Thyroid disease is a common condition, and thyroid hormone excess or deficiency is known to have wide-ranging effects on a variety of organ systems. Our objective is to describe the magnitude, biochemical features, and clinical characteristics of hepatic abnormalities in patients with acute thyrotoxicosis. We performed a retrospective review of all patients admitted to our institution between January 1, 1998 and December 31, 2008 with a discharge diagnosis of acute thyrotoxicosis excluding iatrogenic causes. The records of these patients were reviewed and data extracted regarding demographic, biochemical, and clinical data particularly relevant to liver function. Fourteen patients were identified of which eleven had liver studies performed. The majority (90.9%) had Graves disease. Nine of eleven patients (81.8%) had some degree of hepatic abnormality. Seven patients (63.6%) had an elevation in one or both transaminases, and two (18.2%) had isolated synthetic dysfunction as manifested as an elevated INR and/or decreased albumin without transaminitis. The mean magnitude of deviation from the normal range was greater in the transaminases as compared to bilirubin, INR, or albumin. Definitive treatment was radioiodine ablation in six cases (54.5%) and surgical thyroidectomy in two cases (18.2%). Noniatrogenic acute thyrotoxicosis requiring hospitalization is a rare condition which is most frequently caused by Graves disease. The majority of patients have disordered liver tests of a highly variable nature, making the recognition of this association important in the care of patients presenting with acute thyrotoxicosis.
Thyroid disease is a common condition, and thyroid hormone excess or deficiency is known to have wide-ranging effects on a variety of organ systems. Liver failure in hyperthyroidism was first reported by Habershon in 1874 [
This research was approved by the Mayo Clinic Institutional Review Board (IRB ID no. 08-008253). A review was undertaken of all electronic discharge summaries for patients 18 years of age or older at our institution between January 1, 1998 and December 31, 2008. A search of discharge diagnoses including the terms “acute thyrotoxicosis,” “thyroid storm,” or “thyroid crisis” yielded 130 discharge documents. Fifty-two of these documents were found to be either duplicates or documentation of transfer from one hospital service to another during the same admission. The electronic medical records of the remaining patients were manually reviewed. In an effort to identify patients with clinically important and well-documented thyrotoxicosis as a cause of admission, additional patient records were excluded as follows: thirty patients were excluded because review indicated that hyperthyroidism was not a principal element of their admission, and they did not have clinical instability or organ dysfunction potentially attributable to acute thyrotoxicosis; eight admissions were excluded because they were for elective thyroidectomy in patients who had previously had acute hyperthyroidism; twenty-six patients were excluded because the principal etiology of their hyperthyroidism was determined to be amiodarone effect, making interpretation of liver function test (LFT) abnormalities difficult due to the potential for amiodarone to cause liver enzyme changes directly. This resulted in a pool of fourteen patients, which comprised the group for the review. Data was extracted from patient records including serum thyroid indices and LFTs as well as markers of hepatic synthetic function, namely, serum albumin and international normalized ratio (INR). The presence or absence of ascites was noted. In order to determine the extent to which liver abnormalities were secondary to high-output cardiac failure, echocardiogram and troponin data were recorded when available. As propylthiouracil (PTU) has been associated with cases of liver damage and failure, determination was also made as to whether the patient had been commenced on PTU prior to the liver testing. If patients were treated with PTU following testing, subsequent liver tests were assessed to determine if the values worsened after commencement of PTU.
Fourteen patients were identified as having been admitted for de novo acute thyrotoxicosis over this period. Of this number, eleven patients (78.6%) had LFTs performed during their admission (Table
Demographic, laboratory, echocardiographic results, and propylthiouracil use by patient.
Patient number | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 |
---|---|---|---|---|---|---|---|---|---|---|---|
Age | 81 | 50 | 55 | 21 | 26 | 23 | 42 | 42 | 56 | 55 | 50 |
Gender | F | F | F | M | F | M | M | M | M | F | F |
TSH (mIU/L) [NR 0.3–5.0 mIU/L] | 0.02 | <0.01 | <0.01 | <0.01 | <0.01 | <0.01 | <0.01 | <0.01 | 0.01 | <0.01 | 0.01 |
Thyroxine (total; mcg/dL) [NR 5–12.5 mcg/dL] | NP | NP | 11.5 | NP | 23.2 | 22.3 | NP | NP | 16.2 | NP | NP |
Thyroxine (free; ng/dL) [NR 0.8–1.8 ng/dL] | 10 | 4.6 | 6 | 7.7 | 7 | >12 | 5.4 | 9.4 | 3.2 | 5 | 3.6 |
Triiodothyronine (ng/dL) [NR 80–190 ng/dL] | NP | NP | 462 | 385 | NP | 359 | NP | NP | 150 | NP | NP |
Free T3 (pg/mL) [NR 2.0–3.5 pg/mL] | 20 | 8.6 | NP | 17.6 | 10.3 | >20 | 17.6 | NP | NP | 12.3 | 43 (reverse) |
TPO (IU/mL) [NR < 9.0 IU/mL] | <20 | 287 | NP | >950 | NP | 4420 | NP | 3017 | 706 | NP | NP |
TRAB (%) [NR < 16%] | 82 | 31 | 66 | NP | NP | 93 | 46 | 75 | 52 | NP | 53 |
TSI index (%) [NR ≤ 1.3%] | NP | NP | NP | 7.3 | NP | NP | NP | NP | NP | 5 | 2.6 |
AST (U/L) [NR 8–48 U/L] | 76 | 978 | 636 | 2850 | 39 | 30 | 82 | 34 | 99 | 52 | 69 |
ALT (U/L) [NR 7–55 U/L] | 66 | 920 | 841 | 1895 | 41 | 27 | 102 | 53 | 70 | 41 | 44 |
ALP (U/L) [NR 45–115 U/L] | 135 | 252 | 136 | 180 | 129 | NP | 97 | NP | 194 | NP | 162 |
Bilirubin (Total/Dir.; mg/dL) [NR 0.1–1.0/0.0–0.3 mg/dL] | NP | 1.2/0.6 | 0.9/0.7 | 5.6/1.8 | 0.4/0.1 | NP | 0.7/0.2 | NP | 1.2/0.7 | NP | 3.6/2.5 |
INR [NR 0.8–1.2] | NP | 1.2 | 1.4 | 2.4 | 1.1 | 1.0 | 1.0 | 1.0 | 1.1 | 1.7 | 1.7 |
Albumin (g/dL) [NR 3.5–5.0 g/dL] | NP | 2.9 | 3.6 | 3.6 | NP | NP | 3.3 | NP | 3.4 | 3.1 | 3.0 |
Ascites | N | N | N | Y | N | N | N | N | Y | N | N |
LVEF by ECHO | NP | 48% | 65% | 20% | NP | NP | 68% | 45–50% | 22% | 29% | NP |
Troponin T (ng/mL) | NP | 12.38 | 0.07 | <0.01 | <0.01 | <0.01 | <0.01 | 0.11 | <0.01 | NP | NP |
PTU prior to LFTs? | N | N | N | N | Y | N | N | N | N | Y | N |
PTU after LFTs? | Y | Y | N | N | Y | Y | N | N | Y | Y | Y |
LFTs worsened? | N | N | N/A | N/A | N | N | N/A | N/A | N | N | N |
Y: yes, N: no, NP: not performed, N/A: not applicable.
Mean and median values of hepatic indices.
AST (U/L) | ALT (U/L) | AST : ALT | ALP1 (U/L) | Bilirubin (total/dir.; mg/dL) | INR | Albumin (g/dL) | ||
| ||||||||
All patients ( |
Mean | 450 | 373 |
1.11 | 161 | 1.9/0.9 | 1.4 | 3.3 |
SD | 854.04 | 603.74 | 0.31 | 48.04 | 1.93/0.88 | 0.45 | 0.28 | |
Median | 76 | 66 | 1.11 | 149 | 1.2/0.7 | 1.15 | 3.3 | |
| ||||||||
Patients with abnormalities | Mean | 605 | 564 | 1.18 | 170 | 2.9/1.3 | 1.8 | 3.0 |
SD | 969.73 | 700.19 | 0.34 | 43.83 | 2.13/0.85 | 0.42 | 0.21 | |
Median | 90.5 | 102 | 1.15 | 162 | 2.4/0.7 | 1.7 | 3.1 |
Mean and median degree of variation from the normal range of laboratory indices expressed as a multiple of the upper limit of normal (lower limit of normal for albumin). AST = aspartate aminotransferase; ALT = alanine aminotransferase; ALP = alkaline phosphatase; INR = International Normalized Ratio.
It is known that thyroid hormones affect a variety of tissues via numerous established mechanisms. This investigation highlights the clinically important association between excess thyroid hormone and liver abnormalities. Documented acute thyroid crisis appears to be a rare condition, with only fourteen noniatrogenic cases identified at a large tertiary referral center over an eleven-year period. Of note, the majority of admissions labeled as acute thyrotoxicosis appear to be secondary to amiodarone with the majority of the remaining cases secondary to Graves disease. Importantly, most of these patients present with some degree of hepatic dysfunction and LFT abnormality which is highly variable in character and severity. This is in keeping with previous reports demonstrating that mild derangements in LFTs are common even in patients with subclinical hyperthyroidism but that hepatic failure is rare [
Examination of the LFT patterns in these patients demonstrates a high degree of variability. The most common LFT pattern is characterized by a transaminitis out of proportion to synthetic dysfunction. This is not invariably the case, however, with two patients demonstrating rises in INR without any marked abnormality of the transaminases and one patient with a mixed, predominately conjugated, hyperbilirubinemia and an elevated INR without transaminitis. Other investigators have noted a pattern of cholestatic liver dysfunction, with case reports documenting severe jaundice as the predominant clinical feature [
Notably, while all eleven patients comprising this study population had very depressed or undetectable thyroid-stimulating hormone levels, the circulating levels of thyroxine, free triiodothyronine, and antibodies to thyroid peroxidase varied considerably. There does not appear to be a correlation between the degree of thyroid function abnormalities and either the nature or magnitude of liver dysfunction, an observation that is in keeping with the results of previous investigations [
The mechanism for an association between thyroid hormone excess and hepatic dysfunction is unclear and may be due either to indirect pathways or, alternatively, to direct hormone effects on the target organ, such as an interaction with nuclear receptors to modify gene expression or through receptors at the plasma membrane, mitochondria, and other extranuclear sites [
Investigations of the syndrome of resistance to thyroid hormone (RTH) have been indirectly instructive regarding the implications of thyroid hormone excess in the liver [
Apart from direct hepatocellular effects, thyrotoxicosis may cause secondary hemodynamic insults that can perturb liver function. Congestive heart failure exacerbation, often secondary to atrial fibrillation but also described in sinus tachycardia, has a recognized association with hyperthyroidism. Studies of this association have found it to be an uncommon manifestation of hyperthyroidism generally, although research specifically investigating patients with acute thyrotoxicosis is limited [
It has been recognized since studies performed in the 1950s that splanchnic metabolism, as measured by oxygen consumption, is elevated in hyperthyroidism [
While this analysis represents the largest recent report detailing liver dysfunction in the setting of thyrotoxicosis, there are several limitations to this study. Broad applicability of these findings is limited by the disparate definitions of acute thyrotoxicosis, thyroid storm, or thyroid crisis in the literature, including definitions requiring that the episode be acutely life-threatening, and others dependent on decompensation of one or more organ systems to meet the definition [
Another potential confounder is the increasing recognition of the association between autoimmune thyroid disease and autoimmune liver disease. While none of the patients in our study had diagnoses of autoimmune hepatitis, primary sclerosing cholangitis or primary biliary cirrhosis, testing was not universally conducted to specifically exclude this possibility. Similarly, none of the included patients had known viral or other primary liver diseases but these potential confounders cannot be excluded.
While hyperthyroidism is a common condition, acute, severe thyrotoxicosis without an iatrogenic precipitant is uncommon and most frequently associated with Graves disease. These patients are present with liver abnormalities which are variable in character and magnitude but appear to have a hepatitic predominance. Evidence of thyroid dysfunction should be sought in cases of unexplained hepatic abnormalities, particularly transaminitis. There remains lacking a standardized mechanism for diagnosis and grading of acute thyrotoxicosis and the development of this would facilitate further systematic study of the end-organ effects of this condition.
Drs. Elias, Dean, and Barsness report no material support and financial or personal conflict of interests associated with this paper.
The authors would like to thank Anna Elias for her invaluable assistance in the preparation of the paper.