Association between low serum creatinine and mortality in patients with severe COPD

Introduction: Muscle mass may be a better predictor of mortality than body mass index in chronic obstructive pulmonary disease (COPD). Serum creatinine depends on muscle mass and kidney function. Lower values ​​may predict a higher mortality rate.

OBJECTIVE: To determine whether there is an association between low serum creatinine and mortality in severe COPD.

METHODS: This is a retrospective study of serum creatinine values ​​at admission and during the last year before admission. The measured outcomes were mortality at 30 days and 1 year after admission in patients with acute respiratory failure type 2 secondary to COPD, who were admitted over a 1-year period to the respiratory ward (n=130). Statistics were calculated using chi-squared test.

RESULTS: There appears to be a significant relationship between creatinine values ​​before 1-year admission and 1-year mortality (p = 0.0003).

CONCLUSIONS: The relationship with mortality appears to be stronger with pre-admission creatinine values ​​rather than admission values ​​and appears to predict which patients are most likely to die one year after admission.

an introduction

In chronic obstructive pulmonary disease (COPD), low body mass index (BMI) is an adverse prognostic factor and is associated with increased mortality. [1-4]. A composite score of BMI, airflow obstruction, dyspnea, and exercise tolerance (BODE index) is often used to predict in patients with stable COPD and is considered superior to forced expiratory volume in one second (FEV1) alone for this the purpose [5]. For predicting during acute COPD exacerbations, a composite score of comorbidities including airflow obstruction, dyspnea, exercise tolerance and previous exacerbation severity details (CODEX) were found to be superior to both the BODE index alone or when the BODE index was used in combination with Additional information about previous exacerbations (BODEX) [6].

One study showed that mid-arm muscle area is a better prognostic indicator than BMI in patients with COPD. [7]. The relationship between BMI and lean muscle mass index (FFMI) is not very good and FFMI is shown to give additional prognostic information in patients with COPD. [4].

Serum creatinine values ​​are partly dependent on muscle mass and are usually measured as part of a routine assessment of patients admitted to hospital with various conditions, including COPD exacerbations. These values ​​are also measured very frequently in primary care, as part of chronic disease surveillance. A previous study in unselected patients admitted to intensive care showed that high and low serum creatinine values ​​are associated with an increased risk of mortality. [8].

The DECAF study (dyspnea, leukopenia, consolidation, acidity, and atrial fibrillation) included factors that predict hospital mortality in patients with exacerbation of chronic obstructive pulmonary disease (COPD) baseline serum biochemistry at the time of admission but did not include creatinine values pre-admission [9]. Because creatinine values ​​can also rise sharply, due to several other conditions such as the presence of renal insufficiency or infection, the values ​​measured at the time of admission may be less reliably correlated with an individual’s muscle mass and this may explain why in the DECAF study this was not A low creatinine value is identified as a risk factor for death. If this is the case, creatinine values ​​measured during stabilization periods, such as samples taken routinely in primary care, may have a better correlation with muscle mass in the individual patient.

We hypothesize that low serum creatinine values ​​either at the time of admission or before admission may be associated with an increased risk of death in patients with severe COPD. If this is found to be the case, it offers the advantage of the ready availability of a prognostic indicator, as this test is frequently performed in both the primary care and the secondary care setting. If so, it may help to immediately identify patients at risk of adverse outcomes.

Patients receiving non-operative ventilation (NIV) for acute respiratory failure type 2 (usually secondary to COPD) are at significantly increased risk of death with an inpatient mortality rate of 34% in the British Thoracic Society’s national review on NIV, 2013. We wanted To see if there would be a significant difference in mortality in this group of patients based on accepted or pre-admission creatinine levels.

materials and methods

The study was approved by the Health Research Authority, Healthcare Research Wales (IRAS Project ID: 254977), and the hospital’s Department of Research and Development. We have conducted a retrospective cohort study in patients with severe COPD, who were admitted with acute type 2 respiratory failure and treated with NIV. All patients who received NIV for acute respiratory failure type 2 secondary to COPD in the respiratory ward of Royal Blackburn Teaching Hospital (Lancashire, UK) in 2013 were included in the study. The diagnosis of COPD was confirmed by review of discharge summaries, clinic letters, or patient notes, and any patients in whom they used NIV for indications other than COPD (eg obesity-related hypopnea or neuromuscular disease resulting in respiratory failure) were excluded. type 2). The study was conducted in 2019 and the reason for selecting the cohort from 2013 was to ensure that the complete data for assessing mortality was obtained for one year after hospital discharge.

A total of 130 patients fulfilled the criteria for inclusion in the study. The normal reference ranges set by the hospital laboratory for creatinine were 46-92 μmol/L for women and 58-110 μmol/L for men. For the study, a patient was considered to have low serum creatinine values ​​if his creatinine values ​​were 45 or less for females and 57 or less for males.

Mortality rates at 30 days and 1 year after hospital discharge were compared between patients with low creatinine values ​​during the year prior to discharge and patients with normal or high creatinine values. The first sample taken at the time of admission for blood biochemistry was taken as the input creatinine value and any creatinine value that was abnormally low between the previous year and the time of admission was taken as evidence of a pre-admission creatinine low value for the purpose of our calculations.


We performed a chi-square test to see if there were significant differences in mortality rates observed in patients with low admission or baseline creatinine values, compared to patients with normal values, at 30 days and 1 year after discharge.

The patients included in the study were 35% male (n = 46) with a mean age of 68.9 years (median = 69 years). The mean blood gas pH at the time of NIV initiation was 7.29 (range = 7.19–7.34). The results are summarized in the table 1.

Died within 30 days Alive in 30 days q value Died within one year Alive at 1 year q value
Acceptance Low Creatinine (N=31) 7 24 0.107 17 14 0.068
Normal or elevated creatinine (N = 99) 11 88 36 63
Creatinine decrease within 1 year prior to admission (N=48) 6 42 0.734 30 18 0.0003
Normal or elevated creatinine level in the year prior to admission (N=82) 12 70 25 57

Based on admission serum biochemistry, 31 (23.85%) of the total 130 patients in the study had low serum creatinine values. The mortality rate in this group was 22.58% (n = 7) at 30 days and 54.84% (n = 17) at 1 year (Fig. 1). Differences did not reach statistical significance (p-values ​​0.107 and 0.068 for 30 days and 1 year, respectively) when compared with patients with normal or high creatinine values ​​based on admission values ​​where the 30-day mortality was 11.11% (N = 11) or 36.36% at one year (n = 36).

When patients were classified based on the presence of low creatinine values ​​in blood tests performed within one year prior to admission, the mortality rate in patients with low creatinine (N = 6) at 30 days was 12.5%, compared to 14.63% in the others. . (p = 0.734). However, one-year mortality was significantly higher (p = 0.0003) in patients with low creatinine before admission as 62.5% of patients died (N = 30) compared with 30.49% mortality (N = 25) in patients with They have normal or high creatinine values ​​(Fig 2).


Low BMI is known to be a negative prognostic indicator in patients with COPD, but muscle mass may be better than BMI in this regard. [7] Or to show additional information from BMI [4]. Measuring lean muscle mass or mid-arm muscle area is complex. In studies, this was calculated using equations incorporating various other factors such as mid-arm circumference, BMI, skin thickness, and fat mass estimate. [7]. This complexity makes the use of endpoints such as these problematic in everyday use in the care of patients with COPD. In addition, although there is a proposed method for classifying cuts for low lean muscle index [10]There are no official guidelines on this topic.

Serum creatinine values ​​are routinely checked during hospital admission and frequently measured in primary care. Although serum creatinine values ​​can be affected by a variety of factors such as intrinsic kidney disease, use of nephrotoxic drugs, and the presence of systemic disease, the values ​​are also dependent on baseline muscle mass. [11]. Since some of these factors are likely to be present during the hospitalization period, it is expected that measurement of serum creatinine values ​​during stable conditions, perhaps in primary care, will be more correlated with actual muscle mass than those measured during hospitalization periods, when elevated The latter level erroneously due to factors such as acute kidney injury or use of nephrotoxic drugs. This may explain why in the DECAF study a decrease in serum creatinine was not seen as a significant factor affecting mortality, since only values ​​on admission were considered.

We found in our study that although there were no significant differences in mortality between patients with low creatinine values ​​and others at 30 days after admission, there appeared to be a trend towards significance at 1 year when patients were identified based on admission values. , and statistically significant differences when identifying patients based on pre-admission creatinine values. This suggests that differences in mortality may be evident in the medium to long term rather than in the immediate post-admission period. Previous studies have shown that increasing BMI through nutritional support improves survival [3] and quality of life in COPD patients [12]. If further studies (ideally prospectively) confirm the value of low serum creatinine as a useful prognostic marker in identifying patients at increased risk of death, it may be possible to improve outcomes by providing nutritional support to these patients, especially with increased mortality not in the post-discharge period. directly.


The identification of more patients with low serum creatinine based on pre-admission values ​​rather than admission values ​​appears to indicate that some of these patients have already developed increased creatinine values, which may reflect a severe deterioration in their general health. This may explain the lack of significant differences in mortality based on admission values ​​due in part to the deaths of these patients in the low creatinine comparison group.

It is recommended that the results of our study be confirmed by a prospective study where other important factors such as the patient’s BMI, nutritional status, and severity of airflow obstruction can also be considered to assess differences in mortality.