- Email: [email protected]
Control of breathing in patients with severe hypothyroidism
Control of Breathing in Patients With Severe Hypothyroidism ROBERTODURANTI,M.D.,RICCARDOG.GHERI,M.D.,MASSIMOGORINI,M.D., FRANCESCOGIGLIOTTI,M.D.,ALESSANDROSPINELLI,M.D.,ALESSANDRAFANELLI,M.D., GIORGIOSCANO,M.D., f/orence,/ta\y
PURPOSE: Hypothyroid
patients have been reported to have a blunted ventilatory response to carbon dioxide stimulation. However, previous data did not clarify the localization of abnormalities responsible for that disorder. The present investigation was aimed at evaluating to what extent central (neural) and/or peripheral (muscular) factors are involved in the abnormalities of the ventilatory control system in hypothyroid patients. PATIENTSANDMETHOD~: Westudied13patients with severe hypothyroidism before and after 6 to 9 months of replacement therapy; 7 age- and sex-matched normal subjects were also studied as a control. In each subject, we assessed (1) inspiratory muscle strength by measuring maximal inspiratory pressure (MIP), and (2) respiratory control system during a carbon dioxide rebreathing test by measuring minute ventilation (VE), tidal volume (VT), mean inspiratory flow (VT/TI), and electromyographic (EMG) activity of the diaphragm (Ea) and intercostal (E& muscles. RESULTS: Compared with the normal control group (Group C), patients exhibited similar MIP, and similar VE and EMG response slopes to carbon dioxide. However, evaluating individual VE response slopes, we were able to identify two subsets of patients: Group A (six patients) with low VE response (less than mean - SD 1.65 of Group C) and Group B (seven patients) with normal VE response. Compared with both Groups B and C, Group A exhibited significantly lower VT/TI, E& and Eht response slopes; the difference between Groups B and C was not significant. Six patients (two from Group A and four from Group B) exhibited low MIP values compared with that in Group C. After replacel
From the lstituto di Clinica Medica III (RD, AS, AF, GS). Unita di Endocrinologia-Dipartimento di Fisiopatologia Clinica (RGG). Universita degli Studi di Firenze, and Fondazione Pro-Juventute Don C. Gnocchi (MG, FG). Florence, Italy. This study was supported by grants from the Minister0 dell’Universita e della Ricerca Scientifica e Tecnologica of Italy, Rome, Italy. Requests for reprints should be addressed to Roberto Duranti, M.D., lstituto di Clinica Medica Ill, Universita degli Studi di Firenze, Viale G.B. Morgagni 85. 50134 Florence, Italy. Manuscript submitted May 4, 1992, and accepted in revised form November 19, 1992.
ment therapy, (1) VE, VT/TI, and Ea response slopes increased significantly in Group & and (2) MIP increased, but not significantly in patients with low MIP. CONCLUSIONS: We conclude that: (1) In patients with severe hypothyroidism, the ventilatory control system may be altered at the neural level, as indicated by a blunted chemosensitivity; (2) Impaired respiratory muscle function does not seem to play a major role in the decreased ventilatory response to carbon dioxide stimulation; (3) Replacement therapy appears to normalize the response to hypercapnic stimulation, but not respiratory muscle strength.
R
espiratory manifestations are rarely a major problem in hypothyroid patients, even if several abnormalities in respiratory function may be found in these patients. Fatigue and dyspnea on exertion [l], disordered breathing during sleep [2], impaired ventilatory response to hypoxia and hypercapnia [3,4], decreased inspiratory muscle force [5], and severe diaphragmatic dysfunction [6,7] have been reported. Sometimes, ventilatory involvement is life-threatening, since carbon dioxide retention, respiratory failure, and coma may ensue [l&9]. The reasons why hypothyroid patients can develop alveolar hypoventilation have not been clarified so far. It has been suggested that depression of respiratory centers causing a decreased inspiratory neural drive may be involved [l-3]. On the other hand, decrease in inspiratory muscle force, i.e., muscle weakness, has also been thought to play a role [6,7,10]. In fact, the severe myopathy, described in hypothyroid adults 111-131, may also affect respiratory muscles [14,15]. Recently, Ladenson et a2 [3] have investigated the prevalence of impaired hypoxic and hypercapnic ventilatory responses and the clinical and chemical parameters that might predict the presence of abnormal ventilatory control in hypothyroid patients. These authors reported that only a subset of patients may have a blunted respiratory response to either hypercapnic (34%) or hypoxic (27%) stimulation. Hypothyroid women and patients with higher values of serum thyroid-stimulating hormone (TSH) were significantly more likely to have impaired ventilaJuly 1993
The American
Journal
of Medicine
Volume
95
29
CONTROL
OF BREATHING
IN HYPOTHYROIDISM
/ DURANTI
ET AL
TABLE I Anthropometric, Biologic, and Clinical Data of the 13 Hypothyroid Patients Before and After ReplacementTherapy*
Subject
Age (Y)
p#;0 Before
After
TSH (mu/L) Before After
&i-h, Before After
Tab
Mab
Diagnosis
1
55
83.2
84.0
264
0.70
4.20
10.20
1:1,280
1:80
Chronic autoimmune thyroiditis
2
58
98.6
95.2
>64
1.30
2.10
10.80
1:2,500
1:80
Chronic autoimmune thyroiditis
3
57
82.1
89.4
>64
1.00
0.86
10.50
1:1,280
1:4Q
Chronic autoimmune thyroiditis
4
51
93.4
85.3
>64
1.50
3.14
8.20
Negative
1:80
Chronic autoimmune thyroiditis
5
62
96.2
89.5
>64
0.04
1.77
10.90
Negative
1:80
Chronic autoimmune thyroiditis
6
66
109.0
96.9
>64
2.20
1.10
10.80
1:160
1:80
Chronic autoimmune thyroiditis
7
40
105.3
102.0
>64
0.30
0.75
9.10
Negative
1:40
Chronic autoimmune thyroiditis
8
69
124.3
124.3
>64
9.30
1.00
8.90
1:1,280
1:40
Chronic autoimmune thyroiditis
9
49
111.2
104.9
>64
0.60
2.50
9.20
Negative
1:40
Postablative (radioiodine)
10
44
102.6
100.3
>64
1.10
0.58
6.60
Negative
Negative
Primary idiopathic
11
30
98.8
86.6
>64
1.00
0.80
9.70
1:1,280
Negative
Primary idiopathic
12
57
102.0
96.8
>64
3.10
1.10
8.70
Negative
1:lO
Primary idiopathic
13
64
92.3
96.0
>64
8.30
3.70
9.10
1:80
1:160
Chronic autoimmune thyroiditis
Mean (patients) SEM
54.0 3.10
99.9 3.20
96.3 2.90
>64
2.4 3.0
1.84 1.07
9.37 1.27
Mean (normals) SEM
“E
102.0 3.3
iormal rangesare: serum TSH, 0.6 io 4.6 mu/L; T4,4.5 to 12 kgldt; Tab and Mab, not present. Body weight is expressed as percentageof ideal weight (IW).
tory responses. However, in that study [3] as in a previous one [4], it was not possible to ascertain the localization of the lesion(s) responsible for disordered ventilatory responsiveness; neural respiratory drive was assessed in terms of minute ventilation (VE) and mean inspiratory flow (VT/TI). In fact, in patients with ventilatory disorders, mechanical limitation of the lung and/or chest wall prevents VE and VT/TI from accurately reflecting respiratory center output. Therefore, neither VE nor VT/TI allows ascertainment of whether central or peripheral abnormalities are involved in the decreased ventilatory response observed in patients with hypothyroidism. The relative role of central (neural) and peripheral (muscular) abnormalities might be assessed by recording the electromyographic (EMG) activity of the diaphragm (Edi), an index of the neural drive to the respiratory muscles [16-231, and maximal inspiratory pressure (MIP), an index of inspiratory muscle force. The current investigation was undertaken in patients with severe hypothyroidism, before and after replacement therapy, to evaluate the existence of abnormalities in the ventilatory control system and, if any, to what extent central and/or peripheral factors are involved. 30
July
1993
The American
Journal
of Medicine
Volume
95
PATIENTS AND METHODS Patients Thirteen female patients with clinical and laboratory signs of severe primary hypothyroidism (overt hypothyroidism by the classification of Evered and Hall [24]) were studied before and after thyroid hormone replacement therapy. A group of seven age- and sex-matched healthy normal subjects was also studied as a control. Informed consent was obtained from patients and normal subjects. The anthropometric, biologic, and clinical data of patients are provided in Table I. All were outpatients and were examined at the Endocrinological Unit of the University of Florence. None of them had evidence of pulmonary, cardiac, or neurologic diseases on the basis of clinical evaluation, routine laboratory testing, chest radiography, and electrocardiography. None of the patients fulfilled the diagnostic criteria of asthma, chronic bronchitis, or emphysema according to the American Thoracic Society [25], and none of them exhibited dyspnea at rest or during exercise. Smokers were excluded from the study. Thyroid Function Study Thyroid function was evaluated on the day of the
CONTROL
study by measuring serum TSH concentration and total serum thyroxine (Td). The hormonal and immune determinations were performed with “routine” methods. The measurement of TSH was performed by the Allegro TM HS-TSH Immunoassay System kit (Nichols Institute, San Juan Capistrano, CA [normal values: 0.6 to 4.6 mu/L]). Tq was determined by radioimmunoassay methods (ARIA HT Tq system [normal values: 4.5 to 12 pg/dL], Becton Dickinson Immunodiagnostic, Salt Lake City, UT). Thyroglobulin antibodies (Tab) were determined by a passive hemoagglutination test (Wellcome Diagnostics, Dartford, England); microsomal thyroid antibodies (Mab) were detected by an immunofluorescence method using cryostatic sections of human thyroid. Pulmonary Function Study Routine spirometry was performed as previously described [21]. The normal values for lung volumes are those proposed by the European Community for Coal and Steel [26]. MIP was measured using a differential pressure transducer (Statham SC 1001, Gould Inc., Oxnard, CA) as already reported [27]. After baseline routine testing, ventilatory pattern and respiratory drive were evaluated during a carbon dioxide rebreathing test. Each subject, placed in a comfortable supine position, breathed through a mouthpiece and a Fleisch (no. 3) pneumotachograph (Beckman, Geneva, Switzerland) attached to a one-way valve that separated the inspiratory from the expiratory line. Flow signal was integrated to yield tidal volume (VT). Inspiratory time (TI) and respiratory cycle duration (TT) were measured from the volume tracing. VT/TI and VE (VE = VT a 6O/TT) were also calculated. Expired carbon dioxide was continuously measured at the mouth by an infrared carbon dioxide meter. The EMG activity of the respiratory muscles was recorded from the second parasternal intercostal (E& and diaphragm (Edi) muscles via large surface electrodes; the technique employed to record EMG activity has previously been described in detail [21,27,28]. Briefly, the differentially amplified and filtered muscle action potentials were full wave rectified and “integrated” over time (time constant 100 milliseconds) in order to provide a measure of changes in average electrical activity as a function of time, referred to as “moving time average” [19,20]. Inspiratory activity was quantified both as peak of activity (directly measured in arbitrary units) and as rate of increase in activity (obtained by dividing the peak activity by the inspiratory time). Because of the variability of the impedance between diaphragm and electrodes, absolute values (mV) are not comparable in different subjects.
OF BREATHING
IN HYPOTHYROIDISM
/ DURANTI
ET AL
Therefore, in order to obtain a reference value, EMG activity was measured while the subject was connected to the pneumotachograph and performed an inspiratory maneuver breathing in up to the total lung capacity (TLC) [16,27]. This maneuver was repeated at least three times, and in each subject, both inspiratory volume and peak of EMG activity at TLC were closely reproducible (less than 5% variability). The mean level of this EMG activity was taken as the reference, and all successive measurements have been expressed as a percentage of the reference value obtained at TLC. The output of the carbon dioxide analyzer, flow signal, integrated flow signal, and the moving time average (Edi and E& were continuously recorded on a multichannel chart recorder. After a period of 5 to 10 minutes, which allowed the subjects to adapt to the circuit, as shown by the stability of carbon dioxide recording, each subject underwent a carbon dioxide rebreathing test following the procedure recommended by Read [29]. A gas mixture (7% carbon dioxide, 93% oxygen) was inhaled over 3 to 5 minutes from an 8-L bag. The response to carbon dioxide stimulation was evaluated in terms of slope of the regression line of studied variables versus end-tidal partial pressure of carbon dioxide (PETIT,). In each normal subject, the rebreathing test was repeated on two to three different days, whereas in patients, it was duplicated on the same day with an interval of 60 minutes between each test. Study Protocol Respiratory and thyroid function tests were performed prior to the beginning of therapy. All the patients underwent thyroid replacement therapy with L-thyroxine (L-Td); the patients initially received 0.025 mg daily. The dose of L-T4 was increased by 0.025-mg increments up to a maximal dose of 0.150 mg. After 6 to 9 months, when euthyroid as judged on the basis of both clinical and thyroid function tests, patients were reexamined as before therapy. Data Analysis For each rebreathing run, changes in VE, VT, VT/TI, Edi, and Eint were plotted against corresponding values of PET~~, and subjected to leastsquare linear regression analysis. To obtain a normal reference range, an average slope value of the runs carried out by each normal subject was considered. In no case was the response exhibited on 1 day twice greater or less than half of the response obtained on the other days. The mean slope for the two runs was calculated for each patient. Data were averaged for patients and normal subjects and are July
1993
The American
Journal
of Medicine
Volume
95
31
CONTROL OF BREATHING IN HYPOTHYROIDISM / DURANTI ET AL
TABLE II Pulmonary Function Data and Maximal lnspiratory Pressuresof the 13 Hypothyroid Patients Before and After Replacement Therapy*
vc+
Before ’ After
MIP km Hz01 Before After
Before
After
1
104.8
102.3
102.7
103.5
75.0
85.0
91.9
95.5
101.9
102.6
80.2
85.5
120
118
2
114.0
130.7
87.7
85.3
88.1
81.9
98.6
107.5
114.9
125.5
81.7
78.2
80
122
3
112.3
113.4
111.1
117.5
92.2
87.7
112.6
111.2
115.9
120.3
84.6
86.9
106
90
4
95.1
102.6
110.2
108.9
88.6
92.6
102.3
96.5
103.4
94.6
75.0
76.1
54
82
5
119.9
125.2
93.4
92.9
92.7
92.1
103.7
102.5
125.3
121.7
83.9
77.8
108
108
6
97.6
103.4
124.6
91.7
120.8
97.9
98.4
95.9
88.3
97.8
72.0
75.0
52
62
7
99.3
98.5
96.3
96.0
91.1
92.1
96.2
96.2
88.3
93.4
75.2
80.0
48
64
8
97.7
85.7
120.5
146.1
127.6
127.1
109.6
116.5
91.1
84.8
73.5
77.6
70
68
9
115.3
115.1
92.5
92.0
73.6
84.5
104.9
103.8
112.5
110.1
81.4
80.2
94
90
10
96.5
99.0
101.6
117.7
104.8
107.1
98.0
110.1
119.8
117.1
86.5
82.8
52
45
11
104.3
103.1
96.0
122.6
90.3
111.1
101.0
106.2
109.9
110.9
90.0
92.1
88
82
12
97.9
104.6
126.7
112.1
113.7
107.3
96.9
114.1
98.1
115.0
73.3
75.6
50
88
13
95.0
94.7
113.0
112.7
88.8
104.8
95.0
94.6
85.8
94.8
72.5
80.0
52
60
105.8
105.3
95.9
97.8
103.9 2.1
104.2 3.6
106.8 3.6
79.3 1.7
80.6 1.4
74.1 6.8
83.0 6.4
103.8
Mean (normals) SEM
116.2
2.4 2.8
106.0
3.4
3.5
3.4
102.4
3.5
4.5
3.6
98.5
3.8
TLC+ Before After
F%vc
Subject
Mean (patients) SEM
RV+ Before After
FE&+ Before After
FRC+ Before After
100.6
1.6 103.7
2.2
99.7
2.7
81.2
1.9
83.0 5.8
: = vital capacity; FRC = functional residual capacity; RV = residual volume; TLC = total lung capacity, FEVl = forced expiratory volume in 1 second; MIP = maximal inspiratory pressure. *individual data and mean values r SEM are reported. tpercent of predicted value.
presented as the mean f SEM. Comparisons between patients and normal control subjects were performed by Mann-Whitney U test for unpaired samples; significance of variations induced by replacement therapy was evaluated by Wilcoxon rank tests for paired samples. VE response slope and MIP values more than SD * 1.65 below the mean value for a normal control group were considered to be significantly reduced. Comparisons among the three groups (Groups A and B patients and normal control group) were made by one-way analysis of variance (ANOVA); Newman-Keuls tests were used to assess the significance of differences between pairs of groups where ANOVA showed statistically significant differences. Differences were considered significant when p was less than 0.05.
RESULTS Anthropometric, biologic, and clinical characteristics of the patients before and after replacement therapy are summarized in Table I. A hypothyroid condition was ascertained on the basis of both clinical and biologic data. All patients presented with high values of TSH and low values of TJ; significant levels of serum Tab and Mab were observed in 7 and 32
July 1993 The American Journal of Medicine
Volume 95
11 patients, respectively (Table I). After replacement therapy, 11 patients exhibited normal values of TSH and Tq, and were judged to be euthyroid (Table I). Two patients (Patients 8 and 13), who showed slightly elevated TSH values and normal T.J values, were considered to be euthyroid (preclinical hypothyroidism by the classification of Evered and Hall [24]). Study Before Treatment The average body weight of patients, expressed as percentage of the ideal weight [30], was 99.9% f 3.2%. Patients invariably exhibited normal spirometric values (Table II). No patient appeared to retain carbon dioxide; PETIT, was 40.6 f 0.77. As a mean, no significant difference was observed in terms of MIP (Table II) between patients and the normal control group (Group C); however, in 6 patients (Patients 4,6,7,10,12, and 13, see Table II), MIP values were less than mean - SD * 1.65 of the values calculated for Group C. During carbon dioxide rebreathing, as a whole, hypothyroid patients did not exhibit any significant difference in terms of VE, VT, VT/TI, Edi, and Eint response slopes to increasing carbon dioxide
CONTROL
1.6
OF BREATHING
IN HYPOTHYROIDISM
0.06
/ DURANTI
ET AL
0.08
22 q
G q
\ 1 1.2 73 0.6
Gm 0.06 : 1 0.04
=:
;8
If 3 0.4 5
g3 0.02 Q
, B 0.045 i. 7=q 0.03
”0”
3r” 0.015 \ 5
0
0
a
Figure 1. Slopes of the relationships of VE, VT, VT/Tl, and Edi and Ei”t against corresponding PETco, values, during carbon dioxide rebreathing, in the 13 hypothyroid patients, before and after therapy, and in the normal control group. Values are mean i SEM.
Normal subjects
TABLE III Slopes of the Relationshipsof VE,VT,VT/TI, Edi,and EintVersus Corresponding PETCO,Values During Rebreathing Test in Groups A and B Patients Before ReplacementTherapy, and in Normal Control Group* AVE/APETCO
AVT/APETCO~
AWTITI)/APETCIJ~
No.
([L/mini/mm i&I
(L/mm Hg)
UL/sl/mm Hg)
Group A patients
6
Group B patients
7
Normal subjects
7
0.52 (0.03) 1.24 (0.17) 1.22 (0.11)
0.036 (0.008) 0.065 (0.011) 0.066 (0.007)
0.018 (0.003) 0.041 (0.006) 0.043 (0.006)
Subjects
One-way ANOVA F P
9.42 <0.002
3.04 <0.07
AEdi/APETCO ([%TLC/sl/mmkg) 0.482 (0.135) 1.57 (0.36) 1.46 (0.24) 4.50 <0.027
AEdAPETCo
([%TLC/sl/mmkg) 0.36 (0.09) 0.99 (0.25) 1.55
(0.39) 4.04 (0.037
Newman-Keuls Group A versus normals Group B versus normals Group A versus Group B
I
VE = minute ventilation; VT = tidal volume; WTI = mean inspiratory flow; PETCO,= end-tidal partial pressure of carbon dioxide; Ed, = EMG activity of the diaphragm; E,,t = EMG activity of the parasternal intercostal muscles. Ediand Elntwere quantified as the rate of Increase of EMGactivity obtainedby dividing the peak of inspiratory activity by inspiratory time (TI). *Valuesare mean i- SEM.
compared with the normal control group (Figure 1); nevertheless, as indicated in Figure 1, there was a tendency for these variables to be lower in patients than in normal subjects. Assessing the individual VE responses to carbon dioxide, we were able to distinguish two subsets of patients: one (Group A, six patients: Patients 1 through 4,7, and 11) with a blunted VE response slope (less than mean - SD . 1.65 of the values calculated for the normal control group) and the other (Group B, seven patients: Pa-
tients 5,6,8 through 10,12, and 13) with a normal VE response slope. VE, VT/TI, Edi, and Eint response slopes to increasing PET~O, did significantly differ (one-way ANOVA) among Groups A, B, and C (Table III). Intergroup comparisons (Newman-Keuls tests) showed that VE, VT/TI, and Edi response slopes were significantly lower in Group A than in Groups B and C, whereas Eint was significantly lower in Group A than in Group C. In Group A, the VT response slope was lower compared with that in July
1993
The American
Journal
of Medicine
Volume
95
33
CONTROL
OF BREATHING
IN HYPOTHYROIDISM
/ DURANTI
ET AL
TABLEIV VE,VT,WTt,
Edi, and Eint for a PETCO, of 60 mm Hg During Carbon Dioxide Rebreathing in Normal Control Group*
I
No.
Group A patients
6
Group B patients
7
(L/min)
13.0 (1.24) 17.10
1.02 (0.10) 1.19
(0.20) 1.26 (0.13)
(0.91) 21.08 (2.29)
7
Normal subjects One-way ANOVA F P
Therapy,
0.43 (0.04) 0.75 (0.02) 0.72 (0.09)
5.74 (2.31) 13.95 (1.68) 12.48 (2.85)
10.2 (2.81) 12.47 (4.45) 15.98 (5.70) 0.39 CO.68
to,011
0.57 to.57
6.35 < 0.009
5.10 to.018
<0.05
NS
-co.05
to.05
5.89
Newman-Keuls Group A versus normals Group B versus normals Group A versus Group B
Before Replacement
and
VT/TI L/s)
VE
Subjects
in Groups A and B Patients
NS N”s”
<0.05
N”s”
NS
<0.05
K NS
*Values are mean f SEM. Same abbreviations as in Table III.
Figure 2. Individual slopes of the relationships of VE, VT, and VT/T1 against corresponding PETco, values, during carbon dioxide rebreathing, in the six hypothyroid patients with low ventilatory response to carbon dioxide stimulation (Group A), before (b) and after (a) replacement therapy. Mean values f SEM for the six patients, before (0) and after (A) replacement therapy, and for the normal control group (m) are also shown.
b
a
b
both Groups B and C, but it did not reach the level of statistical significance. No significant difference in any of the studied variables was observed between Groups B and C. At a PETIT, of 60 mm Hg, VE, VT/TI, and Edi significantly differed among the three groups (ANOVA, Table IV). Newman-Keuls 34
July
1993 The American
Journal
of Medicine
Volume
95
a
Figure 3. Individual slopes of the relationships of Edi and Ei,t against corresponding PETco, values, during carbon dioxide rebreathing, in the six hypothyroid patients with low ventilatory response to carbon dioxide stimulation (Group A), before (b) and after (a) replacement therapy. Mean values f SEM for the six patients, before (0) and after (A) replacement therapy, and for the normal control group (m) are also shown.
tests showed that VE was lower in Group A compared with Group C, and VT/TI and Edi were significantly lower in Group A compared with both Groups B and C. No difference was observed in any of the above variables between Groups B and C. Anthropometric, clinical, and biologic character-
CONTROL
OF BREATHING
IN HYPOTHYROIDISM
/ DURANTI
ET AL
TABLE V Slopesof the Relationships of VE, VT, VT/TI, Edi, and EintVersusCorresponding PETCO,Values During Rebreathing Testin Groups A and B Patients After ReplacementTherapy, and in Normal Control Group* AVE/APETCQ Subjects
No.
KL/minl/mm Hg)
Group A patients
6
Group B patients
7
Normal subjects
7
1.40 (0.21) 1.33 (0.22) 1.22 (0.11)
One-way ANOVA F P
0.22 to.80
AVT/APETCO,
A(VT/TI)IAPETCO,
&/APETCo2
AEdAPETCO
(L/mm Hg)
([L/sl/mm Hg)
(I%TLC/sl/mmHg)
U%TLC/sl/mmkg)
0.067 CO.0161 0.066 (0.007)
0.047 (0.007) 0.047 (0.007) 0.043 (0.006)
1.197 (0.230) 1.90 (0.63) 1.46 (0.24)
0.908 (0.118) 1.55 (0.39)
0.04 to.96
0.23 <0.80
0.69 co.52
1.24 to.31
*Values are mean f SEM. Same abbreviations as in Table III,
istics, lung volumes, and MIP did not significantly differ between Groups A and B. Only two Group A patients (Patients 4 and 7) showed low MIP values. Study After Treatment Patients were reexamined after 6 to 9 months of replacement therapy; as a whole, no significant changes in ideal weight, lung volumes, MIP (Table II), and PETITE response slopes (Figure 1) were found. In the six patients with low MIP values, MIP increased without, however, reaching the level of statistical significance (from 51.33 f 0.84 to 66.83 f 6.41). In Group A, compared with the pretreatment values, replacement therapy resulted in a significant increase in VE (p <0.03), VT/T1 (p <0.03), and Edi (p <0.03) response slopes; VT and Eint response slopes also rose, but the increase was not significant. Mean and individual response slopes in Group A, before and after replacement therapy, are depicted in Figures 2 and 3. When euthyroid, Group A did not differ compared with both Groups B and C (one-way ANOVA, Table V). In Group B, replacement therapy did not result in any significant ventilatory and EMG changes (Table V). At a PET~~, of 60 mm Hg, VE (p <0.03), VT/T1 (p <0.03), and Edi (p <0.03) significantly increased in Group A; the values no longer differed from those observed in both Groups B and C (ANOVA, Table VI). In the patients as a whole, the relationship of the VE response slope with the Edi response slope, obtained before and after treatment, was found to be significant (r = 0.62, p <0.025, n = 26). In contrast, the VE response slope did not relate to the MIP (r = 0.1, NS, n = 26).
COMMENTS Our data in female hypothyroid subjects are consistent with previous studies showing that patients with hypothyroidism may or may not demonstrate a low ventilatory response to increasing carbon dioxide [3,4]. In our study, 46% of the total population
r-~
TABLE VI VE, VT, W/T!, Edi, and Eintfor a PETCOof 60 mm Hg During Carbon Dioxide Rebreathing in GroupsA and fPatients After Replacement Therapy, and in Normal Control Group* VE
VT/TI
(L/s)
Subjects No. (L/min)
Edi
(%TLC/s) C%?;,s)
I Group A patients GroupB patients Normal subjects One-way ANOVA F P
6 7 7
20.88 (3.91) 22.26 (2.20) 21.08 (2.29)
1.15 (0.14) 1.27 (0.10) 1.26 (0.13)
0.82 (0.12) 0.82 (0.07) 0.72 (0.09)
14.4 (3.20) 16.08 (3.30) 12.48 (2.85)
12.41 (5.61) 13.01 (2.32) 15.98 (5.70)
0.07 to.93
0.20 co.82
0.36 to.70
0.35 to.71
0.16 to.85
I
*Valuesare mean f SEM. Same abbreviations as in Table Ill.
exhibited a low ventilatory response, a percentage very similar to that reported in the study of Ladenson et al [3], where only women were observed to exhibit a decreased ventilatory response to hypercapnic stimulation. The most important finding in our study is that the subset of hypothyroid patients with a low VE response slope to carbon dioxide stimulation also exhibited significantly lower VT/TI, Edi, and Eint response slopes. Two main factors are likely to play a role in the decreased hypercapnic VE response in hypothyroid patients: (1) impairment in neural function either at central (decreased neural respiratory drive) or peripheral (motor neuropathy) levels [1,3]; (2) respiratory muscle weakness, i.e., the failure to generate force, due to hypothyroid myopathy [6,7,10]. Impairment in Neural Function It has been previously supposed that neural factors play an important role in determining respiratory abnormalities in hypothyroid patients [1,3]. In this connection, moderate to marked reduction in the EMG response slope to carbon dioxide observed in patients with a low VE response slope (Figure 3) July
1993
The American
Journal
of Medicine
Volume
95
35
CONTROL OF BREATHING IN HYPOTHYROIDISM / DURANTI ET AL
indicates decreased neural activation of both diaphragm and intercostal muscles; possible reasons for this alteration may be either central brainstem dysregulation or peripheral neuropathy, or both. The hypothesis that central brainstem dysregulation contributes to the impairment in respiratory function in hypothyroid patients has been raised previously [2,3,31]. In this connection, Ladenson et al [3] observed that TSH concentration, but not Tq levels, predicted the occurrence of an abnormal ventilatory response, and accordingly, TSH secretion could reflect the impact of thyroid hormone deficiency on the central nervous system. We cannot contribute to this hypothesis owing to the lack of precise values of TSH when its concentration was greater than 64 U; thus, a relationship between TSH values and VE response to carbon dioxide stimulation could not be assessed. In agreement with the findings of Ladenson et al [3], we did not observe any significant relationship between hypercapnic ventilatory response and Tq values. Our data support evidence against a role of anthropometric, biologic, and clinical characteristics, given the lack of any difference in this regard between Groups A and B. In the present study, the association of a low VE response with a low EMG response suggests the possibility of a decreased responsiveness of respiratory centers to carbon dioxide stimulation. This conclusion may be correct if one considers EMG activity of the inspiratory muscles as a reliable index of neural inspiratory drive. A close correlation between changes in electrical activity of the phrenic nerve and the diaphragm has been reported in a dog, during both normal breathing and obstructed breathing [32]. Nevertheless, the use of either surface or esophageal EMG recordings to assess the neural drive to inspiratory muscles in humans has been thoroughly criticized [21-23,271. However, as we recently outlined [27], many data support the contention that the EMG can actually reflect the neural drive to the respiratory muscles, and that the slope of the “moving time average” of Edi is a reliable measure to assess neural inspiratory drive to the diaphragm in human subjects [19-23,271. Therefore, we think that Group A patients really showed a blunted carbon dioxide responsiveness. The hypothesis of a decreased responsiveness of respiratory centers is also supported by the observation of the significantly lower VE, VT/TI, and Edi values observed in Group A patients for a PETITE of 60 mm Hg, compared with those in both Groups B and C. All these observations support the hypothesis of a central dysregulation if we assume the absence of a peripheral neuropathy. In fact, the presence of a peripheral neuropathy involving motor fibers to respiratory muscles may cause a decrease in muscle activation that is not due to a reduction in 36
July
1993
The American
Journal
of Medicine
Volume
95
central respiratory drive. This may be the case in hypothyroid patients. In fact, inexcitability, delayed conduction time, and, in one case, fibrosis and demyelination of the phrenic nerve have been reported in hypothyroid patients [6,33]. Considering that our patients showed a low response of both diaphragm and intercostal muscles, it would be necessary to hypothesize the existence of a neuropathy involving both phrenic and intercostal nerves. However, four of six patients (Patients 1, 2, 3, and 11) with a low EMG response exhibited normal MIP values, and in light of the fact that patients with diffuse peripheral neuropathy have been reported to develop low MIP [34,35], it seems difficult to hypothesize that patients with a neuropathy involving the main respiratory muscles can develop normal MIP values. Despite few exceptions of patients with low MIP (Patients 4 and 7), we believe that peripheral neuropathy contributed to a less extent to the low EMG response we observed. Hypothyroid Myopathy Hypothyroid myopathy is the second important factor that may be responsible for the observed decreased VE response to carbon dioxide. Several reports indicate that respiratory muscle dysfunction, mostly diaphragm dysfunction, may be responsible for the impaired respiratory function in hypothyroid patients [6,7,10]. In these patients, Wilson and Bedell [lo] observed normal lung volumes but reduced maximal breathing capacity and decreased ventilatory response to hypercapnia; they suspected a major role of respiratory muscle dysfunction in these alterations. Six of our patients showed low MIP values, thus confirming the possibility that hypothyroid patients may demonstrate respiratory muscle weakness. However, the following has to be considered: (1) Four patients with low MIP had normal VE, VT, VT/T& and Edi response slopes; the only two patients with low MIP values who showed a decreased VE response slope also had a low EMG response slope; moreover, four of the six patients with a blunted VE response slope exhibited normal MIP values. (2) Unlike Edi, MIP did not significantly correlate with VE response slopes in the patients as a whole. (3) Hypothyroid patients reported to have severe diaphragm dysfunction [6,7,31] also experienced breathlessness, orthopnea, and in some cases a decrease in vital capacity and increase in PaCOz; none of these manifestations was present in any of our patients. In this context, Weiner et al [31] reported the case of a patient with severe hypothyroidism, hypercapnia, and significant respiratory muscle weakness; with replacement therapy, the patient showed rapid resolution of hypercapnia despite persistent severe respiratory muscle weakness.
CONTROL
All these findings are in accordance with a previous study of ours [36] in which patients with shortterm primary hypothyroidism exhibited low MIP but normal Edi and VE response slopes. Therefore, many lines of evidence seem to indicate that moderate respiratory muscle involvement may not be sufficient, per se, to determine a low hypercapnic ventilatory response, and that more severe respiratory muscle impairment would probably be necessary. The effects of replacement therapy confirm that respiratory alterations depend upon the hormonal deficiency in these patients. In particular, we observed that 6- to g-month replacement therapy was sufficient to restore hypercapnic response (Figures 2 and 3) but not respiratory muscle strength. This is in accordance with the observation that in hypothyroid patients, despite being euthyroid for a mean period of 1 year, treatment could induce only modest increases of quadriceps muscle strength [37]. Thus, in agreement with previous reports [6,31], it appears that hypothyroid patients need a longer time to recover from respiratory muscle weakness. Moreover, the normalization of the hypercapnic respiratory response after replacement therapy further supports the hypothesis that the reduced activation of respiratory muscles (EMG) plays a more important role in determining the reduced VE response than does respiratory muscle weakness. In conclusion, our study confirms previous observations [3] that only some hypothyroid patients may demonstrate an impaired respiratory center responsiveness to hypercapnia, which was not predicted by clinical, biologic, or functional characteristics. Decreased activation of the inspiratory muscles seems to be the major factor in determining a blunted hypercapnic ventilatory response’ in patients with severe hypothyroidism.
ACKNOWLEDGMENT We wish to thank the staff of the Nuclear Medicine Unit of the University Florence, Florence, Italy, for performing all the hormonal determinations.
of
REFERENCES 1. lngbar DH. The respiratory system. In: lngbar SH, Braverman LE, editors. The thyroid. Philadelphia: JB Lippincott, 1986: 1130-g. 2. Millman R, Bevilacqua J, Peterson DD, Pack Al. Central sleep apnea in hypothyroidism. Am Rev Respir Dis 1983; 127: 504-7. 3. Ladenson PW, Goldenheim PD. Ridgway EC. Prediction and reversal of blunted ventilatory responsiveness in patients with hypothyroidism. Am J Med 1988; 84: 877-83. 4. Zwillich CW, Pierson DJ, Hofeldt FD, Lufkin EG, Weil JV. Ventilatory control in myxedema and hypothyroidism. N Engl J Med 1975; 292: 662-5. 5. Freedman S. Lung volumes and distensibility, and maximum respiratory pressures in thyroid disease before and after treatment. Thorax 1978; 33: 785-90. 6. Laroche CL, Cairns T, Moxham J. Green M. Hypothyroidism presenting with respiratory muscle weakness. Am Rev Respir Dis 1988; 138: 472-4. 7. Martinez FJ, Bermudez-Gomez M, Celli BR. Hypothyroidism. A reversible cause of diaphragmatic dysfunction. Chest 1989; 96: 1059-63. 8. Massumi RA, Winnacker JL. Severe depression of the respiratory center in myxedema. Am J Med 1964; 36: 876-82. 9. Nordqvist P, Dhuner KG, Stenberg K, Orndahl G. Myxoedema coma and CO2
retention.
OF BREATHING
Acta Med Stand
IN HYPOTHYROIDISM
/ DURANTI
ET AL
1960; 166: 189-94. abnormalities
10. Wilson WR, Bedell GN. The pulmonary
in myxedema. J Clin Invest 1960; 39: 42-55. 11. Norris FH Jr, Panner BJ. Hypothyroid myopathy. Arch Neurol 1966; 14: 574-88. 12. Adams RD, DeLong GR. The neuromuscular system and brain. In: lngbar SH, Braverman LE, editors. The thyroid. Philadelphia: JB Lippincott, 1986: 1168-80. 13. Khaleeli AA, Griffith DG, Edwards RH. The clinical presentation of hypothyroid myopathy and its relationship to abnormalities in structure and function of skeletal muscle. Clin Endocrinol (Oxf) 1983; 19: 365-76. 14. lannuuo CD, Chen V, O’Brien P, Keens TG. Effect of experimental dysthyroidism on the enzymatic character of the diaphragm. J Appl Physiol 1984; 56: 117-21. 15. Johnson MA, Olmo JL, Mastaglia FL. Changes in histochemical profile of rat respiratory muscles in hypo- and hyperthyroidism. Q J Exp Physiol 1983; 68: 1-13. 16. Grassino AE, Goldman MD, Mead J, Sears TA. Mechanics of the human diaphragm during voluntary contraction: statics. J Appl Physiol 1978; 44: 829-39. 17. Begle RL, Skatrud JB, Dempsey JA. Ventilatory compensation for changes in functional residual capacity during sleep. J Appl Physiol 1987: 62: 1299-306. 16. Reid MB, Banzett RB, Feldman A, Mead J. Reflexcompensation of spontaneous breathing when immersion changes diaphragm length. J Appl Physiol 1985; 58: 1136-42. 19. Lopata M, Evanich MJ. Lourenco RV. Quantification of diaphragmatic EMG response to CO2 rebreathing in humans. J Appl Physiol 1977; 43: 262-70. 20. Lopata M, Zubillaga G. Evanich MJ, Lourenco RV. Diaphragmatic EMG response to isocapnic hypoxia and hyperoxic hypercapnia in humans. J Lab Clin Med 1978; 91: 698-709. 21. Scano G, Duranti R, Spinelli A, Gorini M, Lo Conte C, Gigliotti F. Control of breathing in normal subjects and in patients with chronic airflow obstruction. Bull Eur Physiopathol Respir 1987; 23: 209-16. 22. Gorini M, Ginanni R, Spinelli A, Duranti R, Andreotti L, Scano G. Inspiratory muscle strength and respiratory drive in patients with rheumatoid arthritis. Am Rev Respir Dis 1990; 142: 289-94. 23. Gorini M, Spinelli A, Ginanni R, Duranti R, Gigliotti F, Scano G. Neural respiratory drive and neuromuscular coupling in patients with chronic obstructive pulmonary disease (COPD). Chest 1990; 98: 1179-86. 24. Evered D, Hall R. Hypothyroidism. BMJ 1972; 1: 290-3. 25. American Thoracic Society. Chronic bronchitis, asthma and pulmonary emphysema. Astatement by the committee on diagnostic standards for nontuberculous respiratory disease. Am Rev Respir Dis 1962; 85: 762-8. 26. European Community for Coal and Steel. Standardization of lungfunction tests. Bull Eur Physiopathol Respir 1983; 19 Suppl 5: 1-95. 27. Spinelli A, Marconi G, Gorini M. Pizzi A, Scano G. Control of breathing in patients with myasthenia gravis. Am Rev Respir Dis 1992; 145: 135966. 28. Duranti R, Pantaleo T, Bellini F, Bongianni F, Scano G. Respiratory responses induced by the activation of somatic nociceptive afferents in humans J Appl Physiol 1991; 71: 2440-8. 29. Read DJC. A clinical method for assessing the ventilatory response to carbon dioxide. Aust Ann Med 1967; 16: 20-32. 30. Halpern SL, Glenn MB, Goodhart R. New height-weight tables: importance of new criteria. JAMA 1960; 173: 1576. 31. Weiner M, Chausow A, Szidon P. Reversible respiratory muscle weakness in hypothyroidism. Br J Dis Chest 1986; 80: 391-5. 32. Lourenco RV, Cherniack NS, Malm JR, Fishman AP. Nervous output from the respiratory center during obstructed breathing. J Appl Physiol 1966; 21: 527-33. 33. Hamly FH. Timms RM, Mihn VD, Moser M. Bilateral phrenic paralysis in myxedema. Am Rev Respir Dis 1975; 111: 911-2. 34. Rochester DF, Findley LJ. The lungs and neuromuscular and chest wall diseases. In: Murray JF, Nadel JA, editors. Textbook of respiratory medicine. Philadelphia: WB Saunders, 1988: 1942-71. 35. De Troyer A, Pride NB. The respiratory system in neuromuscular disorders. In: Roussos C. Macklem PT, editors. The thorax (Lung biology in health and disease, vol 29, part B). New York: Marcel Dekker, 1985: 1089-121. 36. Gorini M, Spinelli A, Cangioli C, eta/. Control of breathing in patients with short-term primary hypothyroidism. Lung 1989; 167: 43-52. 37. Khaleeli AA, Edwards RHT. Effect of treatment on skeletal muscle dysfunction in hypothyroidism. Clin Sci 1984; 66: 63-8.
July 1993
The American
Journal
of Wledicine
Volume
95
37
PURPOSE: Hypothyroid
patients have been reported to have a blunted ventilatory response to carbon dioxide stimulation. However, previous data did not clarify the localization of abnormalities responsible for that disorder. The present investigation was aimed at evaluating to what extent central (neural) and/or peripheral (muscular) factors are involved in the abnormalities of the ventilatory control system in hypothyroid patients. PATIENTSANDMETHOD~: Westudied13patients with severe hypothyroidism before and after 6 to 9 months of replacement therapy; 7 age- and sex-matched normal subjects were also studied as a control. In each subject, we assessed (1) inspiratory muscle strength by measuring maximal inspiratory pressure (MIP), and (2) respiratory control system during a carbon dioxide rebreathing test by measuring minute ventilation (VE), tidal volume (VT), mean inspiratory flow (VT/TI), and electromyographic (EMG) activity of the diaphragm (Ea) and intercostal (E& muscles. RESULTS: Compared with the normal control group (Group C), patients exhibited similar MIP, and similar VE and EMG response slopes to carbon dioxide. However, evaluating individual VE response slopes, we were able to identify two subsets of patients: Group A (six patients) with low VE response (less than mean - SD 1.65 of Group C) and Group B (seven patients) with normal VE response. Compared with both Groups B and C, Group A exhibited significantly lower VT/TI, E& and Eht response slopes; the difference between Groups B and C was not significant. Six patients (two from Group A and four from Group B) exhibited low MIP values compared with that in Group C. After replacel
From the lstituto di Clinica Medica III (RD, AS, AF, GS). Unita di Endocrinologia-Dipartimento di Fisiopatologia Clinica (RGG). Universita degli Studi di Firenze, and Fondazione Pro-Juventute Don C. Gnocchi (MG, FG). Florence, Italy. This study was supported by grants from the Minister0 dell’Universita e della Ricerca Scientifica e Tecnologica of Italy, Rome, Italy. Requests for reprints should be addressed to Roberto Duranti, M.D., lstituto di Clinica Medica Ill, Universita degli Studi di Firenze, Viale G.B. Morgagni 85. 50134 Florence, Italy. Manuscript submitted May 4, 1992, and accepted in revised form November 19, 1992.
ment therapy, (1) VE, VT/TI, and Ea response slopes increased significantly in Group & and (2) MIP increased, but not significantly in patients with low MIP. CONCLUSIONS: We conclude that: (1) In patients with severe hypothyroidism, the ventilatory control system may be altered at the neural level, as indicated by a blunted chemosensitivity; (2) Impaired respiratory muscle function does not seem to play a major role in the decreased ventilatory response to carbon dioxide stimulation; (3) Replacement therapy appears to normalize the response to hypercapnic stimulation, but not respiratory muscle strength.
R
espiratory manifestations are rarely a major problem in hypothyroid patients, even if several abnormalities in respiratory function may be found in these patients. Fatigue and dyspnea on exertion [l], disordered breathing during sleep [2], impaired ventilatory response to hypoxia and hypercapnia [3,4], decreased inspiratory muscle force [5], and severe diaphragmatic dysfunction [6,7] have been reported. Sometimes, ventilatory involvement is life-threatening, since carbon dioxide retention, respiratory failure, and coma may ensue [l&9]. The reasons why hypothyroid patients can develop alveolar hypoventilation have not been clarified so far. It has been suggested that depression of respiratory centers causing a decreased inspiratory neural drive may be involved [l-3]. On the other hand, decrease in inspiratory muscle force, i.e., muscle weakness, has also been thought to play a role [6,7,10]. In fact, the severe myopathy, described in hypothyroid adults 111-131, may also affect respiratory muscles [14,15]. Recently, Ladenson et a2 [3] have investigated the prevalence of impaired hypoxic and hypercapnic ventilatory responses and the clinical and chemical parameters that might predict the presence of abnormal ventilatory control in hypothyroid patients. These authors reported that only a subset of patients may have a blunted respiratory response to either hypercapnic (34%) or hypoxic (27%) stimulation. Hypothyroid women and patients with higher values of serum thyroid-stimulating hormone (TSH) were significantly more likely to have impaired ventilaJuly 1993
The American
Journal
of Medicine
Volume
95
29
CONTROL
OF BREATHING
IN HYPOTHYROIDISM
/ DURANTI
ET AL
TABLE I Anthropometric, Biologic, and Clinical Data of the 13 Hypothyroid Patients Before and After ReplacementTherapy*
Subject
Age (Y)
p#;0 Before
After
TSH (mu/L) Before After
&i-h, Before After
Tab
Mab
Diagnosis
1
55
83.2
84.0
264
0.70
4.20
10.20
1:1,280
1:80
Chronic autoimmune thyroiditis
2
58
98.6
95.2
>64
1.30
2.10
10.80
1:2,500
1:80
Chronic autoimmune thyroiditis
3
57
82.1
89.4
>64
1.00
0.86
10.50
1:1,280
1:4Q
Chronic autoimmune thyroiditis
4
51
93.4
85.3
>64
1.50
3.14
8.20
Negative
1:80
Chronic autoimmune thyroiditis
5
62
96.2
89.5
>64
0.04
1.77
10.90
Negative
1:80
Chronic autoimmune thyroiditis
6
66
109.0
96.9
>64
2.20
1.10
10.80
1:160
1:80
Chronic autoimmune thyroiditis
7
40
105.3
102.0
>64
0.30
0.75
9.10
Negative
1:40
Chronic autoimmune thyroiditis
8
69
124.3
124.3
>64
9.30
1.00
8.90
1:1,280
1:40
Chronic autoimmune thyroiditis
9
49
111.2
104.9
>64
0.60
2.50
9.20
Negative
1:40
Postablative (radioiodine)
10
44
102.6
100.3
>64
1.10
0.58
6.60
Negative
Negative
Primary idiopathic
11
30
98.8
86.6
>64
1.00
0.80
9.70
1:1,280
Negative
Primary idiopathic
12
57
102.0
96.8
>64
3.10
1.10
8.70
Negative
1:lO
Primary idiopathic
13
64
92.3
96.0
>64
8.30
3.70
9.10
1:80
1:160
Chronic autoimmune thyroiditis
Mean (patients) SEM
54.0 3.10
99.9 3.20
96.3 2.90
>64
2.4 3.0
1.84 1.07
9.37 1.27
Mean (normals) SEM
“E
102.0 3.3
iormal rangesare: serum TSH, 0.6 io 4.6 mu/L; T4,4.5 to 12 kgldt; Tab and Mab, not present. Body weight is expressed as percentageof ideal weight (IW).
tory responses. However, in that study [3] as in a previous one [4], it was not possible to ascertain the localization of the lesion(s) responsible for disordered ventilatory responsiveness; neural respiratory drive was assessed in terms of minute ventilation (VE) and mean inspiratory flow (VT/TI). In fact, in patients with ventilatory disorders, mechanical limitation of the lung and/or chest wall prevents VE and VT/TI from accurately reflecting respiratory center output. Therefore, neither VE nor VT/TI allows ascertainment of whether central or peripheral abnormalities are involved in the decreased ventilatory response observed in patients with hypothyroidism. The relative role of central (neural) and peripheral (muscular) abnormalities might be assessed by recording the electromyographic (EMG) activity of the diaphragm (Edi), an index of the neural drive to the respiratory muscles [16-231, and maximal inspiratory pressure (MIP), an index of inspiratory muscle force. The current investigation was undertaken in patients with severe hypothyroidism, before and after replacement therapy, to evaluate the existence of abnormalities in the ventilatory control system and, if any, to what extent central and/or peripheral factors are involved. 30
July
1993
The American
Journal
of Medicine
Volume
95
PATIENTS AND METHODS Patients Thirteen female patients with clinical and laboratory signs of severe primary hypothyroidism (overt hypothyroidism by the classification of Evered and Hall [24]) were studied before and after thyroid hormone replacement therapy. A group of seven age- and sex-matched healthy normal subjects was also studied as a control. Informed consent was obtained from patients and normal subjects. The anthropometric, biologic, and clinical data of patients are provided in Table I. All were outpatients and were examined at the Endocrinological Unit of the University of Florence. None of them had evidence of pulmonary, cardiac, or neurologic diseases on the basis of clinical evaluation, routine laboratory testing, chest radiography, and electrocardiography. None of the patients fulfilled the diagnostic criteria of asthma, chronic bronchitis, or emphysema according to the American Thoracic Society [25], and none of them exhibited dyspnea at rest or during exercise. Smokers were excluded from the study. Thyroid Function Study Thyroid function was evaluated on the day of the
CONTROL
study by measuring serum TSH concentration and total serum thyroxine (Td). The hormonal and immune determinations were performed with “routine” methods. The measurement of TSH was performed by the Allegro TM HS-TSH Immunoassay System kit (Nichols Institute, San Juan Capistrano, CA [normal values: 0.6 to 4.6 mu/L]). Tq was determined by radioimmunoassay methods (ARIA HT Tq system [normal values: 4.5 to 12 pg/dL], Becton Dickinson Immunodiagnostic, Salt Lake City, UT). Thyroglobulin antibodies (Tab) were determined by a passive hemoagglutination test (Wellcome Diagnostics, Dartford, England); microsomal thyroid antibodies (Mab) were detected by an immunofluorescence method using cryostatic sections of human thyroid. Pulmonary Function Study Routine spirometry was performed as previously described [21]. The normal values for lung volumes are those proposed by the European Community for Coal and Steel [26]. MIP was measured using a differential pressure transducer (Statham SC 1001, Gould Inc., Oxnard, CA) as already reported [27]. After baseline routine testing, ventilatory pattern and respiratory drive were evaluated during a carbon dioxide rebreathing test. Each subject, placed in a comfortable supine position, breathed through a mouthpiece and a Fleisch (no. 3) pneumotachograph (Beckman, Geneva, Switzerland) attached to a one-way valve that separated the inspiratory from the expiratory line. Flow signal was integrated to yield tidal volume (VT). Inspiratory time (TI) and respiratory cycle duration (TT) were measured from the volume tracing. VT/TI and VE (VE = VT a 6O/TT) were also calculated. Expired carbon dioxide was continuously measured at the mouth by an infrared carbon dioxide meter. The EMG activity of the respiratory muscles was recorded from the second parasternal intercostal (E& and diaphragm (Edi) muscles via large surface electrodes; the technique employed to record EMG activity has previously been described in detail [21,27,28]. Briefly, the differentially amplified and filtered muscle action potentials were full wave rectified and “integrated” over time (time constant 100 milliseconds) in order to provide a measure of changes in average electrical activity as a function of time, referred to as “moving time average” [19,20]. Inspiratory activity was quantified both as peak of activity (directly measured in arbitrary units) and as rate of increase in activity (obtained by dividing the peak activity by the inspiratory time). Because of the variability of the impedance between diaphragm and electrodes, absolute values (mV) are not comparable in different subjects.
OF BREATHING
IN HYPOTHYROIDISM
/ DURANTI
ET AL
Therefore, in order to obtain a reference value, EMG activity was measured while the subject was connected to the pneumotachograph and performed an inspiratory maneuver breathing in up to the total lung capacity (TLC) [16,27]. This maneuver was repeated at least three times, and in each subject, both inspiratory volume and peak of EMG activity at TLC were closely reproducible (less than 5% variability). The mean level of this EMG activity was taken as the reference, and all successive measurements have been expressed as a percentage of the reference value obtained at TLC. The output of the carbon dioxide analyzer, flow signal, integrated flow signal, and the moving time average (Edi and E& were continuously recorded on a multichannel chart recorder. After a period of 5 to 10 minutes, which allowed the subjects to adapt to the circuit, as shown by the stability of carbon dioxide recording, each subject underwent a carbon dioxide rebreathing test following the procedure recommended by Read [29]. A gas mixture (7% carbon dioxide, 93% oxygen) was inhaled over 3 to 5 minutes from an 8-L bag. The response to carbon dioxide stimulation was evaluated in terms of slope of the regression line of studied variables versus end-tidal partial pressure of carbon dioxide (PETIT,). In each normal subject, the rebreathing test was repeated on two to three different days, whereas in patients, it was duplicated on the same day with an interval of 60 minutes between each test. Study Protocol Respiratory and thyroid function tests were performed prior to the beginning of therapy. All the patients underwent thyroid replacement therapy with L-thyroxine (L-Td); the patients initially received 0.025 mg daily. The dose of L-T4 was increased by 0.025-mg increments up to a maximal dose of 0.150 mg. After 6 to 9 months, when euthyroid as judged on the basis of both clinical and thyroid function tests, patients were reexamined as before therapy. Data Analysis For each rebreathing run, changes in VE, VT, VT/TI, Edi, and Eint were plotted against corresponding values of PET~~, and subjected to leastsquare linear regression analysis. To obtain a normal reference range, an average slope value of the runs carried out by each normal subject was considered. In no case was the response exhibited on 1 day twice greater or less than half of the response obtained on the other days. The mean slope for the two runs was calculated for each patient. Data were averaged for patients and normal subjects and are July
1993
The American
Journal
of Medicine
Volume
95
31
CONTROL OF BREATHING IN HYPOTHYROIDISM / DURANTI ET AL
TABLE II Pulmonary Function Data and Maximal lnspiratory Pressuresof the 13 Hypothyroid Patients Before and After Replacement Therapy*
vc+
Before ’ After
MIP km Hz01 Before After
Before
After
1
104.8
102.3
102.7
103.5
75.0
85.0
91.9
95.5
101.9
102.6
80.2
85.5
120
118
2
114.0
130.7
87.7
85.3
88.1
81.9
98.6
107.5
114.9
125.5
81.7
78.2
80
122
3
112.3
113.4
111.1
117.5
92.2
87.7
112.6
111.2
115.9
120.3
84.6
86.9
106
90
4
95.1
102.6
110.2
108.9
88.6
92.6
102.3
96.5
103.4
94.6
75.0
76.1
54
82
5
119.9
125.2
93.4
92.9
92.7
92.1
103.7
102.5
125.3
121.7
83.9
77.8
108
108
6
97.6
103.4
124.6
91.7
120.8
97.9
98.4
95.9
88.3
97.8
72.0
75.0
52
62
7
99.3
98.5
96.3
96.0
91.1
92.1
96.2
96.2
88.3
93.4
75.2
80.0
48
64
8
97.7
85.7
120.5
146.1
127.6
127.1
109.6
116.5
91.1
84.8
73.5
77.6
70
68
9
115.3
115.1
92.5
92.0
73.6
84.5
104.9
103.8
112.5
110.1
81.4
80.2
94
90
10
96.5
99.0
101.6
117.7
104.8
107.1
98.0
110.1
119.8
117.1
86.5
82.8
52
45
11
104.3
103.1
96.0
122.6
90.3
111.1
101.0
106.2
109.9
110.9
90.0
92.1
88
82
12
97.9
104.6
126.7
112.1
113.7
107.3
96.9
114.1
98.1
115.0
73.3
75.6
50
88
13
95.0
94.7
113.0
112.7
88.8
104.8
95.0
94.6
85.8
94.8
72.5
80.0
52
60
105.8
105.3
95.9
97.8
103.9 2.1
104.2 3.6
106.8 3.6
79.3 1.7
80.6 1.4
74.1 6.8
83.0 6.4
103.8
Mean (normals) SEM
116.2
2.4 2.8
106.0
3.4
3.5
3.4
102.4
3.5
4.5
3.6
98.5
3.8
TLC+ Before After
F%vc
Subject
Mean (patients) SEM
RV+ Before After
FE&+ Before After
FRC+ Before After
100.6
1.6 103.7
2.2
99.7
2.7
81.2
1.9
83.0 5.8
: = vital capacity; FRC = functional residual capacity; RV = residual volume; TLC = total lung capacity, FEVl = forced expiratory volume in 1 second; MIP = maximal inspiratory pressure. *individual data and mean values r SEM are reported. tpercent of predicted value.
presented as the mean f SEM. Comparisons between patients and normal control subjects were performed by Mann-Whitney U test for unpaired samples; significance of variations induced by replacement therapy was evaluated by Wilcoxon rank tests for paired samples. VE response slope and MIP values more than SD * 1.65 below the mean value for a normal control group were considered to be significantly reduced. Comparisons among the three groups (Groups A and B patients and normal control group) were made by one-way analysis of variance (ANOVA); Newman-Keuls tests were used to assess the significance of differences between pairs of groups where ANOVA showed statistically significant differences. Differences were considered significant when p was less than 0.05.
RESULTS Anthropometric, biologic, and clinical characteristics of the patients before and after replacement therapy are summarized in Table I. A hypothyroid condition was ascertained on the basis of both clinical and biologic data. All patients presented with high values of TSH and low values of TJ; significant levels of serum Tab and Mab were observed in 7 and 32
July 1993 The American Journal of Medicine
Volume 95
11 patients, respectively (Table I). After replacement therapy, 11 patients exhibited normal values of TSH and Tq, and were judged to be euthyroid (Table I). Two patients (Patients 8 and 13), who showed slightly elevated TSH values and normal T.J values, were considered to be euthyroid (preclinical hypothyroidism by the classification of Evered and Hall [24]). Study Before Treatment The average body weight of patients, expressed as percentage of the ideal weight [30], was 99.9% f 3.2%. Patients invariably exhibited normal spirometric values (Table II). No patient appeared to retain carbon dioxide; PETIT, was 40.6 f 0.77. As a mean, no significant difference was observed in terms of MIP (Table II) between patients and the normal control group (Group C); however, in 6 patients (Patients 4,6,7,10,12, and 13, see Table II), MIP values were less than mean - SD * 1.65 of the values calculated for Group C. During carbon dioxide rebreathing, as a whole, hypothyroid patients did not exhibit any significant difference in terms of VE, VT, VT/TI, Edi, and Eint response slopes to increasing carbon dioxide
CONTROL
1.6
OF BREATHING
IN HYPOTHYROIDISM
0.06
/ DURANTI
ET AL
0.08
22 q
G q
\ 1 1.2 73 0.6
Gm 0.06 : 1 0.04
=:
;8
If 3 0.4 5
g3 0.02 Q
, B 0.045 i. 7=q 0.03
”0”
3r” 0.015 \ 5
0
0
a
Figure 1. Slopes of the relationships of VE, VT, VT/Tl, and Edi and Ei”t against corresponding PETco, values, during carbon dioxide rebreathing, in the 13 hypothyroid patients, before and after therapy, and in the normal control group. Values are mean i SEM.
Normal subjects
TABLE III Slopes of the Relationshipsof VE,VT,VT/TI, Edi,and EintVersus Corresponding PETCO,Values During Rebreathing Test in Groups A and B Patients Before ReplacementTherapy, and in Normal Control Group* AVE/APETCO
AVT/APETCO~
AWTITI)/APETCIJ~
No.
([L/mini/mm i&I
(L/mm Hg)
UL/sl/mm Hg)
Group A patients
6
Group B patients
7
Normal subjects
7
0.52 (0.03) 1.24 (0.17) 1.22 (0.11)
0.036 (0.008) 0.065 (0.011) 0.066 (0.007)
0.018 (0.003) 0.041 (0.006) 0.043 (0.006)
Subjects
One-way ANOVA F P
9.42 <0.002
3.04 <0.07
AEdi/APETCO ([%TLC/sl/mmkg) 0.482 (0.135) 1.57 (0.36) 1.46 (0.24) 4.50 <0.027
AEdAPETCo
([%TLC/sl/mmkg) 0.36 (0.09) 0.99 (0.25) 1.55
(0.39) 4.04 (0.037
Newman-Keuls Group A versus normals Group B versus normals Group A versus Group B
I
VE = minute ventilation; VT = tidal volume; WTI = mean inspiratory flow; PETCO,= end-tidal partial pressure of carbon dioxide; Ed, = EMG activity of the diaphragm; E,,t = EMG activity of the parasternal intercostal muscles. Ediand Elntwere quantified as the rate of Increase of EMGactivity obtainedby dividing the peak of inspiratory activity by inspiratory time (TI). *Valuesare mean i- SEM.
compared with the normal control group (Figure 1); nevertheless, as indicated in Figure 1, there was a tendency for these variables to be lower in patients than in normal subjects. Assessing the individual VE responses to carbon dioxide, we were able to distinguish two subsets of patients: one (Group A, six patients: Patients 1 through 4,7, and 11) with a blunted VE response slope (less than mean - SD . 1.65 of the values calculated for the normal control group) and the other (Group B, seven patients: Pa-
tients 5,6,8 through 10,12, and 13) with a normal VE response slope. VE, VT/TI, Edi, and Eint response slopes to increasing PET~O, did significantly differ (one-way ANOVA) among Groups A, B, and C (Table III). Intergroup comparisons (Newman-Keuls tests) showed that VE, VT/TI, and Edi response slopes were significantly lower in Group A than in Groups B and C, whereas Eint was significantly lower in Group A than in Group C. In Group A, the VT response slope was lower compared with that in July
1993
The American
Journal
of Medicine
Volume
95
33
CONTROL
OF BREATHING
IN HYPOTHYROIDISM
/ DURANTI
ET AL
TABLEIV VE,VT,WTt,
Edi, and Eint for a PETCO, of 60 mm Hg During Carbon Dioxide Rebreathing in Normal Control Group*
I
No.
Group A patients
6
Group B patients
7
(L/min)
13.0 (1.24) 17.10
1.02 (0.10) 1.19
(0.20) 1.26 (0.13)
(0.91) 21.08 (2.29)
7
Normal subjects One-way ANOVA F P
Therapy,
0.43 (0.04) 0.75 (0.02) 0.72 (0.09)
5.74 (2.31) 13.95 (1.68) 12.48 (2.85)
10.2 (2.81) 12.47 (4.45) 15.98 (5.70) 0.39 CO.68
to,011
0.57 to.57
6.35 < 0.009
5.10 to.018
<0.05
NS
-co.05
to.05
5.89
Newman-Keuls Group A versus normals Group B versus normals Group A versus Group B
Before Replacement
and
VT/TI L/s)
VE
Subjects
in Groups A and B Patients
NS N”s”
<0.05
N”s”
NS
<0.05
K NS
*Values are mean f SEM. Same abbreviations as in Table III.
Figure 2. Individual slopes of the relationships of VE, VT, and VT/T1 against corresponding PETco, values, during carbon dioxide rebreathing, in the six hypothyroid patients with low ventilatory response to carbon dioxide stimulation (Group A), before (b) and after (a) replacement therapy. Mean values f SEM for the six patients, before (0) and after (A) replacement therapy, and for the normal control group (m) are also shown.
b
a
b
both Groups B and C, but it did not reach the level of statistical significance. No significant difference in any of the studied variables was observed between Groups B and C. At a PETIT, of 60 mm Hg, VE, VT/TI, and Edi significantly differed among the three groups (ANOVA, Table IV). Newman-Keuls 34
July
1993 The American
Journal
of Medicine
Volume
95
a
Figure 3. Individual slopes of the relationships of Edi and Ei,t against corresponding PETco, values, during carbon dioxide rebreathing, in the six hypothyroid patients with low ventilatory response to carbon dioxide stimulation (Group A), before (b) and after (a) replacement therapy. Mean values f SEM for the six patients, before (0) and after (A) replacement therapy, and for the normal control group (m) are also shown.
tests showed that VE was lower in Group A compared with Group C, and VT/TI and Edi were significantly lower in Group A compared with both Groups B and C. No difference was observed in any of the above variables between Groups B and C. Anthropometric, clinical, and biologic character-
CONTROL
OF BREATHING
IN HYPOTHYROIDISM
/ DURANTI
ET AL
TABLE V Slopesof the Relationships of VE, VT, VT/TI, Edi, and EintVersusCorresponding PETCO,Values During Rebreathing Testin Groups A and B Patients After ReplacementTherapy, and in Normal Control Group* AVE/APETCQ Subjects
No.
KL/minl/mm Hg)
Group A patients
6
Group B patients
7
Normal subjects
7
1.40 (0.21) 1.33 (0.22) 1.22 (0.11)
One-way ANOVA F P
0.22 to.80
AVT/APETCO,
A(VT/TI)IAPETCO,
&/APETCo2
AEdAPETCO
(L/mm Hg)
([L/sl/mm Hg)
(I%TLC/sl/mmHg)
U%TLC/sl/mmkg)
0.067 CO.0161 0.066 (0.007)
0.047 (0.007) 0.047 (0.007) 0.043 (0.006)
1.197 (0.230) 1.90 (0.63) 1.46 (0.24)
0.908 (0.118) 1.55 (0.39)
0.04 to.96
0.23 <0.80
0.69 co.52
1.24 to.31
*Values are mean f SEM. Same abbreviations as in Table III,
istics, lung volumes, and MIP did not significantly differ between Groups A and B. Only two Group A patients (Patients 4 and 7) showed low MIP values. Study After Treatment Patients were reexamined after 6 to 9 months of replacement therapy; as a whole, no significant changes in ideal weight, lung volumes, MIP (Table II), and PETITE response slopes (Figure 1) were found. In the six patients with low MIP values, MIP increased without, however, reaching the level of statistical significance (from 51.33 f 0.84 to 66.83 f 6.41). In Group A, compared with the pretreatment values, replacement therapy resulted in a significant increase in VE (p <0.03), VT/T1 (p <0.03), and Edi (p <0.03) response slopes; VT and Eint response slopes also rose, but the increase was not significant. Mean and individual response slopes in Group A, before and after replacement therapy, are depicted in Figures 2 and 3. When euthyroid, Group A did not differ compared with both Groups B and C (one-way ANOVA, Table V). In Group B, replacement therapy did not result in any significant ventilatory and EMG changes (Table V). At a PET~~, of 60 mm Hg, VE (p <0.03), VT/T1 (p <0.03), and Edi (p <0.03) significantly increased in Group A; the values no longer differed from those observed in both Groups B and C (ANOVA, Table VI). In the patients as a whole, the relationship of the VE response slope with the Edi response slope, obtained before and after treatment, was found to be significant (r = 0.62, p <0.025, n = 26). In contrast, the VE response slope did not relate to the MIP (r = 0.1, NS, n = 26).
COMMENTS Our data in female hypothyroid subjects are consistent with previous studies showing that patients with hypothyroidism may or may not demonstrate a low ventilatory response to increasing carbon dioxide [3,4]. In our study, 46% of the total population
r-~
TABLE VI VE, VT, W/T!, Edi, and Eintfor a PETCOof 60 mm Hg During Carbon Dioxide Rebreathing in GroupsA and fPatients After Replacement Therapy, and in Normal Control Group* VE
VT/TI
(L/s)
Subjects No. (L/min)
Edi
(%TLC/s) C%?;,s)
I Group A patients GroupB patients Normal subjects One-way ANOVA F P
6 7 7
20.88 (3.91) 22.26 (2.20) 21.08 (2.29)
1.15 (0.14) 1.27 (0.10) 1.26 (0.13)
0.82 (0.12) 0.82 (0.07) 0.72 (0.09)
14.4 (3.20) 16.08 (3.30) 12.48 (2.85)
12.41 (5.61) 13.01 (2.32) 15.98 (5.70)
0.07 to.93
0.20 co.82
0.36 to.70
0.35 to.71
0.16 to.85
I
*Valuesare mean f SEM. Same abbreviations as in Table Ill.
exhibited a low ventilatory response, a percentage very similar to that reported in the study of Ladenson et al [3], where only women were observed to exhibit a decreased ventilatory response to hypercapnic stimulation. The most important finding in our study is that the subset of hypothyroid patients with a low VE response slope to carbon dioxide stimulation also exhibited significantly lower VT/TI, Edi, and Eint response slopes. Two main factors are likely to play a role in the decreased hypercapnic VE response in hypothyroid patients: (1) impairment in neural function either at central (decreased neural respiratory drive) or peripheral (motor neuropathy) levels [1,3]; (2) respiratory muscle weakness, i.e., the failure to generate force, due to hypothyroid myopathy [6,7,10]. Impairment in Neural Function It has been previously supposed that neural factors play an important role in determining respiratory abnormalities in hypothyroid patients [1,3]. In this connection, moderate to marked reduction in the EMG response slope to carbon dioxide observed in patients with a low VE response slope (Figure 3) July
1993
The American
Journal
of Medicine
Volume
95
35
CONTROL OF BREATHING IN HYPOTHYROIDISM / DURANTI ET AL
indicates decreased neural activation of both diaphragm and intercostal muscles; possible reasons for this alteration may be either central brainstem dysregulation or peripheral neuropathy, or both. The hypothesis that central brainstem dysregulation contributes to the impairment in respiratory function in hypothyroid patients has been raised previously [2,3,31]. In this connection, Ladenson et al [3] observed that TSH concentration, but not Tq levels, predicted the occurrence of an abnormal ventilatory response, and accordingly, TSH secretion could reflect the impact of thyroid hormone deficiency on the central nervous system. We cannot contribute to this hypothesis owing to the lack of precise values of TSH when its concentration was greater than 64 U; thus, a relationship between TSH values and VE response to carbon dioxide stimulation could not be assessed. In agreement with the findings of Ladenson et al [3], we did not observe any significant relationship between hypercapnic ventilatory response and Tq values. Our data support evidence against a role of anthropometric, biologic, and clinical characteristics, given the lack of any difference in this regard between Groups A and B. In the present study, the association of a low VE response with a low EMG response suggests the possibility of a decreased responsiveness of respiratory centers to carbon dioxide stimulation. This conclusion may be correct if one considers EMG activity of the inspiratory muscles as a reliable index of neural inspiratory drive. A close correlation between changes in electrical activity of the phrenic nerve and the diaphragm has been reported in a dog, during both normal breathing and obstructed breathing [32]. Nevertheless, the use of either surface or esophageal EMG recordings to assess the neural drive to inspiratory muscles in humans has been thoroughly criticized [21-23,271. However, as we recently outlined [27], many data support the contention that the EMG can actually reflect the neural drive to the respiratory muscles, and that the slope of the “moving time average” of Edi is a reliable measure to assess neural inspiratory drive to the diaphragm in human subjects [19-23,271. Therefore, we think that Group A patients really showed a blunted carbon dioxide responsiveness. The hypothesis of a decreased responsiveness of respiratory centers is also supported by the observation of the significantly lower VE, VT/TI, and Edi values observed in Group A patients for a PETITE of 60 mm Hg, compared with those in both Groups B and C. All these observations support the hypothesis of a central dysregulation if we assume the absence of a peripheral neuropathy. In fact, the presence of a peripheral neuropathy involving motor fibers to respiratory muscles may cause a decrease in muscle activation that is not due to a reduction in 36
July
1993
The American
Journal
of Medicine
Volume
95
central respiratory drive. This may be the case in hypothyroid patients. In fact, inexcitability, delayed conduction time, and, in one case, fibrosis and demyelination of the phrenic nerve have been reported in hypothyroid patients [6,33]. Considering that our patients showed a low response of both diaphragm and intercostal muscles, it would be necessary to hypothesize the existence of a neuropathy involving both phrenic and intercostal nerves. However, four of six patients (Patients 1, 2, 3, and 11) with a low EMG response exhibited normal MIP values, and in light of the fact that patients with diffuse peripheral neuropathy have been reported to develop low MIP [34,35], it seems difficult to hypothesize that patients with a neuropathy involving the main respiratory muscles can develop normal MIP values. Despite few exceptions of patients with low MIP (Patients 4 and 7), we believe that peripheral neuropathy contributed to a less extent to the low EMG response we observed. Hypothyroid Myopathy Hypothyroid myopathy is the second important factor that may be responsible for the observed decreased VE response to carbon dioxide. Several reports indicate that respiratory muscle dysfunction, mostly diaphragm dysfunction, may be responsible for the impaired respiratory function in hypothyroid patients [6,7,10]. In these patients, Wilson and Bedell [lo] observed normal lung volumes but reduced maximal breathing capacity and decreased ventilatory response to hypercapnia; they suspected a major role of respiratory muscle dysfunction in these alterations. Six of our patients showed low MIP values, thus confirming the possibility that hypothyroid patients may demonstrate respiratory muscle weakness. However, the following has to be considered: (1) Four patients with low MIP had normal VE, VT, VT/T& and Edi response slopes; the only two patients with low MIP values who showed a decreased VE response slope also had a low EMG response slope; moreover, four of the six patients with a blunted VE response slope exhibited normal MIP values. (2) Unlike Edi, MIP did not significantly correlate with VE response slopes in the patients as a whole. (3) Hypothyroid patients reported to have severe diaphragm dysfunction [6,7,31] also experienced breathlessness, orthopnea, and in some cases a decrease in vital capacity and increase in PaCOz; none of these manifestations was present in any of our patients. In this context, Weiner et al [31] reported the case of a patient with severe hypothyroidism, hypercapnia, and significant respiratory muscle weakness; with replacement therapy, the patient showed rapid resolution of hypercapnia despite persistent severe respiratory muscle weakness.
CONTROL
All these findings are in accordance with a previous study of ours [36] in which patients with shortterm primary hypothyroidism exhibited low MIP but normal Edi and VE response slopes. Therefore, many lines of evidence seem to indicate that moderate respiratory muscle involvement may not be sufficient, per se, to determine a low hypercapnic ventilatory response, and that more severe respiratory muscle impairment would probably be necessary. The effects of replacement therapy confirm that respiratory alterations depend upon the hormonal deficiency in these patients. In particular, we observed that 6- to g-month replacement therapy was sufficient to restore hypercapnic response (Figures 2 and 3) but not respiratory muscle strength. This is in accordance with the observation that in hypothyroid patients, despite being euthyroid for a mean period of 1 year, treatment could induce only modest increases of quadriceps muscle strength [37]. Thus, in agreement with previous reports [6,31], it appears that hypothyroid patients need a longer time to recover from respiratory muscle weakness. Moreover, the normalization of the hypercapnic respiratory response after replacement therapy further supports the hypothesis that the reduced activation of respiratory muscles (EMG) plays a more important role in determining the reduced VE response than does respiratory muscle weakness. In conclusion, our study confirms previous observations [3] that only some hypothyroid patients may demonstrate an impaired respiratory center responsiveness to hypercapnia, which was not predicted by clinical, biologic, or functional characteristics. Decreased activation of the inspiratory muscles seems to be the major factor in determining a blunted hypercapnic ventilatory response’ in patients with severe hypothyroidism.
ACKNOWLEDGMENT We wish to thank the staff of the Nuclear Medicine Unit of the University Florence, Florence, Italy, for performing all the hormonal determinations.
of
REFERENCES 1. lngbar DH. The respiratory system. In: lngbar SH, Braverman LE, editors. The thyroid. Philadelphia: JB Lippincott, 1986: 1130-g. 2. Millman R, Bevilacqua J, Peterson DD, Pack Al. Central sleep apnea in hypothyroidism. Am Rev Respir Dis 1983; 127: 504-7. 3. Ladenson PW, Goldenheim PD. Ridgway EC. Prediction and reversal of blunted ventilatory responsiveness in patients with hypothyroidism. Am J Med 1988; 84: 877-83. 4. Zwillich CW, Pierson DJ, Hofeldt FD, Lufkin EG, Weil JV. Ventilatory control in myxedema and hypothyroidism. N Engl J Med 1975; 292: 662-5. 5. Freedman S. Lung volumes and distensibility, and maximum respiratory pressures in thyroid disease before and after treatment. Thorax 1978; 33: 785-90. 6. Laroche CL, Cairns T, Moxham J. Green M. Hypothyroidism presenting with respiratory muscle weakness. Am Rev Respir Dis 1988; 138: 472-4. 7. Martinez FJ, Bermudez-Gomez M, Celli BR. Hypothyroidism. A reversible cause of diaphragmatic dysfunction. Chest 1989; 96: 1059-63. 8. Massumi RA, Winnacker JL. Severe depression of the respiratory center in myxedema. Am J Med 1964; 36: 876-82. 9. Nordqvist P, Dhuner KG, Stenberg K, Orndahl G. Myxoedema coma and CO2
retention.
OF BREATHING
Acta Med Stand
IN HYPOTHYROIDISM
/ DURANTI
ET AL
1960; 166: 189-94. abnormalities
10. Wilson WR, Bedell GN. The pulmonary
in myxedema. J Clin Invest 1960; 39: 42-55. 11. Norris FH Jr, Panner BJ. Hypothyroid myopathy. Arch Neurol 1966; 14: 574-88. 12. Adams RD, DeLong GR. The neuromuscular system and brain. In: lngbar SH, Braverman LE, editors. The thyroid. Philadelphia: JB Lippincott, 1986: 1168-80. 13. Khaleeli AA, Griffith DG, Edwards RH. The clinical presentation of hypothyroid myopathy and its relationship to abnormalities in structure and function of skeletal muscle. Clin Endocrinol (Oxf) 1983; 19: 365-76. 14. lannuuo CD, Chen V, O’Brien P, Keens TG. Effect of experimental dysthyroidism on the enzymatic character of the diaphragm. J Appl Physiol 1984; 56: 117-21. 15. Johnson MA, Olmo JL, Mastaglia FL. Changes in histochemical profile of rat respiratory muscles in hypo- and hyperthyroidism. Q J Exp Physiol 1983; 68: 1-13. 16. Grassino AE, Goldman MD, Mead J, Sears TA. Mechanics of the human diaphragm during voluntary contraction: statics. J Appl Physiol 1978; 44: 829-39. 17. Begle RL, Skatrud JB, Dempsey JA. Ventilatory compensation for changes in functional residual capacity during sleep. J Appl Physiol 1987: 62: 1299-306. 16. Reid MB, Banzett RB, Feldman A, Mead J. Reflexcompensation of spontaneous breathing when immersion changes diaphragm length. J Appl Physiol 1985; 58: 1136-42. 19. Lopata M, Evanich MJ. Lourenco RV. Quantification of diaphragmatic EMG response to CO2 rebreathing in humans. J Appl Physiol 1977; 43: 262-70. 20. Lopata M, Zubillaga G. Evanich MJ, Lourenco RV. Diaphragmatic EMG response to isocapnic hypoxia and hyperoxic hypercapnia in humans. J Lab Clin Med 1978; 91: 698-709. 21. Scano G, Duranti R, Spinelli A, Gorini M, Lo Conte C, Gigliotti F. Control of breathing in normal subjects and in patients with chronic airflow obstruction. Bull Eur Physiopathol Respir 1987; 23: 209-16. 22. Gorini M, Ginanni R, Spinelli A, Duranti R, Andreotti L, Scano G. Inspiratory muscle strength and respiratory drive in patients with rheumatoid arthritis. Am Rev Respir Dis 1990; 142: 289-94. 23. Gorini M, Spinelli A, Ginanni R, Duranti R, Gigliotti F, Scano G. Neural respiratory drive and neuromuscular coupling in patients with chronic obstructive pulmonary disease (COPD). Chest 1990; 98: 1179-86. 24. Evered D, Hall R. Hypothyroidism. BMJ 1972; 1: 290-3. 25. American Thoracic Society. Chronic bronchitis, asthma and pulmonary emphysema. Astatement by the committee on diagnostic standards for nontuberculous respiratory disease. Am Rev Respir Dis 1962; 85: 762-8. 26. European Community for Coal and Steel. Standardization of lungfunction tests. Bull Eur Physiopathol Respir 1983; 19 Suppl 5: 1-95. 27. Spinelli A, Marconi G, Gorini M. Pizzi A, Scano G. Control of breathing in patients with myasthenia gravis. Am Rev Respir Dis 1992; 145: 135966. 28. Duranti R, Pantaleo T, Bellini F, Bongianni F, Scano G. Respiratory responses induced by the activation of somatic nociceptive afferents in humans J Appl Physiol 1991; 71: 2440-8. 29. Read DJC. A clinical method for assessing the ventilatory response to carbon dioxide. Aust Ann Med 1967; 16: 20-32. 30. Halpern SL, Glenn MB, Goodhart R. New height-weight tables: importance of new criteria. JAMA 1960; 173: 1576. 31. Weiner M, Chausow A, Szidon P. Reversible respiratory muscle weakness in hypothyroidism. Br J Dis Chest 1986; 80: 391-5. 32. Lourenco RV, Cherniack NS, Malm JR, Fishman AP. Nervous output from the respiratory center during obstructed breathing. J Appl Physiol 1966; 21: 527-33. 33. Hamly FH. Timms RM, Mihn VD, Moser M. Bilateral phrenic paralysis in myxedema. Am Rev Respir Dis 1975; 111: 911-2. 34. Rochester DF, Findley LJ. The lungs and neuromuscular and chest wall diseases. In: Murray JF, Nadel JA, editors. Textbook of respiratory medicine. Philadelphia: WB Saunders, 1988: 1942-71. 35. De Troyer A, Pride NB. The respiratory system in neuromuscular disorders. In: Roussos C. Macklem PT, editors. The thorax (Lung biology in health and disease, vol 29, part B). New York: Marcel Dekker, 1985: 1089-121. 36. Gorini M, Spinelli A, Cangioli C, eta/. Control of breathing in patients with short-term primary hypothyroidism. Lung 1989; 167: 43-52. 37. Khaleeli AA, Edwards RHT. Effect of treatment on skeletal muscle dysfunction in hypothyroidism. Clin Sci 1984; 66: 63-8.
July 1993
The American
Journal
of Wledicine
Volume
95
37