in time plots occurred prior to the start of treatment starting on April 18 around 07:30 (consisting of 12-hourly 10-rag doses of phenoxybenzamine), and, if so, whether a decrease continued on treatment. For this purpose, the span prior to treatment was subdivided into two subspans of approximately one day. In order to take two non-overlapping adjacent spans that did not include data collected after the start of medication, their length had to be restricted to 19-20 hours each; the first data subspan after the initiation of treatment lasted about 25 hours. During the 2 days prior to treatment, an about 3- to 4-hour gap occurred in each subspan. The total span during which data were actually collected in each subspan was about 16 hours, each gap not exceeding 5 hours. With these qualifications, a separate cosinor analysis [15] was repeated on each pretreatment subspan. Parametric tests [16] were then applied for a comparison of BP and HR between the first two subspans prior to the institution of treatment and between subspans 2 and 3 for an assessment of any prompt treatment effect. In order to reduce problems related to serial correlation, the tests were applied on data averaged over consecutive hourly intervals. From 16 April to 12 July 1993, the patient resumed around-the-clock monitoring. These data were analyzed in consecutive 3-day intervals and results were compared with gender-, age-, ethnicity-, geography- and social class-comparable peers.
3. R e s u l t s
Fig. 1 shows the time course and great variability of the patient's systolic (S) and diastolic (D) BP during the whole monitoring span. Before treatment, a decreasing trend for SBP and DBP was observed. Results from a model consisting of a 24-hour cosine and a linear trend, fitted concomitantly to the data, are summarized in Table 1. Tables 1 and 2 show first a statistically significant decrease in MESOR of BP between the first two subspans prior to the institution of medication and, second, no further decrease during the immediately following third subspan after initiation of treatment. There is also a decrease in cir-
S101
L.D. Sabath et al. / Biomedicine & Pharmacotherapy 58 (2004) $99-S106
(F, 26y) 180 Systolic BP
1
150
"~ 1 2 0
E
g 90
60
30
16~avr
26-avr
06-rnai 16-mai 26-rnai 05-juin 15-juin 25-juin 05-juil Time (calendar date in 1993)
Fig. 1. Time course of systolic and diastolic blood pressure during several months. Note an initial drop, which started before initiation of treatment, as documented in Tables 1 and 2. See also concomitant drop for several days of both systolic and diastolic blood pressure in the middle of the longer monitoring span. © Halberg.
cad±an amplitude b e t w e e n the first two subspans for BP but not for HR. No differences are seen in circadian acrophase, either b e t w e e n the first two subspans or b e t w e e n subspans 2 and 3.
A least-squares s p e c t r u m was also calculated for each v a r i a b l e d u r i n g e a c h s u b s p a n . A l t h o u g h the c i r c a d i a n r h y t h m is i n v a r i a b l y statistically h i g h l y significant during the third subspan on medication, an increase in p r o m i n e n c e of the 12-hour c o m p o n e n t is observed, perhaps due at least in part to the fact that medication was given at about 12-hour intervals. The 24-hour c o m p o n e n t was the most p r o m i n e n t for SBP and DBP, whereas for m e a n arterial pressure, the 24- and 12-hour c o m p o n e n t s were of about the same prominence, the second h a r m o n i c having a slightly larger amplitude than the fundamental. For HR, the 12-hour c o m p o n e n t is not statistically significant, whereas the 24-hour compon e n t is detected with a high level of statistical significance. These results show that although treatment m a y be administered at 12-hour intervals, overall, the circadian c o m p o n e n t remains the most p r o m i n e n t one, and is not detectably affected in acrophase b y the 12-hourly a d m i n i s t r a t i o n of phen o x y b e n z a m i n e , with the methods here used. A n a l y s e s of the patient's data during consecutive 3-day intervals reveal that the M E S O R of her S B P tends to be mostly but not invariably higher than the upper 95% prediction limit for clinically healthy w o m e n of her age (Fig. 2). The M E S O R of her D B P is mostly below the upper limit of acceptability, but tends to be 'high n o r m a l ' . Fig. 3 shows that she is occasionally diagnosed with C H A T . Fig. 4 shows
Table 1 Circadian characteristics of blood pressure and heart rate of patient with pheochromocytoma before and during treatment with phenoxybenzamine* Subspan
PR
P
MESOR + SE
Double amplitude ± SE
Acrophase*(95% CI)
Hour
Systolic blood pressure (mmHg) 1. (1: before Rx)
47
~x l ~i~' i ~~ 31 -mai
20-juin
" t O-juil
30-juii
Time (calendar date in 1993)
Fig. 3. Intermittent systolic and diastolic circadian hyper-amplitude-_tension (CHAT) in data summarized for successive 3-day intervals on a patient with intermittent excessive epinephrine excretion. © Halberg.
$103
L.D. Sabath et al./ Biomedicine & Pharmacotherapy 58 (2004) $99-Sl06
Table 3 Changes as a function of time in the circadian acrophase of blood pressure (BP) and heart rate (HR) of 26-year-old woman with putative pheochromocytoma, assessed in consecutive 3-day spans
(F, 26y) 120
MESOR 1oo ~ ~ 80
~
Acrophase (h:min)
O0
24h Double Amplitude Limits of acceptability
,,o ;
~,
j! },/
.... \ - . . . . . . . . . . . . . . . . . .
0
01-avr
.
21-avr
11-ma[
.
:-'TT-'i-
.
3t -mai
Systolic BP
Diastolic BP Heart rate
01-03
15:44
15:33
13:59
13-15
14:30
14:02
12:24
16-18
14:13
13:53
13:59
19-21
14:39
14:03
14:05
22-24
16:14
15:52
14:44
25-27
15:19
15:04
16:11
28-30
14:47
14:57
15:17
31-33
13:53
14:39
15:06
34-36
12:49
13:00
13:21
37-39
14:59
14:12
13:16
40-42
07:18
09:30
09:32
43-45
15:12
14:54
12:26
46-48
16:11
14:55
13:26
49-51
11:32
11:54
13:48
52-54
13:55
13:35
14:15
55-57
13:29
14:16
11:23
58-60
16:03
15:34
15:04
61-63
18:26
01:32
02:05
64-66
13:20
13:26
i4:05
67-69
04:00
19:37
03:42
70-72
16:26
15:35
19:28
73-75
15:48
15:12
17:38
76-78
15:13
15:32
14:59
79-81
15:34
15:32
07:06
82-84
15:16
15:45
14:54
85-87
14:37
14:59
09:55
Limits of acceptability (healthy peers matched by gender and a g e ) :
09:25-19:37 11:28-17:30 10:45-18:29
........
.
20-juin
Times (days from start of monitoring)
10-juil
30-juil
Time (calendar date in t993)
Fig. 4. Time course of heart rate in data summarized for successive 3-day intervals on a patient with intermittent excessive epinephrine excretion. Note elevated MESOR of heart rate during most 3-day spans and the occasional excessive circadian double amplitude of heart rate. © Halberg.
nal heart rate fall is not responsible for it, since the nocturnal blood pressure fall remained unchanged in patients undergoing cardiac pacing." The observations here indicate that the fundamental chronobiology of BP variation previously quantified in health [ l 5] can persist in the likely presence of a pheochromocytoma, and that this fundamental circadian rhythmicity [15, 19-21] can override the effect of an alphaa d r e n e r g i c r e c e p t o r - b l o c k i n g drug, p h e n o x y b e n z a m i n e , when the observations are based upon a few days of aroundthe-clock monitoring only. In consecutive 3-day spans over several months, however, opposite patterns are seen both by the naked eye and by t i m e - m i c r o s c o p i c analysis, so that prior different values can be readily reconciled. Figs 2 and 3 show, by sphygmochron analysis, M E S O R - n o r m o t e n s i o n vs M E S O R - h y p e r t e n s i o n , and C H A T vs the a b s e n c e o f C H A T during different spans monitored in the same subject. A circadian amplitude that is too high can precede one that is extremely low. The months-long monitoring shows, overall, that one encounters in the same patient all of the contrary reports concerning circadian rhythmic behavior. For the preChrousos-Pacak history of pheochromocytoma [7], the interested reader is referred to T o r m e y [22] and the developments of magnetic resonance imaging and computed t o m o g r a p h y for i m a g i n g for tumor localization. W e here report temporal imaging of chronomes in the service of following the functional dynamics of BP variability in a case of intermittent epinephrine excess. Alesci, Chrousos and Pacak discuss the a p p r o a c h by g e n o m i c m e d i c i n e to endocrine hypertension [23]. Whatever the underlying mechanisms may be, long-term monitoring reveals chronomic (time structural) [4] alterations that m a y point to alterations in the network of pinealhypothalamic-pituitary-adrenal interactions in the coordination of BP [24-26] and that point to the h y p o t h a l a m i c suprachiasmatic nuclei (SCN). Indeed, for Stoynev et al. [27], a zero-amplitude test for a group of SCN-ablated rats
did not allow rejection of the 'no circadian rhythm' assumption, whereas for a group of sham-operated (control) rats, this was readily the case (Fig. 5). Lenders et al. [28] have noted that pheochromocytomas are dangerous tumors and noted that false-negative tests can occur in asymptomatic patients and recommend measurements of plasma concentrations of free metanephrines and 24-hour urinary outputs o f f r a c t i o n a t e d n o r m e t a n e p h r i n e and m e t a n e p h r i n e that "almost always reveal the tumor" [28].
S 104
L.D. Sabath et al./Biomedicine & Pharmacotherapy 58 (2004) $99-S106
Table 4 Nonparametric results of blood pressure and heart rate profiles of 26-year-old woman with putative pheochromocytoma, assessed in consecutive 3-day spans Percent time elevation (%)
Amount of excess HBI (ram Hg x h) TCI (bpm x h) (during 24 hours)
Timing of excess (hr:min)
Times (days from start of monitoring)
SBP
DBP
HR
SBP
DBP
HR
SBP
DBP
HR
01-03
3.9
0.0
0.0
2
0
0
18:29
--
--
13-15
54.6
9.2
0.0
75
2
0
22:40
14:51
--
I6-18
36.2
0.9
0.0
18
