This retrospective study collected data from 372 consecutive adult patients who underwent craniotomy for brain tumors at the Department of Neurosurgery in the Southwest Hospital of the Third Military Medical University (Army Medical University) from January 2017 to December 2019. All study procedures and protocols involving human participants were in accordance with the ethical standards of the 1964 Helsinki Declaration, and the the Ethics Committee of the Southwest Hospital of Third Military Medical University has approved this study. For retrospective study, formal consent is not required and the requirement for the informed consent of patients was waived prior to the collection of their medical data in this study.
Inclusion and exclusion criteria
In this study, 372 adult patients with normal preoperative total (≥ 60 g/L) and albumin (≥ 35 g/L), who underwent craniotomy due to intracranial solid tumors, such as gliomas, meningiomas, acoustic neuromas, and metastatic tumors, were included. Patients who underwent craniotomy due to non-solid tumors, such as aneurysm, vascular malformation, intracerebral hemorrhage, hydrocephalus, cranioplasty or skull lesion, had been excluded. Patients younger than 18 years old, who had hypoalbuminemia (< 35 g/L) before operation were also excluded. Patients with abnormal proteinuria that may lead to protein loss were also excluded. For the purpose of this study, patients with postoperative serum albumin of < 35 g/L were defined as hypoalbuminemia group, while those with postoperative serum albumin of ≥ 35 g/L were defined as non-hypoalbuminemia group. The postoperative albumin level was checked on the first blood samples taken upon after craniotomy.
Clinical data collection
For each patients, demographic data [age, sex, body mass index (BMI), histories of smoking and drinking], comorbidities (hypertension, diabetes mellitus, coronary heart disease, viral hepatitis, and history of stroke etc.), as well as pre- and post-operaotive laboratory test were collected. Operation time, intraoperative blood loss, intraoperative total input and intraoperative total urine output composed the operative-related medical data. Pathological classification were documented according to Central Nervous System Tumor Classification by the World Health Organization in 2016. The durations of postoperative neurological intensive care unit (NICU) and hospitalization and postoperative complications (pneumonia, epilepsy, incision infection, respiratory failure, renal failure, hydrocephalus, deep vein thrombosis) were also collected.
The pre- and first post-operative (within 6 hours after surgery) laboratory tests included white blood cell count, red blood cell count, hemoglobin, hematocrit, platelet count, fibrinogen (FIB), prothrombin time (PT), prothrombin time-international normalized ratio (PT-INR), activated partial thromboplastin time (APTT), thrombin time (TT), D-dimers, potassium (K), sodium (Na), chloride (Cl), blood glucose, calcium (Ca), phosphorus (P), magnesium (Mg), serum prealbumin, total protein, albumin and albumin/globulin (A/G) ratio.
All statistical analysis was performed using the SPSS software for Windows (version 25.0, IBM SPSS Inc., Chicago, IL) software. Descriptive statistical analysis of the data (e.g., means, medians, frequencies, and percentages) was performed. Normally distributed quantitative variables are presented as the mean ± standard deviation (SD) and were compared using independent unpaired two-tailed Student’s t test. Non-normally distributed quantitative variables are presented as the median [interquartile range (IQR)] and were compared using independent-sample nonparametric test. Categorical variables were expressed as counts with percentages and compared using the chi-squared test or continuity correction test. Logistics reression analysis was used to investigate the risk factors of postoperative hypoalbuminemia and pneumonia. The variables were analyzed by univariate binary logistic regression analysis. Multi-colinearity was assessed using the Pearson correlation coefficient statistic and by checking the Variance Inflation Factor multiple regression model with the same dependent and independent variables. Multivariate logistic regression model was performed on variables with significant differences (P < 0.05) determined in the univariate analysis to investigate the relationship between variables and postoperative hypoalbuminemia/pneumonia. All statistical tests of hypothesis performed at the 0.05 level of significance.