In the present study, multivariate linear regression analysis showed that serum albumia level was closely related to LOS of ICUs. Trend test and smooth curve fittings suggested that LOS of ICUs was gradually shorten when serum albumia level was gradually increased. The above relationships was also found in subgroup of age, sex and BMI.
The nutritional status of patients with HF in ICUs significantly affects their prognosis and LOS of ICUs. Albumia not only maintains the colloidal osmotic pressure, but also plays an important role in their nutritional status. Hypoalbuminemia, occurred commonly in patients with HF in ICUs, might be related to their poorer heart function and more underlying diseases, and it could result in serious adverse events. Previous studies showed that hypoalbuminemia significantly affected prognosis and LOS of various diseases, such as patients for acute decompensated HF with preserved ejection fraction, acute decompensated chronic obstructive pulmonary disease and requiring surgery.[][][][] In the present study, linear relationships between serum albumia level and LOS of ICUs of patients with HF was found, and elevation of serum albumia level could reduce the LOS of ICUs.
Currently, several studies had suggested that serum albumia level < 3.5g/dL was an independent risk factor for prolonged LOS of ICUs,[13][14][] but the present study found that LOS of ICUs was gradually shorten when serum albumia level was gradually increased, and the above relationship was also showed in subgroup based on age, sex and BMI. In the past studies, age not only was an independent risk factor for mortality of various diseases, but also determined the length of their hospital stay, and with age increasing, the length of hospital stay was prolonged.[][][] The elderly often had multiple organ failure, which lead to poor prognosis and long length of hospital stay. Especially some patients with HF in ICUs, not only having poor heart function, but also presenting other serious underlying diseases. Similar to age, sex difference had also different influence on patients with HF in ICUs. Systolic blood pressure could increase the heart load of patients who had severe diseases, especially patients with HF, which could have an adverse influence on their prognosis and prolong the length of hospital stay. Compared with men, female had higher systolic blood pressure, and as patients grew older, female could have heavier arteriosclerosis.[][] Severe coronary arteriosclerosis reduced blood supply of heart, and made prognosis of patients with HF worse. Unlike gender and age, high BMI might be protective factor for patients in ICUs. Although the above opinion has been controversial, most current studies supported this opinion.[][] Mukhopadhyay, et al. analyzed 273 patients in ICUs that high BMI was associated with low mortality and long LOS of ICUs.[] It could be seen from the above that age, gender and BMI could affect the length of hospital stay to a certain degree. In order to further analyze whether the above three variables had an impact on the relationship between serum albumin level and LOS of ICUs, we conducted subgroup analysis based on the above three variables in the present study, and found that associations of serum albumia level for LOS of ICUs were all negative correlation in subgroups.
Patients with HF in ICUs might had poor prognosis and long LOS of ICUs because of poor nutritional status and high inflammatory status. Indeed, albumia were association with malnutrition and inflammatory factors, this might be an explanation for association between LOS of ICUs and serum albumia level. The present study found that high serum albumia level was protective factor for patients with HF in ICUs, and the result could guide clinicians to reduce LOS of ICUs by changing albumin levels. However, our study also have some limitations. Firstly, the study was a single-center retrospective design, the numbers of patients included are not large, and the selected population is relatively limited, therefore, it might has selection bias, and a multicenter study needs to be conducted to confirm these findings. Secondly, we excluded populations who were younger than 18 years, because big differences were seen in serum albumin levels between adults and minors. Finally, we might not adjust other potential confounding factors.