July 1989, Volume 39, Issue 7

Original Article


Farida Agha  ( PMRC Research Centre, Jinnah Postgraduate Medical Centre, Karachi-35. )
Huma Qureshi  ( PMRC Research Centre, Jinnah Postgraduate Medical Centre, Karachi-35. )
R.A. Khan  ( Atomic Energy Medical Centre, Jinnah Postgraduate Medical Centre, Karachi-35. )


Levels of thyroid hormones, serum thyroxine, triiodothyronine (13), free thyroxine (FF4), free triiodothyronine (FT3) and thyrotropin (TSH) were measured in 55 patients with liver cirrhosis using radioimmunoassay techniques. Results were compared with 78 controls. The mean serum concentration of 13, FT3 and FF4 were significantly decreased in cirrhotics, while no significant change was noted in serum T4 and TSH levels. T3/T4 ratio was also lower than the normal. This indicates an impaired liver conversion of 14 to T3 in peripheral tissues. Serum T3 and FT3 showed an inverse correlation with serum bilirubin and a positive correlation with serum albumin. 13, FF3 and T3/T4 ratios were significantly low in patients who had ascites as compared to those who had no ascites. This study confirms the presence of abnormalities in serum thyroid hormone levels in cirrhosis of liver. Alteration in serum 13 and FT3 levels correlate well with the disease severity and may be useful in assessing the course and prognosis in cirrhotic patients (JPMA 39:179, 1989).


The liver plays an important role in thyroid hormone metabolism being involved in conjuga­tion, excretion, peripheral deiodination and in sy­nthesis of thyroxine binding globulin. It is now generally accepted that the serum concentrations of thyroid hormones will vary in patients with hepatic disorders, especially in pat­ients with liver cirrhosis.1 In some studies of liver cirrhosis where the peripheral convcrsion of thy­roxine (T4) to triiodothyronine (T3) was exam­ined, the liver was found to be one of the major sites of this peripheral conversion2,3. This study was undertaken to examine the changes of serum levels of thyroid hormones in patients with cir­rhosis and to examine whether the abnormality in the metabolism of thyroid hormones in liver cir­rhosis would parallel the degree of hepatic dam­age or not.


Fifty five clinically, biochemically and biopsy proven patients with cirrhosis were studied. Gas­troscopy, ultrasonography and liver scan were also done in some patients. Standard laboratory tests included prothrombin time, total proteins, albu­min, globulin, bilirubin, alkaline phosphatase, serum aspartate transaminase (AST) and serum alanine transaminase (ALT). In addition presence and absence of ascites was also recorded. Samples of blood were drawn in the morn­ing, serum was separated and after biochemical analysis, stored at -20°C. Stored frozen serum sam­ples were assayed for thyroxine (T4), triiodothy­ronine (T3), thyrotropin (TSH), free thyroxine (FT4) and free thyronine (FF3) by radioimmuno­assay techniques using Amersham Amerlex RIA kits (supplied by Amersham International Ltd. U.K.). Each assay was performed on duplicate serum samples. A set of quality control sera (Amersham U.K.) was also analysed with each assay to check the accuracy and performance of the assay. The radioactivity was counted in Multi-detector computerised Gamma counter (model 1612-Nuclear Enterprises) using a 4 parameter non-linear curve fitting model. As a control group 78 controls of the same age group were inves­tigated for thyroid hormones and TSH.


Results were analysed on a MITAC per­sonal computer, comparison between patients and controls were made using the Student’s “C test. Correlation analyses between the different indices were done and P values were calculated using Fisher and Yates statistical table A “F” value of less than 0.05 was considered to be statistically significant. All results were expressed in SI Units i.e. nmol T3/L, nmol T4/L, pmol FT4/L, pmol FT3/L and TSH ml U/L.


Of 55 patients with cirrhosis 28 were males and 37 were females, the mean age was 41 (s.e.±2.8) years with a range of 5 to 80 years. In 78 controls, (54 males and 24 females) the mean age was 41 (s.e. ± 1.73) years and the range was 14 to 87 years. In most of the cases diagnosis of cirrhosis was made on liver biopsy. Other factors used in the diagnosis of cirrhosis included disturbed liver function tests, presence of portal hypertension, Ultrasonography and liver scintigraphy. Upper G.I. endoscopy showed oesophageal various in 40 cases. Ultrasonography showed changes in the liver parenchyma consistent with cirrhosis in 28 cases while ascites was present in 41 cases. Liver scintigraphy showed reduced patchy uptake of the iso-tope by the liver, splenomegaly with increased uptake and marrow uptake in 10 cases.

Table 1 shows the biochemical findings in cirrhotics5. Bilirubin was normal (< 17.0 p mol/L) in 29 (53%) patients while 18 (33%) patients had values between 17 to 100 u mol/L and 7(13%) had values greater than 100 p moIJL. Serum albumin was low (<3.5 gidl) in 30 (56%) and globulins were elevated (>2.5 g/dl) in 41 (76%) cases. Transferases were elevated in the majority of pat­ients. A prolongation of prothrombin time (> 15 minutes) was noted in 10 (18%) patients. As there was no significant difference in various hormone levels between the two sexes both in patients and controls, the data is reported in total subjects.

Table II shows the mean values for the in-dices of thyroid function in patients with cirrhosis and in controls. Cirrhotic patients showed sig­nificantly reduced mean serum levels of T3, FF3 and FF4. Mean T3/T4 ratio was also significantly low in patients with cirrhosis. Though the mean serum T4 level in cirrhosis was low compared with controls but the difference was statistically insig­nificant. Four (10%) patients had raised T4 levels (>136 nmol/L) and in 14 (33%) T4 was low (<77nmol/L), while 24 (57%) patients had nor­mal T4 levels. Significant difference was found between patients and controls in mean FF4 levels. In 14 (25%) patients the FF4 levels were lower than those of controls. The mean levels of T3 an4 FF3 were sig­nificantly lower in cirrhotic patients. Low T3 levels were found in 14 (25%) patients while FF3 levels were low in 48 (87%) patients. Of the total 48 patients with low FF3, 35 (73%) had normal FF4 and 13 (27%) had low FF4 levels. The mean value of TSH in cirrhotics was above the mean in normal controls but without statistical significance. Serum TSH levels were normal in all patients except 4 cases (4.9,63,63 and 93ml Ut) with levels above the normal range. Simple correlation analysis showed that among the liver function tests, bilirubin, albumin and prothrombin time had significant correlation with thyroid hormones but alkaline phosphatase, transferases, total proteins and globulins had no correlation with any of the thyroid hormones. Serum T3 and FT3 both showed an inverse correlation with bilirubin but the relation was sta­tistically significant only between FT3 and bill­rubin (Figure 1).

A significant positive correlation was also seen between T3 and FF3 v/s albumin (Figure 2);

FF3 showed greater significant correlation (P <0.01) with albumin than T3 (P <0.02). Only FF4 showed a positive significant correlation (r = 0.29 and P <0.05) with prothrombin time. The mean thyroid hormone levels in cir­rhotic patients with albumin levels of less than 3.5 g/dl were compared with those who had normal levels (Table III).

A significant reduction in the mean T3 (P <0.01) and FT3 levels (P <0.001) was seen in patients with low albumin levels. Patients with ascites were found to have. Significantly lower levels of T3 and FT3. T3/T4  ratio was also sig­nificantly lower in these patients is (Table III).


Significant alterations is a serum thyroid hor­mone concentrations haveb been frequently re7por-ted in patients with chronic liver disease,1,2,4-7 and in cirrhosis it depends up on the impaired liver conversion of T4 to T32,3,5 T tie formation of T3 from T4 is catalyzed by iodothy ronine-5/- deio-dinase. This enzyme is located predominantly in the microsomes and: plasma membranes of the liver and kidney. 5,8 Decrease i serum T3 may therefore be contributable to a postulated deficiency of hepatic iodothyronine 5-deiodinase activity5. The decrease in serum T3 found in cirrhotic patients in the present study is in agreement with other reports1,5,6,9. The serum levels of T3 in liver cirrhosis vary from 33 ng/100ml to 90.8ng/100ml2,3,5,6. In present study on mean serum T3 level was 68.9 ng/100ml fairly close to the results of Hepner et al5. The low total T3 with generally normal total T4 and TSH concentrations in the absence of clini­cal hypothyroidism has been frequently reported in patients with chronic liver disease as well as in many other non-thyroidal illnesses1,2,4 and it has sion of T4 to T? . Our data also confirm a highly significant reduction of T3 in liver cirrhosis with generally normal total T4 and ThH concentration. The slight mal total T4 and TSH concentration. The slight decrease in serum total T4 in some patients seen in this study is also reported by other workers5,9-11. This appeared to be due to a reduction in plasma thyroid hormone protein binding capacity with increasing liver dysfunction9 as the serum from patients with alcoholic liver cirrhosis bind thyroid hormones less strongly than normal and have nor­mal or low T4 concentration.10,12 As the peripheral conversion of T4 to T3 occurs mainly in the liver, the T3/T4 ratios in liver diseases were reported lower than normal.16 A similar decrease in T3/T4 ratio is noted in the present study, which confirms that patients with cirrhosis have a re­duced peripheral conversion of T4 to T3. There are only few reports on direct meas­urement of free thyroid hormones in liver pat­ients. 1 Many studies performed with equilibrium dialysis, however, showed decreased FF3 and nor­mal or frequently increased FT4 concentra­tions1,3,5. Green et al. 4 found normal FF3 and FF4 in a small group of cirrhotic patients. However, some studies have reported low FF4 levels in cir­rhotic patients.10,11 The finding of decreased FF3 in our patients may have been due in part to de­creased deiodination of T4 to T3, as reported in other studies1,2,8-10,14 but a significant decrease in mean FF4 found in this study is in contrast to increased or normal FF4 concentrations observed in alcoholic cirrhosis.1,9,12 The reason for this dis­crepancy is unclear. This may reflect the hctero­geneityof the patient population studied (in terms of type and severity of the disease). The different methods employed for estimation of FF4 levels may also account for the variation in results.13 Though we did not measure thyroid antibodies for the presence of autoimmune thyroiditis10 how­ever, the presence of hypothyroidism in our cir­rhotics is unlikely because serum TSH was within the normal range in most of the patients indicating euthyroid function. Elevation of baseline serum TSH has been reported in some patients with, cirrhosis but TSH responses to thyroid releasing hormone (TRH) failed to indicate hypothyroi­dism3,4,8 In cirrhosis, a variable correlation has been reported between abnormalities in liver and thy­roid function tests.4-6,9,14 Several reports, how­ever, documented a significant correlation bet­ween the clinical severity and prognosis of the disease and serum thyroid hormone levels5,9,14. Green et al. 4 found that serum T3 levels in cirrhotic patients correlated with serum albumin values and Israel et al4 reported that T3 levels correlated with the prothrombin time. Orzio et al. 1 found a good correlation between T3 and serum albumin, bilirubin and prothrombin time while there wasno correlation with transaminases andy globulins in cir­rhotics. The excellent correlation in our Study be­tween serum T3 and FF3 and the severity of liver dysfunction as well as a progressive increase in T3 and FF3 in these patients eventually displaying a favourable outcome, suggests that T3 and FF3 con­centrations in patients with advanced liver cir­rhosis may be considered as a helpful prognostic indicator. This study confirms the existence of abnor­malities in thyroid function tests in liver cirrhosis. It also appears that the degree of depression of serum T3 and FF3 correlate well with the severity of liver disease, and maybe helpful in assessing the course and diagnosis of liver cirrhosis.


We are grateful to Mr. Ejaz Alam for statis­tical assistance.


1. Borzio,M., Caldara, R, Borzio, F., Piepoli, V., Rampini, P. and Ferrari, C. Thyroid function tests in chronic liver disease; evidence for multiple abnormalities despite cli­pical cuthyroidism. Gut, 1983; 24:631.
2. Chopra, IJ., Solomon, D.H., Chopra, U., Young, R.T. and Chuateco, U.N. Alterations in circulating thyroid hormones in hepatic cirrhosis; evidence for euthyroidism despite subnormal serum triiodothyronine. J. Clin. En­docrinol. Metab., 1974; 39:501.
3. Nomurn, S., Pittman, C.S., Chambers, J.B., Buck, M.W. and Shimizu, T. Reduced peripheral conversion of thy­roxine to triiodothyronine in patients with hepatic cir­rhosis. 3. Clin. Invest., 1975; 56:643.
4. Green, J.R.B., Snitcher, EJ., Mowat, N.A.G., Eikins, R.P., Rees, L.H. and Dawson,A.M. Thyroid function and thyroid regulation in euthyroid men with chronic liver disease: evidence of multiple abnormalities. Clin. En- docrinol., 1977; 7:453.
5. Hcpner, G.W. and Chopra, IJ. Serum thyroid hormone levels in patients with liver disease. Arch. Intern. Med., 1979; 139:1117.
6. Yamanaka, T., Ido, K., Kimura, K and Saito, T. Serum levels of thyroid hormones in liver diseases. Clin. Chim. Acta, 1980; 101:45.
7. Gardner, D.F., Carithers, RL, Jr. and Utiger, RD. Thy-mid function tests in patients with acute and resolved hepatitis B virus infection. Ann. Intern. Med., 1982; 96:450.
8. Chopra, IJ. A study of extrathyroidal conversion of thy­roxine (T4) to 3,3 5-triiodothyronine (17s) in vitro. En­docrinology, 1977; 101:453.
9. Walfish, P.O., Orrego,H., Israel, Y., Blake,J. and Kalant, H. Serum triiodothyronine and other clinical and labo­ratory indices of alcoholic liver disease. Ann. Intern. Med., 1979; 91:13.
10. Schussler, G.C., Scheffner, F. and Korn, F. Increased serum thyroid hormone binding and decreased free hor­mone in chronic active liver disease. N. Engl. 3. Med., 1978; 299:510.
11. Hollander, D., Meek, J.C. and Manning, R.T. Deter­mination of free thyroxine in serum of patients with cirrhosis of the liver. N. EngI. J. Med., 1967; 276:900.
12. Chopra, IJ., Chopra, U., Smith, S.R, Reza, M. and Solo­nion, D.H. Reciprocal changes in serum concentrations of 3, 3,5- triiodothyronine (reverse Ts) and 3, 3,5 - triio­dothyronine fF3) and systemic illnesses. 3. Clin. Endoc­rinol. Metab., 1975; 41:1043.
13. Melmcd, S., Geola, F.L., Reed,A.W., Pekary, A.E., Park, J. and llersbman, J.M. A comparison of methods for assessing thyroid function in nonthyroidal illness. 3. Clin. Endocrinol. Metab., 1982;54:300.
14. Israel,Y.,Walfish, P.O., Orrego, H., Blake, 3. and Kalant, H. Thyroid hormones in alcoholic liver disease; effect of treatment with 6-propylthiouracil. Gastroenterology, 1979; 76:116.

Journal of the Pakistan Medical Association has agreed to receive and publish manuscripts in accordance with the principles of the following committees: