Abstract
Objective
maintenance of an appropriate mean arterial pressure (MAP) is important for critically ill patients. However, the association between MAP and prognosis in acute kidney injury (AKI) patients receiving continuous renal replacement therapy (CRRT) is thus far unclear.
Materials and methods
a total of 1,144 AKI patients who had received CRRT between January 2009 and September 2016 were enrolled and their MAP was measured at CRRT initiation. Patients were categorized into four groups (Quartile 1: MAP < 67.3 mmHg; Quartile 2: 67.3 ≤ MAP < 76.7 mmHg; Quartile 3: 76.7 ≤ MAP < 86.3 mmHg; Quartile 4: MAP ≥ 86.3 mmHg), and 28- and 90-day mortality rates were compared.
Results
our results demonstrate that 204 (72.1%), 187 (63.4%), 174 (62.6%), and 145 (50.3%) deaths occurred in quartiles 1, 2, 3, and 4 within 28 days, respectively (p < 0.001). This trend also exists in 90-day mortality (Quartile 1: 81.3%; Quartile 2: 72.5%; Quartile 3: 72.3%; Quartile 4: 61.1%, p < 0.001). The Kaplan-Meier results indicate that higher MAP is associated with a reduction in 28- and 90-day mortality (both p < 0.001). After adjusting for potential confounders using Cox proportional hazard regression analysis, higher MAP was still associated with a decline in 28 - and 90-day mortality (both p < 0.001).
Conclusion
MAP is a valuable parameter for predicting mortality in AKI patients who are receiving CRRT.
Key messages
This study explored the relationship between mean arterial pressure and the risk of mortality in patients with acute kidney injury receiving continuous renal replacement therapy.
Elevated mean arterial pressure is associated with a lower risk of mortality in acute kidney injury patients receiving continuous renal replacement therapy.
Introduction
Acute kidney injury (AKI) [Citation1–3] is a syndrome that is characterized by a sudden decline in renal function. There is a rapid loss of renal excretion function, resulting in an abnormal accumulation of nitrogen metabolic end products, such as urea and creatinine. Previous studies have shown that AKI occurs in 1.9% of hospitalized patients [Citation4] and is more common in critically ill sepsis patients, with incidence on admission [Citation5] as high as 40%. In addition, AKI occurrence rose to over 60% in an intensive care unit (ICU) hospitalizations [Citation6]. AKI leads to disturbances in hemodynamics and electrolyte metabolism, resulting in a significant increase in mortality. For example, rapidly rising potassium concentrations are known to lead to a sharp increase in the risk of cardiac arrest.
Globally, AKI management and treatment remains extremely challenging, including appropriate volume control, management of nephrotoxic medications, and the timing and type of renal support. Given that AKI is severely life-threatening and can rapidly lead to serious and irreversible adverse events, continuous renal replacement therapy (CRRT) [Citation7–9] is often required, particularly for patients with AKI who are hospitalized in the ICU.
Hypotension can potentially lead to ischemia-reperfusion injury, followed by the dysfunction of several vital body organs [Citation3,Citation10]. Furthermore, the kidneys and heart are the organs that are most sensitive to hypotension, which may contribute further to the exacerbation of systemic organ dysfunction in patients who have already developed AKI. Mean arterial pressure (MAP) is an important indicator to measure blood pressure in critically ill patients and is extensively used in the diagnosis and management of ICU patients. An association between MAP and prognosis has been observed in patients undergoing both cardiac [Citation11,Citation12] and noncardiac surgery [Citation10,Citation13]. However, whether MAP has a potential role in the prognosis of AKI patients receiving CRRT is unclear. Therefore, the purpose of this study is to investigate the relationship between MAP and the prognosis of AKI patients receiving CRRT.
Method
Study population
This study is a secondary analysis of a retrospective cohort study [Citation14]. In the Jung et al. [Citation14] study the medical records of 1,144 patients who received CRRT in the Yonsei University Health System Severance Hospital ICU and National Health Insurance Service Medical Center Ilsan hospital were obtained. The recruitment period was from January 2009 to September 2016. Inclusion criteria: 1) Patients over 18 years old; 2) Patients diagnosed with AKI according to relevant guidelines; 3) Patients who had received CRRT treatment; 4) ICU inpatients. Exclusion criteria: 1) Patients under 18 years old; 2) Pregnant or lactating patients; 3) Patients with a history of chronic kidney disease or who have received dialysis or CRRT treatment; 4) Patients with a history of post renal obstruction or previous renal transplantation. Jung et al. [Citation14]. study was approved by Yonsei University Health System, Severance Hospital, Institutional Review Board and followed the provisions of the Declaration of Helsinki (approval no. 4-2016-1073). Because this study was a secondary analysis of a previous retrospective study and the identities of the patients were anonymized, the Institutional Review Board (IRB) of Hengyang Medical School waived the need for informed consent.
Data source
The data used in this study can be downloaded for free and used in the Dryad database (https://datadryad.org/stash/dataset/doi:10.5061/dryad.6v0j9). The database (https://datadryad.org/) is a public data repository containing raw data uploaded by various authors to render their research data discoverable, freely reusable, and referable.
Collection of clinical and biochemical variables
Demographic and clinical data including age, sex, body mass index (BMI), and comorbidities (e.g., myocardial infarction, heart failure, cerebrovascular disease, and diabetes) were recorded for all patients, before the initiation of CRRT. In addition, the following biochemical laboratory data was collected at 0 h: hemoglobin, white blood cells, serum creatinine, albumin, bicarbonate, potassium, lactic acid, aspartate aminotransferase, alanine aminotransferase, and total bilirubin levels. In addition, we collected data on the age-adjusted Charlson Comorbidity Index (CCI), Sequential Organ Failure Assessment (SOFA) score, and Acute Physiology and Chronic Health Assessment II (APACHE II) score. The SOFA score, APACHE II score, MAP, C-reactive protein (CRP), and lactate were also assessed at 0 h before commencing CRRT.
CRRT protocol
When AKI arises in patients admitted to the ICU, the nephrologist decides whether to start CRRT treatment for these patients. In general, indicators for CRRT include persistent oliguria, uncontrolled volume overload, intractable hyperkalemia, and metabolic acidosis. As previously described [Citation14], patients received continuous venous to venous hemodiafiltration through the internal jugular vein, subclavian vein, or femoral vein using a Multifiltrate (Fresenius Medical care, Bad Homburg, Germany) or Prismaflex (Baxter International Inc., Lundia AB, Sweden) machine. The dialyzer surface area ranged from 1.0 to 1.4 m2 and the albumin and ß 2-microglobulin screening coefficients were 0.001 and 0.58 to 0.65, respectively. Initially, CRRT was delivered at a blood flow rate of 100 mL/min, before being gradually increased to 150 mL/min. The total outflow was the sum of the dialysis and replacement doses, providing ≥ 35 mL · kg−1 · h−1 for each participant. The average duration of the first CRRT was set to 40 h.
Study endpoints
The endpoints of this investigation were defined as mortality that occurred within 28 and 90 days after the initiation of CRRT.
Statistical analysis
Patients were classified into four groups based on the MAP interquartile range (IQR) (Quartile 1: MAP < 67.3 mmHg; Quartile 2: 67.3 ≤ MAP < 76.7 mmHg; Quartile 3: 76.7 ≤ MAP < 86.3 mmHg; Quartile 4: MAP ≥ 86.3 mmHg). The continuous variables were described as the median and IQR and differences between groups were compared using Mann-Whitney and Kruskal-Wallis tests. The categorical variables were expressed as the number and percentage and differences between groups were compared using a chi-squared test. Time-to-event curves were drawn using the Kaplan-Meier method and a log-rank test was used to determine the differences in event rates between the four groups. Univariable Cox proportional hazard regression analysis and Lasso regression analysis were used to screen out variables associated with 28-day and 90-day mortality. In multivariable Cox proportional hazard regression analysis, the 28-day and 90-day mortality rates were taken as dependent variables, the MAP as the independent variable, and the screened variables above were adjusted as covariates to investigate the independent relationship between MAP and 28-day and 90-day mortality. Furthermore, we also examined the association between MAP and 28 - and 90-day mortality by employing restricted cubic splines. The analysis encompassed evaluations with and without accounting for potential confounding factors chosen by Lasso regression analysis. In this study, RStudio (version 4.2.0) was used for statistical analysis and two-sided p < 0.05 represented a statistically significant difference between groups.
Results
Baseline characteristics of the four groups
A total of 1144 participants (61.6% male) were included in this study, with a median age and body mass index (BMI) of 66.0 (54.0, 74.0) years and 23.6 (20.9, 26.2) kg/m2, respectively. The number of participants in Quartiles 1, 2, 3, and 4 was 283, 295, 278 and 288, respectively. The baseline characteristics of the four groups are shown in . There were clear significant differences in age (p = 0.002), systolic blood pressure (SBP, p < 0.001), and diastolic blood pressure (DBP, p < 0.001) among the four groups, which were characterized by a negative correlation between MAP and age, and a positive correlation with SBP and DBP. Furthermore, significant differences in cerebrovascular disease (p = 0.031), the reason for continuous renal replacement therapy (CRRT, p = 0.011), hemoglobin (p = 0.017), APACHE II score (p < 0.001), and SOFA score (p = 0.011) were also observed among the four groups. However, differences in other variables, such as gender, BMI, and myocardial infarction, between the four groups were not found.
Table 1. Baseline characteristics of the four groups.
Comparison of 28-day and 90-day mortality among the four groups
Deaths in the four groups within 28 days were 204, 187, 174, and 145, respectively, and the corresponding 28-day mortality rates were 72.1% (204/283), 63.4% (187/295), 62.6% (174/278), and 50.3% (145/288), respectively. Notably, our results identified a significant negative correlation between MAP and 28-day mortality (p < 0.001), as shown in the left panel of . 230, 214, 201, and 176 patients died within 90 days in Quartiles 1, 2, 3, and 4, respectively, with 90-day mortality rates of 81.3% (230/283), 72.5% (214/295), 72.3% (201/278), and 61.1% (176/288), respectively. Similarly, the results also indicate that elevated MAP is associated with a decreased risk of 90-day mortality (p < 0.001), as presented in the right panel of .
Figure 1. Comparison of 28-day (left panel) and 90-day mortality (right panel) among four groups. Quartile 1: MAP < 67.3 mmHg; Quartile 2: 67.3 ≤ MAP < 76.7 mmHg; Quartile 3: 76.7 ≤ MAP < 86.3 mmHg; Quartile 4: MAP ≥ 86.3 mmHg.
Kaplan–meier analysis of MAP and 28-day and 90-day mortality
Kaplan–Meier analysis illustrated that compared with Quartile 1, the 28-day mortality in Quartiles 2, 3, and 4 showed a decreasing trend, and the difference between groups was significant (log-rank test = 36.82, p < 0.001, as displayed in the left panel of ). Similarly, an increase in MAP was associated with a lower 90-day mortality risk (log-rank test = 34.27, p < 0.001, shown in the right panel of ).
Figure 2. Kaplan–meier analysis of mean arterial pressure and 28-day (left panel) and 90-day (right panel) mortality. Quartile 1: MAP < 67.3 mmHg; Quartile 2: 67.3 ≤ MAP < 76.7 mmHg; Quartile 3: 76.7 ≤ MAP < 86.3 mmHg; Quartile 4: MAP ≥ 86.3 mmHg.
Univariable Cox proportional hazard regression analysis is associated with 28-day and 90-day mortality
describes the variables associated with 28-day and 90-day mortality in univariable Cox proportional hazard regression analysis. Our results demonstrate that BMI, SBP, DBP, heart failure, diabetes mellitus, hypertension, MV, reason for CRRT, cause of AKI, phosphate, Charlson comorbidity index (CCI), hemoglobin, creatinine, albumin, APACHE II score, SOFA score, and MAP were associated with 28-day mortality (p < 0.05 for each comparison). Other variables were not associated with 28-day mortality (p > 0.05 for each comparison). The majority of variables associated with 90-day mortality were similar to those associated with 28-day mortality, as described in .
Table 2. Univariable Cox proportional hazard regression analysis associated with 28 - and 90-day mortality.
Lasso regression analysis associated with 28-day and 90-day mortality
Lasso regression analysis was performed based on the parameters chosen by univariable Cox proportional hazard regression analysis to further screen the parameters related to 28-day and 90-day mortality. The results are presented in and . As illustrated in , Lasso regression analysis revealed that 6 parameters including hypertension, CRRT cause, CCI, AKI cause, phosphate, and MAP were associated with 28-day mortality. Similarly, Lasso regression analysis screened out 5 parameters including hypertension, CCI, AKI cause, phosphate, and MAP were associated with 90-day mortality ().
Figure 3. Lasso regression analysis associated with 28-day mortality.
Figure 4. Lasso regression analysis associated with 90-day mortality.
Independent association of MAP with 28-day and 90-day mortality
Finally, multivariable Cox proportional hazard regression analysis was used to investigate the independent association of MAP with 28-day and 90-day mortality. The results are shown in and . It was found that an increase in MAP was associated with a decrease in 28-day mortality when adjusting for potential confounders (Quartile 2 vs Quartile 1: HR = 0.72, 95% CI: 0.59 to 0.89; Quartile 3 vs Quartile 1: HR = 0.69, 95% CI: 0.56 to 0.85; Quartile 4 vs Quartile 1: HR = 0.51, 95% CI: 0.41 to 0.64). Similar to the 28-day mortality results, elevated MAP remained associated with a reduction in 90-day mortality when adjusted for potential confounders (Quartile 2 vs Quartile 1: HR = 0.73, 95% CI: 0.60 to 0.89; Quartile 3 vs Quartile 1: HR = 0.71, 95% CI: 0.58 to 0.86; Quartile 4 vs Quartile 1: HR = 0.56, 95% CI: 0.45 to 0.68). In addition, restricted cubic splines illustrate the outcomes, demonstrating a non-linear association between MAP and both 28-day and 90-day mortality with and without accounting for potential confounding factors (). Notably, no significant threshold was observed in the investigation.
Figure 5. Independent association of MAP with 28-day (left panel) and 90-day (left panel) mortality. Quartile 1: MAP < 67.3 mmHg; Quartile 2: 67.3 ≤ MAP < 76.7 mmHg; Quartile 3: 76.7 ≤ MAP < 86.3 mmHg; Quartile 4: MAP ≥ 86.3 mmHg.
Figure 6. Restricted cubic splines illustrate the outcomes, demonstrating a non-linear association between MAP and both 28-day and 90-day mortality with and without accounting for potential confounding factors. MAP: mean arterial pressure. Hypertension, continuous renal replacement therapy cause, Charlson comorbidity index, acute kidney injury cause, and phosphate were adjusted in the adjusted model of 28- mortality. Hypertension, Charlson comorbidity index, acute kidney injury cause, and phosphate were adjusted in the adjusted model of 90- mortality.
Table 3. Multivariable Cox proportional hazard regression analysis associated with 28 - and 90-day mortality.
Discussion
Based on a large cohort study, this study discovered a strong association between MAP and outcomes in AKI patients receiving CRRT. The relationship between MAP and prognosis in this population is characterized by an elevated MAP leading to lower 28-day and 90-day mortality, independent of other risk factors. The clinical implication of this conclusion is that intensive care physicians should pay attention to MAP levels in AKI patients undergoing CRRT to reduce mortality in vulnerable populations.
A considerable number of patients admitted to the ICU developed AKI and required CRRT [Citation15,Citation16]. Consequently, CRRT has become an important treatment strategy for AKI patients admitted to ICU and is widely used in clinical practice. Hypotension is a common complication in AKI patients undergoing CRRT, leading to serious adverse events and even death [Citation17,Citation18]. Therefore, the monitoring and management of blood pressure are critical for AKI patients receiving CRRT. To date, only a limited number of relevant studies have investigated the relationship between MAP and prognosis in AKI patients receiving CRRT. Based on a cohort study of 2,292 participants, Kim et al. [Citation19] found that higher MAP was associated with reduced in-hospital mortality in patients receiving CRRT. Corresponding with previous results, the findings of this study indicate that elevated MAP is associated with a significant reduction in 28-day and 90-day mortality in AKI patients receiving CRRT. Furthermore, this trend persisted after adjusting for potential confounders. Interestingly, as shown in , there was no significant difference in the mortality rates at 28 days and 90 days between the second and third quartile groups. This suggests that maintaining MAP within the range of 67 to 86 mmHg is consistently associated with similar mortality rates, which may be related to improved neuroperfusion or overall microcirculatory circulation. Further investigation is warranted to explore the specific mechanisms underlying the observed outcomes. Additionally, assessing other relevant factors such as organ perfusion, inflammatory response, and oxygen delivery could provide valuable insights into the relationship between MAP and patient prognosis in this population.
Although CRRT has been widely used in AKI patients, the management of blood pressure during CRRT is still unexplored. The ambiguity of blood management during CRRT may also partly explain why numerous studies propose that CRRT treatment or the intensity of CRRT does not significantly improve the prognosis of AKI patients [Citation20,Citation21]. In addition, to our knowledge, there are currently no relevant guidelines clarifying what level of blood pressure should be maintained in AKI patients receiving CRRT. This may potentially lead to uncertainty about patient blood pressure management and maintenance. The findings of this study provide an important basis for blood pressure management of AKI patients undergoing CRRT. The results reveal that MAP is inversely associated with mortality in AKI patients receiving CRRT and maintaining the MAP above 86.3 mmHg significantly reduces mortality within this vulnerable population. These results provide important evidence for blood pressure management in AKI patients during CRRT. Additionally, considering that the measurement of MAP is ubiquitous in intensive care units, this further enhances the application value of this conclusion. Therefore, while receiving CRRT, intensivists can use vasoactive drugs appropriately to maintain the MAP at an appropriate level and consequently reduce mortality in AKI patients.
This study has the following limitations. Firstly, this study is a retrospective cohort study and there is insufficient evidence to establish a causal relationship between MAP and 28-day and 90-day mortality. Secondly, as this is a single-center study, whether the findings can be generalized to other populations still needs to be clarified by more subsequent studies. Thirdly, as a secondary analysis of a retrospective study, it is unclear whether higher MAP potentially contributes to the development of other complications, such as stroke. Lastly, since this current study is a secondary analysis of Jung’s study [Citation14], while detailed information regarding vasopressor use was not provided in Jung’s study [Citation14]. Consequently, we do not have access to data on the utilization of vasopressors among the included patients.
Conclusion
This study found that MAP is negatively correlated with mortality in AKI patients during CRRT and maintaining a high MAP level helps reduce the risk of death in AKI patients. In addition, considering that measuring MAP is widespread in intensive care units, this inexpensive indicator warrants further promotion and use in AKI patients.
Author contributions
Sheng Yi and Hong Xiao conceptualized and designed the study; Sheng Yi and Hong Xiao collected data; Sheng Yi and Hong Xiao analyzed and interpretated the results; Sheng Yi and Hong Xiao prepared the draft manuscript; All authors reviewed and edited the manuscript.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Data availability statement
All the data can be freely downloaded from public databases (https://datadryad.org/).
Additional information
Funding
References
- Zarbock A, Kellum JA, Schmidt C, et al. Effect of early vs delayed initiation of renal replacement therapy on mortality in critically ill patients with acute kidney injury: the ELAIN randomized clinical trial. JAMA. 2016;315(20):1–9.
- Ronco C, Bellomo R, Kellum JA. Acute kidney injury. Lancet. 2019;394(10212):1949–1964. doi:10.1016/S0140–6736(19)32563–2.
- Walsh M, Devereaux PJ, Garg AX, et al. Relationship between intraoperative mean arterial pressure and clinical outcomes after noncardiac surgery: toward an empirical definition of hypotension. Anesthesiology. 2013;119(3):507–515. doi:10.1097/ALN.0b013e3182a10e26.
- Liangos O, Wald R, O’Bell JW, et al. Epidemiology and outcomes of acute renal failure in hospitalized patients: a national survey. Clin J Am Soc Nephrol. 2006;1(1):43–51. doi:10.2215/CJN.00220605.
- Bagshaw SM, George C, Bellomo R, the ANZICS Database Management Committee. Early acute kidney injury and sepsis: a multicentre evaluation. Critical Care. 2008;12(2):R47. doi:10.1186/cc6863.
- Hoste EA, Clermont G, Kersten A, et al. RIFLE criteria for acute kidney injury are associated with hospital mortality in critically ill patients: a cohort analysis. Crit Care. 2006;10(3):R73. doi:10.1186/cc4915.
- Oh HJ, Park JT, Kim J-K, et al. Red blood cell distribution width is an independent predictor of mortality in acute kidney injury patients treated with continuous renal replacement therapy. Nephrol Dial Transplant. 2012;27(2):589–594. doi:10.1093/ndt/gfr307.
- Neri M, Villa G, Garzotto F, et al. Nomenclature for renal replacement therapy in acute kidney injury: basic principles. Crit Care. 2016;20(1):1–11. doi:10.1186/s13054–016–1489–9.
- Nash DM, Przech S, Wald R, et al. Systematic review and meta-analysis of renal replacement therapy modalities for acute kidney injury in the intensive care unit. J Crit Care. 2017;41:138–144. doi:10.1016/j.jcrc.2017.05.002.
- Mascha EJ, Yang D, Weiss S, et al. Intraoperative mean arterial pressure variability and 30-day mortality in patients having noncardiac surgery. Anesthesiology. 2015;123(1):79–91. doi:10.1097/ALN.0000000000000686.
- Azau A, Markowicz P, Corbeau J, et al. Increasing mean arterial pressure during cardiac surgery does not reduce the rate of postoperative acute kidney injury. Perfusion. 2014;29(6):496–504. doi:10.1177/0267659114527331.
- Kanji HD, Schulze CJ, Hervas-Malo M, et al. Difference between pre-operative and cardiopulmonary bypass mean arterial pressure is independently associated with early cardiac surgery-associated acute kidney injury. J Cardiothorac Surg. 2010;5:71. doi:10.1186/1749–8090–5-71.
- van Lier F, Wesdorp F, Liem V, et al. Association between postoperative mean arterial blood pressure and myocardial injury after noncardiac surgery. Br J Anaesth. 2018;120(1):77–83. doi:10.1016/j.bja.2017.11.002.
- Jung S-Y, Kwon J, Park S, et al. Phosphate is a potential biomarker of disease severity and predicts adverse outcomes in acute kidney injury patients undergoing continuous renal replacement therapy. PloS One. 2018;13(2):e0191290. doi:10.1371/journal.pone.0191290.
- Tolwani A. Continuous renal-replacement therapy for acute kidney injury. N Engl J Med. 2012;367(26):2505–2514. doi:10.1056/NEJMct1206045.
- Wald R, Shariff SZ, Adhikari NK, et al. The association between renal replacement therapy modality and long-term outcomes among critically ill adults with acute kidney injury: a retrospective cohort study. Crit Care Med. 2014;42(4):868–877. doi:10.1097/CCM.0000000000000042.
- Iwagami M, Yasunaga H, Noiri E, et al. Current state of continuous renal replacement therapy for acute kidney injury in japanese intensive care units in 2011: analysis of a national administrative database. Nephrol Dial Transplant. 2015;30(6):988–995. doi:10.1093/ndt/gfv069.
- Shawwa K, Kompotiatis P, Jentzer JC, et al. Hypotension within one-hour from starting CRRT is associated with in-hospital mortality. J Crit Care. 2019;54:7–13. doi:10.1016/j.jcrc.2019.07.004.
- Kim Y, Yun D, Kwon S, et al. Target value of mean arterial pressure in patients undergoing continuous renal replacement therapy due to acute kidney injury. BMC Nephrol. 2021;22(1):20. doi:10.1186/s12882–020–02227–4.
- Fayad AI, Buamscha DG, Ciapponi A. Intensity of continuous renal replacement therapy for acute kidney injury. Cochrane Database Syst Rev. 2016;10(10):CD010613. doi:10.1002/14651858.CD010613.pub2.
- Zhang P, Yang Y, Lv R, et al. Effect of the intensity of continuous renal replacement therapy in patients with sepsis and acute kidney injury: a single-center randomized clinical trial. Nephrol Dial Transplant. 2012;27(3):967–973. doi:10.1093/ndt/gfr486.