Clin Biochem, Vol. 21, pp. 245-247, 1988 0009-9120/88 $3,00 + .00 Printed in Canada. All rights reserved. Copyright © 1988 The Canadian Society of Clinical Chemists.

Glycemic Control and Raised Serum Alanine Aminotransferase Activity in Treated Diabetes Mellitus

JEAN HENDERSON, WILLIAM D. NEITHERCUT, RICHARD J. SPOONER, and BRIAN M. FRIER

Departments of Pathological Biochemistry and Diabetes, Gartnavel General Hospital, Glasgow, Scotland

The prevalence of raised serum liver-associated enzyme activity in stabilised, treated diabetic outpatients without concurrent hepa- tobiliary disease was investigated using a retrospective computer search of biochemical data. The frequency of raised alanine ami- notransferase, aspartate aminotransferase, or gamma glutamyl transferase activity found among diabetic, general medical and res- piratory outpatients was compared with that found in apparently healthy controls. It was established that a raised activity of any of the three enzymes occurred with a similar frequency in each out- patient group. However, only with alanine aminotransferase did the frequency of elevation (7.1%) in the patients with previously diag- nosed hepatobiliary disease exceed that of healthy controls. A raised alanine aminotransferase activity in diabetic outpatients was asso- ciated with good glycemic control (hemoglobin A~ < 8%, p < 0.02) and treatment with oral hypoglycaemic agents (p < 0.001).

KEY WORDS: diabetes mellitus; alanine aminotrans- ferase; glycaemic control.

R aised activities of the serum enzymes, aspartate aminotransferase (AST; EC 2.6.1.1.), alanine ami-

notransferase (ALT; EC 2.6.1.2), and gamma glutamyl transferase (GGT; EC 2.3.2.2) have been found in un- selected stabilized diabetic outpatients (1-3). The re- lationship of these abnormalities to diabetes mellitus remains unclear, as no specific liver disease is associ- ated with diabetes mellitus (4). Significant liver disease is uncommon among unselected diabetics (2). Raised activity of these enzymes is found in newly diagnosed diabetics and in preketoacidotic states (2,3). The liver has a major role in glucose homeostasis and raised ac- tivity of these liver-associated enzymes in stabilized diabetic outpatients may also be related to glycemic control. The development of these enzyme abnormali- ties in diabetics may however reflect the fact that di- abetes is a multisystem disease and that diabetics may not only be receiving treatment to optimize glycemic control but also for other complications. We, therefore report a retrospective study in which the prevalence of raised enzyme activities in a treated diabetic outpatient

Correspondence: Dr. R. J. Spooner, Department of Patho- logical Biochemistry, Western Infirmary, Glasgow Gll 6NT, Scotland.

Manuscript received September 8, 1987; revised January 4, 1988; accepted January 4, 1988.

group was compared with medical and respiratory out- patients. The relationships between enzyme activities and glycemic control and therapeutic regimen in these stable diabetics were also examined.

Materia ls and m e t h o d s

The archived data from diabetic, general medical, and respiratory clinics stored on the departmental com- puter were identified by computer searches over a four month period (May to August). Abnormalities of AST, ALT, and GGT in serum were identified in these clinic populations and the percentage glycated hemoglobin (HbA1) coincident with the enzyme measurements was extracted for the diabetic outpatients. Serum AST, ALT, and GGT activities had been measured at 37°C using the International Federation of Clinical Chemistry rec- ommended methods on a SMAC I (Technicon, Tarry- town, NY, U.S.A.). Percentage HbA1 was measured by electroendosmosis and densitometry (Corning Medical, Halstead, U.K). The enzyme activities were defined as abnormal if they were outside the departmental reference ranges (AST < 36 IU/L, ALT < 50 IU/L, GGT < 50 IU/L).

The case records of the 66 diabetic patients with en- zyme abnormalities were reviewed to ascertain their age, diabetic treatment regimen, treatment with other drugs, the presence of hepatic or biliary disease, and alcohol consumption.

A survey of 1635 individuals registered with a family doctor from an apparently healthy population provided control data.

Resul t s

The number of patients attending diabetic, general medical, and respiratory clinics who had raised serum AST, ALT, and GGT activities are shown in Table 1. These data were similar at each of the clinics (×2NS), but when compared with the healthy population, the prevalence of abnormality in all outpatient populations was greater for each enzyme (p < 0.001). When diabetic patients with known hepatobiliary disease were ex- cluded, the frequency with which AST and GGT were

CLINICAL BIOCHEMISTRY, VOLUME 21, AUGUST 1988 245

HENDERSON, NEITHERCUT, SPOONER, AND FRIER

TABLE 1 Incidence of Raised Serum Enzyme Activities from Medical, Respiratory, and

Diabetic Outpatients (OP)

Percentage Incidence of Abnormality a

Patient Group ALT AST GGT

Healthy Individuals 1.4 [22/1630] 3.3 [54/1627] 9.1 [148/1635] General Medical OP 9.5 b [28/293] 5.8 b [17/293] 17.7 b [49/276] Respiratory OP 7.8 b [9/114] 7.8 b [9/114] 19.1 b [22/115] Diabetic OP 9.8 b [37/376] 7.3 b [33/447] 15.6 b [66/421] Diabetic OPc 7.1 b [26/365] 3.0 [13/427] 7.7 [30/385]

aAbsolute figures given in brackets. bp < 0.001, ×2, when compared with healthy individuals. CDiabetics without identifiable hepatobiliary disease.

raised was not significantly grea ter than tha t found in the hea l thy group. However, ALT activit ies were still elevated more frequently (p < 0.001). The apparent ly heal thy population provided sex and age matched data. Matching for sex showed no significant differences for any enzyme abnormali ty . ALT activity showed no cor- relat ion with age.

The diabetic pa t ients wi thout hepatobi l iary disease were subdivided on the basis of HbA~ level (Table 2). Excellent glycemic control (HbA1 < 8%, within our non- diabetic range) was associated with a significantly in- creased prevalence of raised ALT and GGT activity compared to less s t r ingent glycemic control (HbA1 I> 8%) (p < 0.02). This difference was not observed if the diabetic group was subdivided on the basis of "accept- able" glycemic control (HbA~ < 9.5%) (5) versus poor control (HbA~/> 9.5%).

The relat ionship of raised enzyme activi ty to diabetic therapy was examined in the diabetic pat ients without hepatobi l iary disease. Of the pat ients with raised en- zyme activities, 64% (21 of 33 patients) were tak ing oral hypoglycemic agents (mean age 63 -+ 8 years), 18% (6 of 33) were insul in-dependent (mean age 57 +- 16 years), and 18% (6 of 33) were on diet alone (mean age 68 _+ 15 years). These proportions were significantly different (p < 0.001) from diabetic pat ients with normal enzyme activit ies with respect to t r ea tmen t regimen, with 46% (100 of 217) t ak ing oral hypoglycemic agents,

51% (111 of 217) requir ing insulin, and 3% (6 of 217) t rea ted by diet alone.

Discuss ion

An increased incidence of raised activity of the three enzymes, AST, ALT, and GGT, in se rum from diabetic outpat ients has previously been reported (1-13). How- ever, the incidence of increased activity of the three enzymes was the same in serum from pat ients from three different outpat ient groups, indicating a back- ground of raised enzyme activity which, in diabetic out- patients, could have been due to other concurrent illnesses. When diabetic outpat ients with previously diagnosed hepatobi l iary disease or other factors rec- ognized to cause increased liver-associated enzyme activity were excluded, then unlike previous in- vest igators (1,2) we found a small increase in the in- cidence of raised se rum ALT activity (7.1%).

The proportion of abnormali t ies found for GGT ac- t ivi ty in the hea l thy population exceeded tha t statis- tically predicted from the 95% confidence interval used in our laboratories. As these individuals were clinically well, the raised enzyme activities probably represent induction of l iver associated enzymes by the consump- tion of alcohol in the population studied. As the hospital outpat ient groups are drawn from a similar population

TABLE 2 Relationship of the Incidence of Raised Enzyme Activity in Diabetic

Outpatients, Without Identifiable Hepatobiliary Disease, to Glycemic Control

Percentage Incidence of Abnormality a

ALT AST GGT

HbAI < 8% 11.7 b [14/119] 4.2 [6/140] 12.6 b [16/126] HbA~ ~> 8% 4.8 [12/246] 2.4 [7/287] 5.4 [14/259]

HbA1 < 9.5% 6.3 [15/242] 2.5 [7/278] 7.0 [18/255] HbA1 >~ 9.5% 8.9 [11/123] 4.0 [6/149] 9.2 [12/130]

aAbsolute figures given in brackets. bp < 0.02, ×2, vs HbA1/> 8%.

246 CLINICAL BIOCHEMISTRY, VOLUME 21, AUGUST 1988

DIABETES MELLITUS

a better assessment of the significance of raised enzyme activity is possible with the hea l thy population t han with our laboratory reference intervals.

Raised serum ALT activi ty in diabetics was related to good glycemic control, as assessed by coincidental HbA1 values, as well as to t r e a t m e n t with oral hypo- glycemic agents. These relat ionships may have oc- curred secondary to t r ea tmen t with oral hypoglycemic agents or in association with Type II diabetes melli tus. No single oral hypoglycemic agent could be implicated among the small numbers of patients. Nor were these patients main ta ined on max imal doses of oral hypo- glycemic agents.

Prospective identification and invest igat ion of sta- bilized diabetic outpat ients t rea ted with oral hypogly- cemic agents and with good glycemic control would indicate whether this small number of pa t ients with raised ALT activity represent those with undiagnosed hepatobiliary disease or a re lat ive toxicity with oral hypoglycemic agents.

In conclusion, raised liver-associated enzyme activity due to diabetes is uncommon. Raised ALT activity in diabetic outpat ients may be secondary to t r ea tmen t with oral hypoglycemic agents if there is no under lying hepatobiliary pathology.

Acknowledgements

We thank Dr. K. Cunningham, Monklands District General Hospital, Airdrie for the provision of data on healthy indi- viduals, and Albert Tytherleigh and Liz McCulloch for the archive analysis routines.

References

1. Salmela PI, Sotaniemi EA, Niemi M, Maentausta O. Liver function tests in diabetic patients. Diabetes Care 1984; 7: 248-54.

2. Foster KJ, Griffiths AH, Dewbury K, et al. Liver disease in patients with diabetes mellitus. Postgraduate Med J 1980; 56: 767-75.

3. Goldberg DM, Martin JV, Knight AH. Elevation of serum alkaline phosphatase activity and related enzymes in di- abetes mellitus. Clin Biochem 1977; 10: 8-11.

4. Stone BG, van Thiel DH. Diabetes mellitus and the liver. Semin Liver Dis 1985; 5: 8-28.

5. Goldstein DE, Parker KM, England ID, et al. Clinical ap- plication of glycosylated hemoglobin measurements. Dia- betes 1982; 31 (Suppl 3): 70-8.

CLINICAL BIOCHEMISTRY, VOLUME 21, AUGUST 1988 247