Carboxyhemoglobin and its correlation to disease severity in cirrhotics

GOAL: To assess the correlation of serum carboxyhemoglobin (CO-Hb) to severity of liver disease as compared with Model for End Stage Liver Disease (MELD) score, Child Pugh score, and clinical parameters. BACKGROUND: There are 2 sources of carbon monoxide (CO) in humans, exogenous sources include those such as tobacco smoke and inhaled motor vehicle exhaust. The endogenous source is via the heme-oxygenase pathway, in which a heme molecule is broken down into biliverdin with release of an iron (Fe) and CO molecule. Normal serum CO-Hb levels in nonsmokers is 0% to 1.5% and 4% to 9% in smokers. Activity of the heme-oxygenase pathway may be increased in the cirrhotic patient, as measured indirectly by exhaled CO and serum CO-Hb. This may be due to alterations in vascular tone in the splanchnic circulation in cirrhotics that may lead to elevated CO production. One published study also showed that those with spontaneous bacterial peritonitis had higher levels of both CO and CO-Hb. The MELD score uses prothrombin time (INR), creatinine, and bilirubin in the prediction of short-term mortality in decompensated cirrhotics while awaiting liver transplant. Measurement of endogenous CO-Hb may correlate to severity of liver disease. STUDY: Retrospective analysis was done of 113 adult patients who were evaluated for liver transplantation between September 1996 and July 2003 and had pulmonary function testing with CO-Hb as part of their evaluation. We excluded any patients with a history of smoking. Clinical parameters used for comparison included grade of esophageal varices (n=75), spleen size (n=51) measured on abdominal ultrasound or computed tomography scan, aminotransferases, and disease duration. Serum CO-Hb levels were measured from whole blood, sent refrigerated to ARUP laboratories (Salt Lake City, UT) and analyzed via spectrophotometry. Bivariate analysis was performed by means of the Pearson product moment correlation. RESULTS: The mean CO-Hb level was 2.1%, which is higher than the expected normal population controls. No correlation was found, however, with MELD score, Child Turcotte Pugh score, or other biochemical or clinical measurements of disease severity. CONCLUSIONS: Although CO and CO-Hb production may be increased in the cirrhotic patient, in this study no correlation was found to disease severity as measured by the MELD score. Further studies are needed to assess the role of CO in other complications of cirrhosis including infection and circulatory dysfunction.