Introduction
Tuberculosis remains a worldwide infectious problem [1 ]. Treatment should follow recently published recommendations [2 ].
SIADH is a possible cause of hyponatremia in patients with pulmonary diseases – especially
in granulomatous diseases. However, the number of case reports on patients with a
SIADH exclusively related to pulmonary TB is low and the mechanism poorly understood.
Hypoxia, decreased vascular volume [3 ] and ectopic ADH production have been discussed as possible causes [4 ].
The case
A forty-nine-year-old cachectic woman (height 1.60 m, weight 30.4 kg, BMI 11.9 kg/m2 ) was admitted to our hospital due to strong suspicion of pulmonary tuberculosis (TB)
([Fig. 1 a, b ] ). The patient was an active smoker (17 pack-years) until the day of admission to
our hospital. For a couple of years, she had been abstinent from alcohol abuse.
Fig. 1 Chest radiographs, day after admission 0 (a, b ) and 133 (c, d ). Features of extensive lung emphysema and variable sized nodular areas of consolidation
within the upper parts of both lungs. Moreover, an area of low density visible in
the left upper lobe (a, b ). Marked decrease of consolidation with only minor parenchymal scarring in both upper
lobes (c, d ).
Initial microbiological examination revealed acid-fast bacilli in auramine staining,
confirmed by Ziehl-Neelsen stain. Antituberculotic treatment with isoniazide (H),
rifampicine (R), ethambutole (E) and pyrazinamide (Z) was initiated following approved
protocols [2 ]
[5 ]. Later on, cultural results showed mycobacterium tuberculosis (MTB) sensitive to
H and R.
From the day of her admission, we observed hyponatremia. Later on, the patient developed
severe hypokalemia ([Table 1 ]). Clinically, these electrolyte imbalances caused no specific symptoms. For treatment
of hyponatremia, fluid intake was consequently restricted to a total of 1000 ml per
day. Oral potassium substitution was started but did not result in normalization of
hypokalemia. Therefore, continuous parenteral potassium administration via a central
line was initiated, again without sustained success.
Table 1
Evolution of laboratory parameters
Day after admission
0
17
20
35
57
78
101
118
134
239
Laboratory reference range
Serum Na +
125
125
126
125
130
136
136
140
138
139
135 – 145 mmol/L
Serum K +
4.8
4.1
3.5
2.3
4.0
4.9
5.1
4.1
3.8
4
3.5 – 5 mmol/L
Serum Osmolality
244
244
258
260
296
295
280 – 295 mos/kg
Urine Osmolality
317
272
426
248
50 – 1200 mos/kg
Urine Na +
48
40
< 10
77
80 – 164 mmol/L
Urine K +
53
37
94
17
32 – 74 mmol/L
CRP
1809
161.8
76.1
238.0
161.8
209.4
114.2
57.1
123.8
28.6
0 – 95 nmol/L
ADH
30.5
< 0.5
0.7 – 4.2 pmol/L
SUSPUP
33.66
50.27
7.67
3.6 – 22.6
SUSPPUP
8.21
21.86
1.92
0.6 – 5.3
ACTH
0.4
11.9
2 – 11.4 pmol/L
Cortisol
504.9
1594.7
193 – 689.8 nmol/L
Renin
3.0
0.1 – 0.36 pmol/L
Aldosterone
0.59
0 – 0.42 nmol/L
H through level
< 1.5
< 1.5
< 1.5
< 1.5
5.1
18.2[1 ]
1.5 – 7.3 µmol/L
R through level
2.9
3. 9
< 0.1
0.5
2.1
2.2
0.1 – 12.2 µmol/L
Day 20: Start intravenous antituberculotic treatment, Day 28 – 35: tolvapatan administration
1 H peak level (laboratory reference range 10,9 – 72,9 µmol/l); H through level controlled
on day 135 after admission: 4,4 µmol/l
Biochemical analysis revealed serum osmolality being reduced, urine osmolality normal
and sodium urine concentration reduced, fulfilling formal criteria for syndrome of
inadequate antidiuretic hormone secretion (SIADH) with hypoosmolar hyponatremia. Reduced
sodium urine concentrations with consecutively elevated SUSPUP (serum sodium to urinary
sodium to serum potassium to urinary potassium ratio) and SUSPPUP (serum sodium to
urinary sodium to [serum potassium]2 to urinary potassium ratio) [6 ] suggested reactive hyperaldosteronism as a compensatory mechanism. There were no
clinical hints for an extrarenal sodium loss. Renal function was slightly reduced
during the whole in-hospital stay, thyroid gland function normal. In the absence of
oedema, there were no signs of heart failure. Hemodynamically, the patient was stable
at all times.
Further laboratory evaluations revealed antidiuretic hormone (ADH) being markedly
increased as well as serum renin and aldosterone levels, while aldosterone-renin ratio
was normal. These findings confirmed our diagnosis of SIADH with secondary hyperaldosteronism
as a compensatory mechanism for hyponatremia. With regard to etiology of the SIADH,
there were no hints for medication-related effects or a possible malignant disease.
Also, no central nervous affections were found, so we had reason to assume the SIADH
to be due to the pulmonary TB. At no point of hospitalization, we had clues for extrapulmonary
tuberculosis.
Despite antituberculotic treatment using directly observed therapy (DOT) from the
very beginning, sputum analyzes revealed still significant smear-positive TB and cultural
results MTB being furthermore multi-sensitive to first-line antituberculotic medication.
We excluded interaction with co-medication and reconfirmed alcohol and tobacco-smoke
abstinence. Additionally and pragmatically, we switched antituberculotic treatment
from oral to intravenous administration at this point of the hospitalization to exclude
malabsorption as a reason for treatment failure.
Due to persisting hyponatremia we started treatment with tolvaptan. Consequently,
hydration restriction was stopped. Surprisingly, tolvaptan remained ineffective and
was stopped after eight days of unsuccessful treatment.
After five weeks of intravenous antituberculotic therapy, sodium and potassium levels
increased and normalized finally in parallel with clinical and radiologic response
to the antituberculotic treatment ([Fig. 1 c, d ]). Moreover, several consecutive sputum samples were negative for MTB. As a further
effect of successful treatment, the patient gained weight (39.4 kg at the end of hospitalization).
The reason for the difficult antituberculotic treatment despite proven multi-sensitivity
of MTB may be seen in sub-therapeutic H-levels measured repeatedly during treatment
and necessitating dose adjustments. Probably, the patient belongs to a group of fast-acetylators
(polymorphism of the NAT2 -gene). Written consent for further genetic testing was denied.
In view of the clinical course, we are convinced that the SIADH was directly attributable
to pulmonary TB. As part of follow-up, the patient presented at our hospital three
months after discharge, i. e. 7.5 months after first admission. Under continued antituberculotic
treatment, there were no signs of infectious pulmonary TB. Sodium and potassium levels
were normal without substitution or further specific therapy.
Discussion
Tolvaptan is a nonpeptide, vasopressin V2 receptor antagonist, a new class of pharmaceuticals for treatment of hyponatremia,
especially when SIADH-induced [7 ]. These aquaretics selectively antagonize the antidiuretic effect of vasopressin
by competitively binding to renal V2 receptors.
The mechanism of TB-related SIADH is poorly understood. Notably, a correlation between
floridity of TB and severity of SIADH has not been examined so far. Thus, it is difficult
to conclude unambiguously, why tolvaptan remained ineffective in this case. A probable
explanation is drug-interaction of R and tolvaptan. R is a strong inductor of cytochrome
P450-dependent monooxygenases (CYP), notably CYP3A4, 1A2, 2C9, 2C8 and 2C18/19 in
intestinal epithelium and liver [8 ]. Tolvaptan is a sensitive CYP3A4 substrate with no inhibitory activity. In healthy
subjects, mean maximum concentration Cmax and AUC of tolvaptan were reduced when co-administered with R [9 ].
Gene polymorphisms of N-acetyltransferase2 (NAT2 ) are known to cause individual variation in N-acetylation capacity of H. In a study
of H pharmacokinetics and pharmacodynamics most homozygous fast acetylators needed
a dose twice as high as homozygous slow acetylators to achieve equivalent AUC and
2-hours H serum concentration [10 ].
Acetylation status has not been described to influence the metabolism of tolvaptan,
and CYP3A4 has previously been identified as the only enzyme to be involved in the
metabolism of this aquaretic drug [9 ]. Therefore, it seems unlikely that the reduced effectiveness of tolvaptan is caused
by a polymorphism of NAT2 .
Conclusion
In this case of TB-related SIADH, tolvaptan remained ineffective. A probable reason
is seen in a clinically relevant drug-interaction of rifampicine and tolvaptan. Consequently,
dose adjustments of tolvaptan may be necessary when co-administered with R. This would
be in agreement with results of a previous trial in healthy volunteers [9 ]. Further studies on the mechanisms of TB-related SIADH and interactions of antituberculotic
and aquaretic drugs in this population are necessary.
Author contributions
All authors planned the work and interpreted the results. H. Knoop, U. Knoop, J. W.
Dietrich and J. Behr wrote the article. C. M. Heyer, S. Kuert, D. Roggenland and M.
Suermann made substantive suggestions for revision. All authors approved the final
version.
