Assessing 2-month clinical prognosis in hospitalized patients with advanced solid tumors.

PURPOSE: The aim of this study was to assess clinical, laboratory, and subjective (patient's preferences) prognostic factors in hospitalized patients with advanced solid tumors. PATIENTS AND METHODS: This prospective study surveyed 177 patients from two French hospitals who had not reached the stage of active dying but had an estimated survival of less than 6 months (median survival, 58 days).
RESULTS: Univariate analysis showed that 10 of the 13 clinical and laboratory factors reported in the literature affected survival at 2 months. Poor prognostic factors were number of metastatic sites, cerebral metastasis, low Karnofsky index, dyspnea at rest, anorexia, edema, confusion, low serum albumin, extremely high leukocyte counts, and high lactate dehydrogenase (LDH) levels. The patient's desire to continue curative treatment was also associated with survival. The multivariate analysis selected four independent criteria: Karnofsky index (in three classes: <or= 30%, 40% to 60%, or >or= 70%), number of metastatic sites (>or= two or < two), low serum albumin (in three classes: <or= 24, 24 to 33, and >or= 33 g/L), and LDH concentration (>or= 600 IU or < 600 IU). The combination of these four criteria assessed prognosis better than the Karnofsky index alone, producing three prognostic profiles: one with short survival (< 2 months: no patient survived to 4 months); one with an expectation of intermediate survival (25% were alive at 4 months), and a final group surviving for several months (80% were alive at 4 months).
CONCLUSION: The prognostic profiles defined by combinations of these four factors may be potentially useful but need further validation before their application in the daily practice.