Creatinine production rate is an integrative indicator to monitor muscle status in critically ill patients

Abstract

Background: Both quantitative and qualitative aspects of muscle status significantly impact clinical outcomes in critically ill patients. Comprehensive monitoring of baseline muscle status and its changes is crucial for risk stratification and management optimization. However, repeatable and accessible indicators are lacking. We hypothesized that creatinine production rate (CPR) could serve as an integrative indicator of skeletal muscle status.

Methods: We conducted a series of animal and clinical studies. First, animal experiments were performed to determine whether CPR reflects not only muscle volume, but also qualitative muscle properties. We also evaluated the effects of acute systemic inflammation, a common feature of critical illness, on CPR, as well as its impact on muscle volume and metabolism. In clinical studies, we analyzed CPR, calculated based on urinary creatinine excretion and changes in serum creatinine, of critically ill patients. We assessed the factors affecting CPR on ICU admission and its temporal changes. Finally, we evaluated the clinical utility of CPR by examining the associations of the CPR index (CPR divided by height squared) on ICU admission and its changes with one-year survival.

Results: Animal studies revealed that CPR is determined by muscle volume, creatine content, and metabolic status. Systemic inflammation accompanied by muscle loss led to reduced CPR. Moreover, even without muscle loss, systemic inflammation decreased CPR, likely due to metabolic derangements. In ICU patients, CPR on admission strongly correlated with muscle cross-sectional area (CSA), with age and sex as additional significant factors. In contrast, the percent change in CPR showed a weak correlation with muscle CSA changes. Additionally, the acute-phase CPR trajectories did not show a consistent decline, suggesting multifactorial influences. In a cohort of 629 ICU patients, lower baseline CPR index (hazard ratio [HR] 1.125 per 0.1 g/day/m² less, P < .001) and a decrease in CPR over the first three days (HR 1.028 per 5%, P = 0.032) were independently associated with higher one-year mortality.

Conclusions: CPR represents an integrative indicator of skeletal muscle status in critically ill patients, reflecting both quantitative and qualitative aspects. Monitoring CPR in the ICU may facilitate risk stratification and optimization of patient care.

Keywords: Creatinine; Critical care; Mitochondrial dysfunction; Sarcopenia.

Fig. 1

Overview of the Study Schemes A, B: Schemes of the animal study. A Analysis of the correlation between CPR and psoas muscle cross-sectional area (CSA) as evaluated by CT. B Analysis of the effects of creatine content and mitochondrial inhibition on CPR. C–F Schemes of the clinical study. C Analysis of factors associated with CPR at ICU admission. D Analysis of factors associated with temporal changes in CPR during the ICU stay. E Descriptive analysis of the CPR trajectory. F Analysis of the association of the CPR index on ICU admission and its acute temporal changes with one-year survival. BNx, bilateral nephrectomy; CPR, creatinine production rate; CSA, cross-sectional area; CT, computed tomography; ICU, intensive care unit

Fig. 2

Analysis of CPR determinants in mice. A Experimental scheme for CPR quantification with bilateral nephrectomy in mice. B Correlation between CPR and psoas muscle CSA measured using CT. C Creatine concentration in muscle tissues and D CPR in mice supplemented with creatine in the drinking water. E The effects of mitochondrial inhibition with rotenone administration on CPR. F Illustrated summary of CPR determinants in mice. BNx, bilateral nephrectomy; CPR, creatinine production rate; CSA, cross-sectional area; CT, computed tomography

Fig. 3

Effects of acute systemic inflammation on psoas CSA and CPR. A Psoas muscle CSA measured using CT and B CPR in mice with systemic inflammation induced by intraperitoneal LPS administration. LPS, lippopolyssacharide; CPR, creatinine production rate; CSA, cross-sectional area; CT, computed tomography

Fig. 4

Metabolic analysis of muscle tissue in mice with LPS-induced systemic inflammation. A–E Analysis of mitochondrial complex activity in the muscle tissues. Time-course changes in OCR in (A) analysis of complexes 1 and 4 and (B) analysis of complexes 2 and 4. OCR of (C) complex 1, (D) complex 2, and (E) complex 4. F–K: Metabolomic analysis of muscle tissues. F Normalized creatine levels in the muscle tissue. G Significantly enriched chemical classes of metabolites in mice with LPS-induced systemic inflammation. Normalized levels of kynurenine (H), 3-hydroxykyunurenine (I), leucine (J), and isoleucine (K) in muscle tissue. LPS, lippopolyssacharide; OCR, oxygen consumption rate; FDR, false discovery rate

Fig. 5

Flow diagram of the cohorts of the clinical study. CPR, creatinine production rate; CSA, cross-sectional area; CT, computed tomography; RRT, renal replacement therapy; ECMO, extracorporeal membrane oxygenation; AKI, acute kidney injury; ICU, intensive care unit

Fig. 6

Correlation between CPR on ICU admission and muscle CSA on CT. A Correlation of CPR with psoas muscle CSA (r = 0.836; 95% CI, 0.737–0.900; P < 0.001); and B Correlation of CPR with abdominal muscle CSA (r = 0.770; 95% CI, 0.602–0.881; P < 0.001). All individual data are shown as dots, and a single linear regression with a 95% CI is shown as a blue line. CPR, creatinine production rate; CSA, cross-sectional area; CT, computed tomography; CI, confidence interval

Fig. 7

Correlation between percent change of CPR after ICU admission and those of muscle CSA. A Correlation of percent changes of CPR with those of psoas muscle CSA (r = 0.505; 95% CI, 0.117–0.806; P = 0.046). B Correlation of percent changes of CPR with those of abdominal muscle CSA (r = 0.357; 95% CI, − 0.242–0.712; P = 0.174). All individual data are shown as dots, and a single linear regression with a 95% CI is shown as a blue line. CPR, creatinine production rate; CSA, cross-sectional area; CI, confidence interval

Fig. 8

Trajectories of CPR during the ICU stay. A CPR on days 1–5 of the ICU stay is shown as a violin plot. The horizontal bars in each violin plot indicate median and quartile ranges, respectively. The horizontal dashed red line indicates the median value on day one. B Percent changes of CPR compared with the first day on days 2 through 5 are described in five categories according to its value. The ribbons between the bars indicated the transition in patients between categories. CI, confidence interval; CPR, creatinine production rate; ICU, intensive care unit