Introduction Prognosis in lung cancer patients has not improved significantly in many years. Prior studies that have evaluated the prognostic value of serum albumin included only patients with late-stage lung cancer, did not control for comorbidities, and only used a single serum albumin value. We evaluated whether serum albumin obtained over time after diagnosis may be associated with lung cancer prognosis.
Methods A retrospective cohort study was performed using the New Mexico Veterans Affairs Health Center (NMVAHC) electronic charting system. Inclusion criteria included patients diagnosed with lung cancer at NMVAHC from 1995 to 2004. Data collected were age, smoking history, lung cancer histology, date of diagnosis and death, serum albumin at diagnosis and within 1 year post-diagnosis, other prognostic laboratory markers, comorbid conditions, and receipt of curative surgical treatment. Relationship of albumin to survival was performed using a Cox proportional hazards modeling.
Results 234 lung cancers diagnosed at the NMVAHC were evaluated. Non-small cell lung cancer (NSCLC) comprised 88% of all lung cancer, 34% of which were early stage (I and II) and 30% late stage (III and IV). Average age was 67 years, chronic obstructive pulmonary disease 34%, diabetes 100%, heart failure 6%, renal failure 2%, nephrotic syndrome 1%, hepatitis C 2%, and cirrhosis 2%. After controlling for comorbidities, albumin at diagnosis (p = .18) or the difference in albumin between diagnosis and 1 year post diagnosis (p = .98) were not predictive of survival. However, trending albumin over a year was predictive of survival for all stages of lung cancer (p = .0207), with an associated risk of death of 1.83 with every 1 g/dL drop in albumin 1 year after diagnosis.
Conclusion Trending serum albumin over a year of diagnosis can be used as a prognostic indicator for survival in lung cancer regardless of comorbidities, morphology, and stage of lung cancer. Availability of electronic charting allows us to trend patient data and aid in assessing tools for prognosis.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.