To the Editor:

We read with great interest the article by Roayaie et al. evaluating the 5-year outcome and the main prognostic factors after resection of single, tiny (≤2-cm) hepatocellular carcinoma (HCC) in patients with compensated chronic liver disease (CLD).[1] Analyzing 132 patients from two Western tertiary referral centers, they found that the presence of satellites and a platelet count <150 × 109/L were the independent prognostic factors for survival. However, despite the indubitable prognostic relevance of satellites, even in these patients with very early HCC, from a practical standpoint it should be noted that (1) their presence cannot be documented by imaging techniques before surgery (as clearly demonstrated in the study), (2) they were observed in only 12% of these patients, (3) they did not retain their prognostic meaning in the subset of patients with cirrhosis.[1] On the contrary, platelet count is a bed-side obtainable, inexpensive, routinely assessed parameter that resulted in being significantly associated with survival, regardless of the presence of cirrhosis.[1]

However, the identification of platelet count as a prognostic parameter in patients with well-compensated cirrhosis should not be regarded as a marker of the presence of portal hypertension (PH) alone.[2, 3] In fact, in patients with CLD, other factors, such as decreased liver production of thrombopoietin, may be responsible for decreased platelet count.[4] Thus, when patients with a similar stage of liver disease are evaluated, as in the study by Roayaie et al., thrombocytopenia may be a parameter able to pinpoint patients with subtle decrease in liver function and therefore with a worse prognosis. Indeed, in the Roayaie et al. series, the survival of patients with low platelet counts started to diverge late (>2 years) after resection, thus suggesting that low platelet count may herald late clinical problems, rather than perioperative risks. In this regard, it is interesting to note that surgical resection of HCC—despite the presence of thrombocytopenia—was not associated with an increased early perioperative mortality and that, in one of the two centers involved in the study, liver resection was allowed even in patients with small esophageal varices and platelet count below 100 × 109/L, pending the presence of an indocyanine green retention <20%.

The data by Roayaie et al. further add to the role of platelet count as a prognostic marker in patients with cirrhosis and HCC and indirectly disclose potential meanings that are not limited to the association of this hematological parameter with PH.

  • Edoardo G. Giannini, M.D., Ph.D., FACG

  • Vincenzo Savarino, M.D.

  • Dipartimento di Medicina Interna Unità di Gastroenterologia IRCCS-Azienda Ospedaliera Universitaria San Martino-ISTUniversità di Genova Genova, Italy


  1. Top of page
  2. References