Reply

Authors


  • Potential conflict of interest: Nothing to report.

Reply:

We are pleased with the interest in our article[1] and appreciate the valuable comments made by Dr. S. Singh and Dr. P. P. Singh. They eloquently discuss the development of hepatocellular carcinoma (HCC), including the effects of antidiabetic medications (ADMs), in patients with diabetes mellitus (DM).

In this study,[1] we concluded that DM is associated with a 1.7-fold higher risk of HCC as compared to patients without DM. In addition, the risk of HCC was reduced by 44% in diabetic patients with mean hemoglobin A1c (HbA1c) < 7.0% during follow-up compared with those with mean HbA1c  ≥ 7.0%. However, we could not sufficiently evaluate the difference in clinical backgrounds between diabetic patients with mean HbA1c < 7.0% during follow-up compared with those with mean HbA1c ≥ 7.0% in our previous study. Thus, we evaluated the difference in detail, in clinical backgrounds that included the use of ADMs in both groups.

Table 1 shows the differences in the clinical backgrounds of diabetic patients with a mean HbA1c level < 7.0% during follow-up and diabetic patients with a mean HbA1c level ≥ 7.0%. There were no significant differences between the two groups except in the use of ADMs. Approximately two-thirds of the 127 diabetic patients with a mean HbA1c level < 7.0% were treated with ADMs. On the other hand, 132 of the 140 diabetic patients with a mean HbA1c level ≥ 7.0% were treated with ADMs. Thus, the prevalence of the use of ADMs in diabetic patients with a mean HbA1c level < 7.0% was significantly lower than the prevalence in diabetic patients with a mean HbA1c level ≥ 7.0%. This result suggests that in comparison with the diabetic patients with a mean HbA1c level ≥ 7.0%, diabetes could be controlled with diet and exercise alone for some of the diabetic patients with a mean HbA1c level < 7.0%.

Table 1. Clinical Background of Enrolled Patients With Type 2 DM at the Initiation of Follow-up
 Diabetic Patients With a Mean HbA1c Level < 7.0%Diabetic Patients With a Mean HbA1c Level ≥ 7.0%P Valuea
  1. a

    Clinical differences between diabetic patients with a mean HbA1c level < 7.0% and diabetic patients with a mean HbA1c level ≥ 7.0% were evaluated with the Wilcoxon rank-sum test or Fisher's exact test.

  2. b

    The smoking index is defined as packs per day times years. The total alcohol intake and the smoking index indicate the sums before and after the first consultation.

  3. Abbreviations: LC, liver cirrhosis; SVR, sustained virological response.

Patients (n)127140 
Age (years)56.1 ± 9.155.9 ± 9.20.874
Sex: male/female (n/n)91/36103/370.726
Body mass index (kg/m2)23.1 ± 3.123.7 ± 4.80.617
Total alcohol intake (kg)b151 ± 102147 ± 1010.842
Smoking indexb10.9 ± 12.010.8 ± 11.20.816
Efficacy: SVR/no SVR (n/n)51/7643/970.124
Staging: LC/no LC (n/n)30/9739/1010.484
Aspartate aminotransferase (IU/L)46 ± 3150 ± 500.887
Alanine aminotransferase (IU/L)56 ± 4761 ± 650.513
Antidiabetic medications: yes/no86/41132/8<0.001

As for the development of HCC in diabetic patients treated with ADMs, it was difficult to assess the effect of ADMs because of the study's retrospective nature. The reason for this is that the diabetic patients treated with ADMs were given various types and doses of ADMs for different periods. Thus, we could not evaluate the effect of ADMs in this study. The effect of ADMs on the development of HCC in diabetic patients is a very important issue. From now on, a more prescriptive approach will be needed to solve this issue in patients with hepatitis C virus treated with interferon.

  • Yasuji Arase, M.D., Ph.D.

  • Mariko Kobayashi, B.S.

  • Hiromitsu Kumada, M.D., Ph.D.

  • Department of Hepatology and Okinaka Memorial Institute for Medical ResearchToranomon HospitalTokyo, Japan

Ancillary