EOL Transitions
The trajectories of functional decline in those diseases that commonly cause end organ failure such as COPD, CHF, and ESLD are similar in that they tend to be longer and more erratic with a constant state of poor and declining health that is interspersed with intermittent exacerbations requiring hospitalization. Death is often relatively sudden and unpredictable, generally arising from complications of the underlying disease. Health care delivery is often reactive rather than proactive and is often initiated in response to acute exacerbations rather than based on a proactive plan of care, which further contributes to poor quality care. Patients with ESLD often pursue curative efforts until the EOL, and palliative care and or hospice is frequently not provided or even suggested until the hope of recovery or transplantation is extinguished, which is often in the last weeks of life.
In an attempt to guide treatment decisions and more accurately predict long-term outcomes in ESLD, many classification schemes have been developed for clinical use. The 2 most common indices are the Child-Turcotte-Pugh (CTP) classification and the Model for End Stage Liver Disease (MELD). The CTP classification has been used widely for many years and was originally developed as a prognostic tool for determining operative risk for patients undergoing portosystemic shunt surgery. It is composed of 5 clinical variables: ascites, encephalopathy, serum bilirubin, serum albumin, and prothrombin time, and classifies patients as A, equating to a 90% chance of 5-year survival; B, equating to an 80% chance of 5-year survival; and C, equating to a median survival of about 1 year. There are problems with this classification system because some of these indices are subjective assessments and some are influenced by arbitrary cutoffs. In addition, CTP does not account for renal dysfunction, which has been shown to have prognostic importance in patients with ESLD. Despite these problems, it is useful clinically in that it can provide rapid risk assessment, easily calculated at the bedside, and has been found to correlate with HRQOL.
The MELD classification was developed in 2001 and was also designed to predict 90-day mortality in those undergoing portosystemic shunts. It has since been adopted by the United Network for Organ Sharing to determine priorities for allocating donor livers and has been used to determine prognosis of groups of patients with chronic liver disease. It is composed of 4 variables: serum bilirubin, creatinine, International Normalized Ratio for prothrombin time, and presence or absence of kidney dialysis. The MELD classification has improved ability to predict 90-day mortality risk but is not without limitations. Clinical markers such as ascites or varices that represent portal hypertension are excluded from the model. Patients with portal hypertension are often at higher risk of short-term death compared with those without portal hypertension and similar MELD scores. In addition, the longitudinal ability of MELD to predict survival accurately beyond 3 months is uncertain, and MELD has not been found to correlate well with HRQOL. A study in ambulatory adult patients with ESLD, looking at correlations between MELD and HRQOL, found that even despite low mean MELD scores of 12 (mortality rate of 6% at 3 months), 70% reported their liver disease symptoms moderate to severe and disabling.
Accurate estimates of risk of mortality are important in determining timing of care interventions. Current classifications do not clearly align with patients' reported functional status and sense of well-being and, alone, are not useful in determining individual risk or timing for initiation of palliative or EOL care. However, functional status and ability to manage daily activities are especially important measures in assessing and discussing desired patient-centered outcomes that extend beyond physiological measures and, along with MELD and CTP, should be incorporated into nursing clinical practice.