In the last 35 years tumour markers (TM) have gained currency in clinical practice. However, in the light of indications by international guidelines, their use is often unjustified. Our aim was to quantify the use of some of the most common TM, assessing their appropriateness and their efficacy in an Internal Medicine Unit. METHODS: In the three Internal Medicine Units of the Department of Internal Medicine of Policlinico of Modena we have carried out a retrospective analysis of the assessment of the main TM (CEA, CA19.9, CA 125, CA 15.3, NSE). The analysis was divided into two distinct phases: (I) quantitative phase, in order to assess the scale of the problem in economical terms; (II) qualitative phase, in order to assess the efficacy of the tests and the appropriateness of their use. RESULTS: (I) At last one of the considered TM was requested in 5102 out of the 8253 admitted patients (62%) (period 2001-2003). The trend was similar in all three units examined. (II) The qualitative analyses revealed: (1) the most common motivation for their use (79%) was diagnostic, mostly prior to any other test; (2) a mere 5% of the requests were appropriate according to the international literature; and (3) TM showed a low positive predictive value when used for diagnosis in an unselected population such as that of an Internal Medicine unit. CONCLUSIONS: The results of our study showed that TM determination represents an overall cost for Internal Medicine units and that there is a high inappropriateness in their use compared to what it is suggested by international guidelines. Though the TM is a low-cost test when used correctly, it seems an unnecessary expense if not adequately incorporated into the decision making process

Controlled trials demonstrated efficacy and safety of non-invasive ventilation (NIV) in treatment of acute respiratory failure, initially in Intensive Care Units, then in other care settings (semi-intensive care units, emergency departments, and also in the wards, more often pneumological ones). Few studies have been published about NIV in Italian wards of Internal Medicine with full self-management of NIV by internists in a normal ward setting. We performed a prospective real-life study about the use of NIV in Internal Medicine ward devoid of a critical area of semi-intensive therapy, with the aim of confirming, in this setting, the effectiveness of NIV. During a period of 13 months, 42 patients with hypercapnic respiratory failure of different etiology and acidosis (pH<7.35) were treated by NIV. NIV was successful in 81% of patients. In-hospital mortality was 9.5%. Safety of NIV is demonstrated by the absence of serious complications: only 7 patients showed poor compliance and 2 patients had facial pressure ulcer due to the mask. There were not statistical differences in success rate of NIV according to severity of acidosis at admission (pH<7.25 vs pH>7.25), neither according to the acute physiology and chronic health evaluation II score and the national early warning score, but the modified early warning score only showed statistically significant difference with lower values in the success group: 2.82±1.57 vs 4.13±1.46 (P<0.05). NIV has proven to be effective and safe in Internal Medicine ward.

Hypercalcemia is ideally detected by the measurement of serum ionised calcium. Because this is not widely available, in common clinical practice "albumin-corrected" calcium values are often utilized. Our study investigated whether the method used to measure serum albumin concentration may significantly interfere in the derived serum calcium values and, consequently, in the identification of hypercalcemic patients. In 170 consecutive patients admitted to our Department of Internal Medicine we measured serum total calcium, total protein, and albumin by colorimetric method; albumin concentration was also derived by electrophoresis assessment. After correcting serum calcium for colorimetrically (CA) and electrophoretically (EA) measured albumin values, the detected frequencies of hypercalcemia were compared, utilizing different cut-off limits (i.e. 11.0, 10.4 and 10.2 mg/dl). In our patients, the CA values were significantly lower than EA levels. As a consequence, EA-corrected calcium, as well as total calcium concentration were significantly lower than CA-corrected values. This may also account for the very different prevalence of hypercalcemic patients identified by serum total, EA-corrected and CA-corrected calcium values. Our data therefore indicate the importance of the method of albumin measurement in the determination of "corrected" calcium concentration.

No abstract