Prognosis of hypercalcemia in aerodigestive tract cancers: study of 136 recent cases.

Recent data issuing the prognostic impact of hypercalcemia on outcome of aerodigestive tract cancers are spare. To assess the prognosis and the survival of head and neck cancer patients with hypercalcemia, we reviewed 136 recent successive cases, including also oesophageal and lung cancers. Data were collected from a retrospective database (July 2002-January 2004). Hypercalcemia was defined by calcemia level above 2.55 mmol/l. Univariate analysis for prognosis was performed with Mann-Whitney test (continuous variables) and Odd Ratio with 95% confidence interval (categorical variables). The primary locations were : oropharynx and oral cavity (79, 58%), hypopharynx (13, 9.5%), larynx (10, 7.3%), oesophagus (17, 12.5%) and lung (17, 12.5%). There were 23 females and 123 males, with a median age of 53 (18-86). The incidence of bone metastasis was low: 20/136, 14.5%. At cancer diagnosis, 32 hypercalcemia were observed. With a median follow-up of 88 days (2-553), we observed 98 deaths (overall mortality=72%). The median overall survival was 35 days (2-553+). The pejorative prognostic factors were: male gender (OR=2.64 CI 95% 1.07-6.82), age inferior to 50 (OR=2.67 CI 95% 1.23-5.8), presence of distant metastasis (OR=4.45 CI 95% 1.8-11.01), elevation of alkaline phosphatases (OR=7 CI 95% 2.73-17.9) and need of hospitalization for intravenous hydratation (OR=5.11 CI 95% 1.99-13.17). We observed 39 recurrences of hypercalcemia. The predictive factors for recurrence of hypercalcemia were: age superior to 50 (OR=4.61 CI 95% 2.02-10.52), male gender (OR=38.22 CI 95% 12.2-89), calcemia level superior to 2.7 mmol/l (OR=3.08 CI 95% 1.42-6.64) and absence of diphosphonates (bisphosphonates: OR=2.16 CI 95% 1.01-4.63). Despite use of diphosphonates (infusions of pamidronate), hypercalcemia is associated with very poor prognosis. Tumour location and level of calcemia had no prognostic value.

Causes of fever and value of C-reactive protein and procalcitonin in differentiating infections from paraneoplastic fever.

OBJECTIVE: The aims of our retrospective study were to study first the causes of 245 febrile episodes in cancer patients and then the value of procalcitonin (PCT) and C-reactive protein (CRP) in differentiating infections from paraneoplastic fever. METHOD: The causes of fever were studied in 245 consecutive cases observed between January and December 2002, and PCT and CRP diagnostic value in 155 cases (114 infections and 41 paraneoplastic fever). RESULTS: The two main causes of fever were infection (121 cases) and paraneoplastic fever (43 cases); 77 infections were microbiologically documented. Cocci gram positive caused 41 out of 77 documented infections. Paraneoplastic fever was more frequent in metastatic disease ( p=6.10(-6)). CRP and PCT serum levels at admission did not differ significantly in the infection group and paraneoplastic fever group (respectively p=0.39 and p=0.14 with Mann-Whitney test). The PCT and CRP levels had poor prognostic value in infection (respectively 0.04 and 0.0003 with Mann-Whitney test). CONCLUSION: Causes of fever are very numerous in cancer patients (more than 75 different causes in this study). Bacterial infection is the most frequent cause. PCT and CRP failed to discriminate infection, but both had pejorative prognostic value in infected patients.