Influence of postmenopausal state, solar radiation, drugs and comorbidities on serum calcium and phosphate in 13,000 hospital admissions.

AIMS: Our goal was to determine the relative contributions of demographic variables, drugs, comorbidities, and weather conditions on serum calcium (Ca) and phosphate (Pi) in patients admitted to a tertiary referral center. METHODS: For 12,667 consecutive patients admitted to the Kantonsspital St. Gallen, drug history on admission, age, sex, body weight, ICD-10 diagnoses, and laboratory data were extracted from electronic medical records. Weather parameters prior to hospital admission were also integrated in a regression analysis. RESULTS: Serum Ca was normally distributed with a median (interquartile range) of 2.3 (2.2/2.4) mmol/L. In contrast Pi showed a right tailed distribution of 1.0 (0.9/1.2) mmol/L. Ca was increased in postmenopausal women. Solar radiation prior to admission was associated both with higher Ca and higher Pi. Lower blood pressure was associated with lower Ca and higher Pi. In addition Ca increased by 0.017 mmol/L per g/L increase of albumin (p < 0.0001). CONCLUSIONS: Serum Ca and Pi at hospital admission are highly dependent on patient characteristics, drugs, and comorbidities. In particular, we found higher Ca in postmenopausal women. The commonly applied albumin correction formula of Payne (0.025 mmol/L Ca per g/L albumin) may overestimate the effect of albumin; we propose using 0.017 mmol/L Ca per g/L albumin or measurement of free (ionized) Ca.

Influence of weather conditions, drugs and comorbidities on serum Na and Cl in 13000 hospital admissions: evidence for a subpopulation susceptible for SIADH.

OBJECTIVES: Considerable variation in serum sodium (Na) and chloride (Cl) is found in patients at hospital admission. Our goal was to quantify the respective impact of drugs, comorbidities, demographic factors and weather conditions on serum Na and Cl. DESIGN AND METHODS: For 13277 consecutive patients without terminal kidney disease admitted to the Department of Internal Medicine of the Kantonsspital St. Gallen drug history on admission, age, sex, body weight, ICD-10 diagnoses, and laboratory data were extracted from electronic medical records. Weather parameters prior to hospital admission were also integrated in a multivariate regression analysis. RESULTS: Both serum Na and Cl showed an asymmetric left-tailed distribution. Median (interquartile range) Na was 138 (136/140) and Cl 104 (101/106). The distribution of sodium in patients with one or more risk factors for SIADH was best explained by the presence of two populations: one population with a similar distribution as the unexposed patients and a smaller population (about 25%) shifted to lower sodium levels. Lower weight, lower blood pressure, kidney dysfunction, fever, and diabetes were associated with both lower Na and Cl. Higher ambient temperature and higher air humidity preceding admission were associated with both higher Na and Cl values. CONCLUSIONS: Na and Cl at hospital admission are highly influenced by ambient weather conditions, comorbidities and medication. The bimodal distribution of Na and Cl in persons exposed to risk factors for SIADH suggests that SIADH may only affect a genetically distinct vulnerable subpopulation.