Renal replacement therapy after cardiac surgery; renal function recovers

Abstract

Objectives. To assess renal outcome in patients discharged from hospital following cardiac surgery-associated acute kidney injury (CSA-AKI) with need for renal replacement therapy. Design. In April 2012 we conducted a cross-sectional study of patients treated with renal replacement therapy following cardiac surgery during 2008–2010. We included all adult patients with a pre-operative serum creatinine (sCr) < 200 μM, surviving to discharge. Primary endpoint was use of renal replacement therapy after hospital discharge; secondary endpoint was a sCr > 200 μM at the time of follow-up. Results. We reviewed the records of 3828 patients receiving cardiac surgery in the defined period. A total of 107 adult patients with sCr concentrations < 200 μM were treated with post-operative renal replacement therapy of whom 70 survived to discharge. Fifty-six patients were alive at follow-up and none had required renal replacement therapy after initial discharge. Median sCr concentration at follow-up was 111 [56–257] μM and two patients had sCr above 200 μM. Conclusions. In this study, renal function recovered in patients discharged from hospital following renal replacement therapy after CSA-AKI. No patients needed further renal replacement therapy and only two (4%) had a sCr > 200 μM at follow-up.

Key words::

• acute kidney injury
• cardiac surgery
• renal replacement therapy

Introduction

Cardiac surgery-associated acute kidney injury (CSA-AKI) is a well-known complication to major cardiac surgery and is associated with increased mortality and morbidity (1). Mortality exceeding 50% has been reported in patients with CSA-AKI who require renal replacement therapy (RRT) (2). Whether CSA-AKI occurs early or late as part of a multiple organ-failure syndrome, short- and long-term mortality are significantly higher than in patients without this complication (3). The pathophysiology is not fully understood, but cardiopulmonary bypass with non-pulsatile flow and reperfusion injury causing renal ischaemia are key contributors (4,5).

Pre-operative renal dysfunction has been identified as a major predictor of CSA-AKI, chronic kidney disease and mortality (6–8). Studies on long-term outcome indicate that late mortality is dependent on post-operative renal function deterioration, and mortality is increased even when renal function is re-established at discharge in terms of serum creatinine (sCr) levels (9).

Few studies have investigated patients with pre-operative normal renal function in relation to long-term renal outcome. We performed a cross-sectional study to assess the incidence of long- term renal failure in patients with pre-operative normal to moderately elevated sCr requiring post-operative RRT.

Material and methods

In April 2012, we screened all patients undergoing cardiac surgery in 2008–2010 at a regional referral hospital (Rigshospitalet, Copenhagen University Hospital, Denmark). We included patients that received RRT in the ICU following cardiac surgery using cardiopulmonary bypass who survived to discharge. Exclusion criteria were as follows: age < 18 years, history of renal disease or dysfunction, pre-operative RRT or sCr > 200 μM (2.3 mg/dl). The sCr cut-off value was chosen according to the European System for Cardiac Operative Risk Evaluation (euroSCORE) cut-off value (10). Vital status at follow-up, including date of death if applicable, was retrieved from the Danish Civil Registration System (www.cpr.dk/cpr/). Use of RRT following discharge was retrieved for all patients by telephone interview or hospital records; sCr was measured for all patients alive at follow-up by their general practitioner, by local hospital or at our facility.

The study was approved by the Regional Ethics Committee, the National Data Protection Agency and conducted in accordance with the Helsinki declaration. Informed consent was obtained from patients who were alive at follow-up.

Post-operative RRT was defined as RRT in the Intensive care unit (ICU), and was performed by continuous veno-venous haemodiafiltration (Prismaflex, Gambro, Vallensbæk, Denmark). Settings and anti-coagulants varied with patients according to clinical status and discretion of the attending ICU consultant. General guidelines in our institution follow those described in the modified sequential organ failure assessment score (11).

Information on pre-, per- and post-operative data was retrieved from hospital databases. Co-morbidity was defined by the EuroSCORE criteria (10). Type of surgery, duration of extracorporeal perfusion and aortic cross-clamp time were retrieved from perfusion charts. In-hospital sCr values were obtained from the database at the Department of Biochemistry.

Statistics

Categorical data are presented by absolute numbers with percentages in parentheses. Continuous data are presented as means ± standard deviation (SD) for normally distributed parametric data and otherwise stated with median [minimum–maximum].

Mann–Whitney U test was used to compare creatinine levels at discharge between the group that survived and the one that did not survive until follow-up.

Rank test was used to test the significance between pre-operative and follow-up sCr values.

Statistical analysis was performed using the statistical software package SPSS, version 19.0; SPSS Inc; Chicago, IL. P values < 0.05 were considered statistically significant.

Results

A total of 3828 patients were screened. After exclusion 107 patients remained, 37 (35%) of them died in hospital giving a total of 70 included patients (Figure 1). Demographics and pre-operative risk factors are shown in Table I. Median pre-operative sCr was 95 [47–90]. Intra- and post-operative data are shown in Table II.

Figure 1. Flowchart of patient exclusion.

Table I. Demographics.

n	70
Age	72 [26–90]
Male gender	48 (69%)
BMI	27.1 ± 4.5
Euroscore, additive	9 [3–20]
Euroscore II	5 [1–70]
NYHA	3 [1–4]
Ejection fraction	50 [20–60]
NIDDM	14 (20%)
IDDM	3 (4%)
Hypertension	43 (61%)
Hypercholesterolemia	32 (46%)
COPD	14 (20%)
Vascular disease	9 (13%)

All values are given as n (percentage of total), mean ± SD or median [min–max].

BMI, Body–mass index; COPD, Chronic obstructive pulmonary disease; IDDM, Insulin-dependent diabetes; NYHA, New York heart association classification; NIDDM, Non-insulin-dependent diabetes.

Table II. Intra- and post-operative data.

Acute surgery	27 (40%)
CPB time (min)	133 [30–428]
Clamp time (min)	89 [0–282]
Continuous infusion of vasopressors	48 (69%)
Reoperation	26 (37%)
Blood transfusions (liter)	2.4 [0–22.8]
Length of stay in hospital (days)	22 [1–129]
Days in respirator	9 [0–125]
Days in dialysis	5 [1–31]
Sepsis	28 (40%)
sCr concentration at RRT initiation (μM)	231 [47–583]
Days from surgery to initiation of RRT
< 2 days	17 (24%)
2–5 days	37 (53%)
> 5 days	16 (23%)

All values are given as n (percentage of total) or median [min–max].

The study was conducted at median 2.5 years [1.2–4.1] following discharge. During this period, none of the patients had received RRT and 13 patients had died before study: four due to cardiac failure, three due to infections, two due to haematological malignancy and one due to complications to a thoracic aorta dissection. For three patients, the specific cause of death could not be retrieved.

Of the 57 patients alive at follow-up, we successfully retrieved sCr and estimated glomerular filtration rate (eGFR) values in 56; median sCr was 111 [56–257] μM and eGFR was 63 [19–90] ml/min. Median change in sCr from prior to surgery to study was 5 [− 90–130] μM (p = 0.24). Two patients had sCr levels above 200 μM (237 and 257 μM, respectively). We found no difference in sCr values at discharge when comparing patients who survived to study to those who did not (99 [43–245] and 114 [81–281], respectively, p = 0.52).

Discussion

The main finding of this study was that no patients, with normal or moderately elevated baseline sCr, were treated with RRT after discharge from hospital following CSA-AKI. Furthermore sCr levels at follow-up were similar to pre-operative levels, and only two patients had a sCr above 200 μM at follow-up. Finally sCr values at discharge of patients who survived to follow-up did not differ when compared with those who did not survive and none of the discharged patients died of renal failure; however cause of death was unknown in three patients. These findings suggest that CSA-AKI requiring RRT is an acute disorder, and that chronic renal failure is rare in patients surviving to discharge. In addition these findings support the current appraisal that CSA-AKI primarily is a marker of generalized organ dysfunction but not per se the cause of the associated increased mortality.

Our findings support those of a recent cohort study of 82 patients discharged following CSA-AKI requiring RRT. In this study, three patients required RRT in the follow-up period (mean 5.2 years, range 1–13.4 years) but all of those patients had a pre-operative sCr above 200 μM (12).

We present a single-centre cross-sectional study, with elaborate data on pre- and post-operative risk factors, and a high rate of completed follow-up (n = 56/57). The population had a relatively high mean age of 67 years and 69% were male (Table I), 40% of the patients underwent acute surgery and most had long cardiopulmonary bypass- and clamp time (Table II), which are all known risk factors for developing CSA-AKI and subsequent need of RRT.

In more than half of the patients RRT was initiated more than 48 h post-operatively (Table II). Several biomarkers have been introduced recently (13) that may facilitate detection of CSI-AKI within hours of surgery and one may speculate whether these may have expedited initiation of RRT to the benefit of the patients; however, as most of these are general markers of organ injury and inflammation, the potential beneficial effect of early RRT initiation based on the measures remains debated (14) and needs further evaluation in clinical trials.

Limitations of this study are the retrospective design, the limited number of patients, and a variable and relatively short time to follow-up.

In conclusion, RRT after CSA-AKI is known to increase mortality, but in this study, patients with pre-operative sCr below 200 μM surviving to discharge following CSA-AKI have little risk of developing chronic renal failure. This is important when informing these patients pre-operatively, for whom a life in chronic dialysis would have a major impact on quality of life.

Acknowledgements

We would like to thank Bach. Med. Frederik Eric Olof Holmberg for his help with data collection.

Declaration of interest: The authors report no declarations of interest. The authors alone are responsible for the content and writing of the paper.