Acute kidney injury (AKI) is a common complication both before and after liver transplantation (LT) with a reported incidence ranging from 10% to 20% in the pretransplant setting and up to 64% after transplantation.1-7 AKI in this setting influences postoperative morbidity and mortality.7-10 Many earlier studies had demonstrated a direct relationship between preoperative AKI and patient survival after LT.4,11-13 These early findings coupled with the adaptation of the Model for End-stage Liver Disease (MELD) system as an organ allocation tool, which prioritizes patients with renal dysfunction for liver transplants, raised concerns about a potential trend for lower overall survival after LT. However, the adaptation of MELD score did not lead to worsened liver transplant outcomes.11 This is likely explained by the fact that AKI episodes encompass a range of renal insults that differ in terms of etiology, ongoing kidney damage and prognosis. Pretransplant AKI episodes due to hepatorenal syndrome normally resolve post-LT without any untoward effects on patient outcomes while renal insults causing acute tubular necrosis may not resolve fully, leading to chronic kidney disease (CKD), end-stage renal disease, and ultimately decreased patient survival.14,15
Similarly, postoperative renal dysfunction is an independent predictor of morbidity and mortality in liver transplant recipients.3,5,9,10 However, the available data lacks granularity in terms of etiology of AKI, severity, and reversibility of each episode. It is clear that not all AKI episodes are the same and that AKI encompasses a heterogenous group of etiologies with differing likelihoods for reversibility and/or progression. Although determining the exact etiology for each episode of AKI may not be possible or feasible, capturing their severity and reversibility, as harbingers of prognosis can be reliably accomplished. Interestingly, studies in the general population, have documented that regardless of disease severity, an episode of AKI that resolves rapidly is associated with better outcomes than one which persists for a longer duration.16 However such data are lacking in the liver transplant population. In this study, we attempted to determine the impact of an episode of AKI that is persistent or progressive in course, on patient outcomes and liver allograft survival posttransplantation.
This was a retrospective cohort study, including all patients who received a deceased donor liver transplant between January 1, 2011. and December 31, 2015, at a single Canadian transplant center. Our study protocol was approved by The University of British Columbia's Clinical Ethics Review Board.
Medical charts of LT recipients were reviewed for baseline characteristics including: Age, sex, BMI, underlying liver disease, the MELD and Child-Pugh Scores, inpatient status at the time of transplantation, medical comorbidities, including history of diabetes mellitus, hypertension, and CKD.
The following data were collected on donor and graft variables: age, sex, BMI, cause of death, donor origin (local, provincial, national, or international), donor warm ischemic time (defined as the period between systolic blood pressure <60 mm Hg, mean arterial pressure <50 mm Hg or oxygen saturations <60% and cold reperfusion with preservation solution), and cold ischemic time.
Intraoperative variables collected included Operative time, surgical technique (classic or piggyback), total estimated blood loss, intraoperative transfusion requirements (packed red cell concentrate, fresh frozen plasma, and platelet transfusions), autologous red cell transfusions, and recipient warm ischemic time.
Infectious complications postoperatively were defined as an episode of bacterial or fungal sterile site infections (for example, blood stream, deep incisional or organ space infections) as defined by Centers for Disease Control and National Healthcare Safety Network criteria17 during the first hospital admission for LT.
Episodes of moderate to severe AKI were defined according to the AKI Network (AKIN) classification as peak serum creatinine (SCr) 2.0 to 2.9 times (stage II) or ≥3.0 times (Stage III) the baseline SCr levels.18 Patients who required renal replacement therapy were also classified as AKIN stage III. Baseline serum creatinine before transplant was defined as the lowest creatinine available within 3 months before transplantation. Patients who did not have a history of CKD or a documented episode of moderate to severe AKI pretransplant, were assumed to have had normal renal function before LT. Moderate to severe AKI episodes (stage II or III) were recorded immediately before transplantation and after surgery until hospital discharge. Only AKI episodes during the initial hospital admission for LT were captured and designated as “early” posttransplant AKI. Persistent or progressive AKI (ppAKI) was defined as either a preoperative episode of moderate or severe AKI that failed to improve (normalization of SCr to baseline value or a decrease in SCr sufficient enough for downstaging of AKIN stage), or worsened (AKIN stage II progressing to III) by the time of hospital discharge, or the development of a new episode of moderate or severe AKI postoperatively in patients with normal renal function before transplantation. Patients who never developed AKI (either preoperatively or postoperatively), or had their AKI episode improve (downstaging of AKIN classification) or completely resolved post-LT, served as the comparator control group. All patients were followed up until either the administrative censoring at the end of the follow-up period on March 31, 2017, or until their death, or re-transplantation.
In our center, standard immunosuppression protocol consists of tacrolimus aiming for trough levels of 4 to 8 ng/mL, mycophenolate mofetil 1 g twice daily, and tapering corticosteroids for the first 4 months post-LT. In patients with a history of kidney dysfunction, a renal sparing immunosuppression protocol can be initiated at the discretion of the treating physician. This protocol includes induction with interleukin-2 receptor antagonist, basiliximab, and delayed introduction of calcineurin inhibitors in addition to mycophenolate and steroids.
Numerical data is presented as mean with standard error when normally distributed and nonparametric data as median with the interquartile range (IQR). Categorical data are presented with their corresponding percentages. Continuous numerical data are compared using the Student t test; categorical data is compared using the χ2 test or Fisher exact test when applicable. Nonparametric data were compared using the Kruskal-Wallis test.
Our primary outcome was the rate of graft loss that included a composite of death or retransplantation due to graft failure. We reported the incidence density rate (IDR) that accounts for variable lengths of follow-up for each patient. Incidence density rate for graft failure and the time to graft failure was compared in patients with ppAKI against controls. Patients that had graft loss as a result of recurrent hepatocellular cancer or death from reasons unrelated to their graft function were included in the IDR analysis, but their deaths were not counted as part of the composite outcome. We also performed a time to event analysis. As some patients could have been transplanted multiple times, we used a time to recurrent event analysis using a counting process. In a counting process, when a subject experiences multiple events, their follow-up time is reset at the time of the recurrence as opposed to be continued from the time of inclusion into the study. The analysis time started at the time of transplantation and ended when a patient died or was retransplanted as a consequence of graft dysfunction. Patients who died for reasons unrelated to graft function or from recurrent hepatocellular carcinoma were censored at the time of death. Administrative censoring occurred on the last day of follow-up set for this cohort (March 31, 2017). Survival curves were estimated with the Kaplan-Meier method and the Cox proportional regression model was used to estimate the effect of ppAKI on graft survival adjusted with variables that have been associated with increased risk of graft loss, including development of ischemic cholangiopathy (IC), hepatic artery thrombosis and MELD score at the time of transplant. Analyses were performed using STATA 14 (StataCorp, College Station, TX).
Two hundred seventy-nine patients received 301 deceased donor liver allografts. Before transplantation, 23% (70/301) developed moderate to severe AKI and 9% (28/301) required renal replacement therapy, while posttransplantation, more than half developed AKI (157/301) with 15% (46/301) requiring dialysis. Progressive or persistent AKI was documented in approximately 50% of transplant cases (152/301). A number of patient, donor and intraoperative variables were associated with ppAKI (Table 1). Recipients' age, BMI, and etiology of liver disease were similar between ppAKI group and the controls. However, patients in the ppAKI group had higher MELD scores, higher baseline SCr and more frequently had a history of CKD before LT. A significantly higher proportion of patients in the ppAKI group were inpatients just before transplantation as compared to the controls (54% vs 29.5%; P < 0.001). The majority of donor variables collected were similar between the 2 groups, except donor origin and cold ischemic time. More grafts were procured from out of country or out of province donors in the ppAKI group as compared with the controls (17.8% vs 9.4%) but this was not statistically significant (P = 0.14). The cold ischemic time was significantly longer in the ppAKI group (429.5 vs 378 min, P = 0.0018). Total estimated blood loss, intraoperative transfusions and transfusions during the first 24 hours in intensive care unit (ICU) were significantly higher in ppAKI group as compared to the controls (P < 0.001) (Table 1).
More patients in the ppAKI group (20.4%) also had infectious complications postoperatively as compared with the controls (6.7%; P < 0.001). Renal sparing immunosuppression regimen with basiliximab induction and delayed tacrolimus initiation was utilized in 65% (98/152) of ppAKI group and 25% of controls. Patients with ppAKI had a prolonged length of ICU stay (mean, 17.4 days; 95% confidence interval [95% CI], 12.0-22.8) as compared with the controls (mean, 4.3 days; 95% CI, 3.3-5.2) (P < 0.0001). Length of hospital stay was also significantly prolonged in ppAKI group (mean, 37.9 days; 95% CI, 31.4-44.3) when compared to the controls (mean, 20.8 days; 95% CI, 18.2-23.4) (P < 0.001).
Over a median follow-up of 23.9 months (IQR 8.9, 41.2), a significantly higher proportion of grafts were lost in the ppAKI group (38/152) as compared with the controls (13/149)—the IDR of graft failure was 13.78 and 3.79 per 100 case-years in the ppAKI group and the controls respectively (IDR ratio, 3.64; 95% CI, 1.88-7.50). There was also a significant difference in the time to graft loss between the ppAKI group and the controls. Figure 1 shows the Kaplan-Meier curves of graft loss for the 2 groups. Multiorgan failure (MOF), primary graft nonfunction (PNF), hepatic artery thrombosis (HAT), and IC were the major causes of graft loss in the ppAKI group, accounting for more than 90% of graft loss in this cohort (Table 2). In multivariate Cox analysis, after adjusting for hepatic artery thrombosis, IC, infectious complications, and MELD score, ppAKI was independently associated with a decreased graft survival time (hazard ratio, 3.13; 95% CI, 1.67-5.86) (Table 3).
In this large, single-center retrospective study, we have demonstrated that acute kidney injury is a dynamic process and a harbinger of poor graft survival. Based on our results, lack of resolution of pretransplant AKI episodes or development of moderate to severe AKI episodes after LT is an independent predictor of graft loss, with an impact only surpassed by hepatic artery thrombosis and IC.
Acute kidney injury was a common complication in our cohort both before and after LT with an incidence of 23% and 52%, respectively. Although the incidence of AKI in our study is congruent with the literature, we found that in nearly 50% of our patients, the AKI episode either failed to resolve or worsened by the time of hospital discharge or death. Patients with such progressive or persistent AKI, not only had an extended length of ICU and hospital stay, but also suffered a significantly higher graft loss. In our study, 25% of the grafts were lost in the ppAKI group versus only 8.7% in the controls.
The impact of AKI on posttransplant morbidity and mortality is well documented.1,5,9,10,14 However, to our knowledge, no study has addressed AKI as a dynamic process in this setting. It is increasingly accepted that a diagnosis of AKI represents a heterogenous group of etiologies with widely differing clinical course, propensity for resolution and patient outcomes.16 Nadim and colleagues14 elegantly showed that pretransplant patients with severe AKI due to hepatorenal syndrome had much improved 1- and 5-year survivals posttransplant, as compared with those with pre-LT acute tubular necrosis. This again highlights that the reversibility of AKI, rather than its occurrence, determines its impact on patient outcomes. Similarly, post-LT AKI has been linked to poor patient and graft survival; however, the literature lacks granularity in terms of whether the AKI's “phenotype” and its duration or reversibility would influence patient outcomes. Here, we were able to show that persistent or progressive AKI predicts a high rate of graft loss post-LT rather than the occurrence of an AKI episode itself.
Although we have demonstrated a significant association between ppAKI and graft loss, our study was not designed to determine a possible pathophysiologic mechanism whereby ppAKI negatively impacts graft survival. It is interesting to speculate that ppAKI could result in a physiologic “milieu” that may result in prolonged and more difficult transplant surgery, delayed surgical recovery, coagulopathy at the microvascular level that may result in microvascular organ ischemia, an increased predisposition for postoperative infections that may not be clinically obvious and general impaired recovery. These subtle factors may become exaggerated when the donor organ is less than optimal, creating a complex multifactorial pathophysiologic situation that results in graft loss. Alternatively, the kidney could be simply a bystander that serves as a surrogate marker, during the acute events surrounding and leading to graft loss, such as delayed graft function, sepsis, and MOF. Recent studies have linked AKI to high risk allografts, donation after circulatory death (DCD) organs, or donation after neurological death (NDD) organs with prolonged cold ischemic time.6,19-21 It is thought that the kidney is affected by hepatic ischemia reperfusion injury, which is associated with a systemic inflammatory response that can cause AKI through hemodynamic alteration and direct renal tubular damage.19 Certainly in our cohort, the cold ischemic time was significantly longer in the ppAKI group versus the controls, signaling perhaps a more severe ischemic reperfusion injury in high risk allografts. In addition, more than 50% (20/38) of graft loss in the ppAKI group were due to MOF and PNF, again supporting a relationship between high risk allografts and AKI.
Just as important are identifying risk factors for the development of AKI, especially because there are no effective treatments to modify the course of AKI once renal insult occurs. Identifying risk factors will allow the clinician to anticipate AKI and to resume a kidney centric therapeutic approach, such as avoiding nephrotoxic agents, if possible, and close monitoring of patients' volumes status, in an attempt to halt the progression of AKI. We were able to identify recipient (high MELD score, high medical acuity at the time of transplant, and history of CKD), donor (longer cold ischemic times), and intraoperative (high transfusion requirements) factors for ppAKI in our population. Previous studies had also demonstrated that DCD LT is associated with an increased risk of AKI.19 In our study, we did not find such association although there was a trend toward higher incidence of ppAKI with DCD versus NDD donors (20% vs 14%), but this difference did not reach statistical significance, likely due to small number of DCD transplant in our cohort (50 vs 251).
Our study is limited by its retrospective design, and our results should be confirmed with a prospective study that attempts to explore physiologic variables that are associated with renal dysfunction in LT. In particular, the interaction between these variables and the organ procurement/surgical implantation process should be explored. Furthermore, elucidating the long-term renal outcomes and their impact on graft survival would be of interest. It is increasingly known that AKI and CKD are interconnected syndromes rather than separate entities, with one being a risk factor for the other.22 Although the relationship between AKI and CKD is well documented, it would be interesting to delineate the impact of AKI and CKD on patient and graft survival in future studies.
In summary, we have demonstrated that acute kidney injury, especially early persistent/progressive kidney injury is associated with decreased graft survival time and could serve as a marker for potential graft loss post-LT.
1. Charlton MR, Wall WJ, Ojo AO, et al. Report of the first international liver transplantation society expert panel consensus conference on renal insufficiency in liver transplantation. Liver Transpl
2. Chen J, Singhapricha T, Hu K, et al. Postliver transplant acute renal injury and failure by the RIFLE criteria in patients with normal pretransplant serum creatinine concentrations: a matched study. Transplantation
3. Rahman S, Davidson BR, Mallett SV. Early acute kidney injury after liver transplantation: predisposing factors and clinical implications. World J Hepatol
4. Nair S, Verma S, Thuluvath PJ. Pretransplant renal function predicts survival in patients undergoing orthotopic liver transplantation. Hepatology
5. O'Riordan A, Wong V, McQuillan R, et al. Acute renal disease, as defined by the RIFLE criteria, post-liver transplantation. Am J Transplant
6. Leithead JA, Armstrong MJ, Corbett C, et al. Hepatic ischemia reperfusion injury is associated with acute kidney injury following donation after brain death liver transplantation. Transpl Int
7. Klaus F, Keitel da Silva C, Meinerz G, et al. Acute kidney injury after liver transplantation: incidence and mortality. Transplant Proc
8. Barreto AG, Daher EF, Silva Junior GB, et al. Risk factors for acute kidney injury and 30-day mortality after liver transplantation. Ann Hepatol
9. Hilmi IA, Damian D, Al-Khafaji A, et al. Acute kidney injury following orthotopic liver transplantation: incidence, risk factors, and effects on patient and graft outcomes. Br J Anaesth
10. Zhu M, Li Y, Xia Q, et al. Strong impact of acute kidney injury on survival after liver transplantation. Transplant Proc
11. Gonwa TA, McBride MA, Anderson K, et al. Continued influence of preoperative renal function on outcome of orthotopic liver transplant (OLTX) in the US: where will MELD lead US? Am J Transplant
12. Gonwa TA, Klintmalm GB, Levy M, et al. Impact of pretransplant renal function on survival after liver transplantation. Transplantation
13. Horvatits T, Pischke S, Proske VM, et al. Outcome and natural course of renal dysfunction in liver transplant recipients with severely impaired kidney function prior to transplantation. United European Gastroenterol J
14. Nadim MK, Genyk YS, Tokin C, et al. Impact of the etiology of acute kidney injury on outcomes following liver transplantation: acute tubular necrosis versus hepatorenal syndrome. Liver Transpl
15. Ojo AO, Held PJ, Port FK, et al. Chronic renal failure after transplantation of a nonrenal organ. N Engl J Med
16. Chawla LS, Bellomo R, Bihorac A, et al. Acute kidney disease and renal recovery: consensus report of the acute disease quality initiative (ADQI) 16 workgroup. Nat Rev Nephrol
17. Horan TC, Andrus M, Dudeck MA. CDC/NHSN surveillance definition of health care-associated infection and criteria for specific types of infections in the acute care setting. Am J Infect Control
18. Mehta RL, Kellum JA, Shah SV, et al. Acute kidney injury network: report of an initiative to improve outcomes in acute kidney injury. Crit Care
19. Leithead JA, Tariciotti L, Gunson B, et al. Donation after cardiac death liver transplant recipients have an increased frequency of acute kidney injury. Am J Transplant
20. Leithead JA, Rajoriya N, Gunson BK, et al. The evolving use of higher risk grafts is associated with an increased incidence of acute kidney injury after liver transplantation. J Hepatol
21. Kalisvaart M, Schlegel A, Umbro I, et al. The impact of combined warm ischemia time on development of acute kidney injury in donation after circulatory death liver transplantation: stay within the golden hour. Transplantation
22. Chawla LS, Eggers PW, Star RA, et al. Acute kidney injury and chronic kidney disease as interconnected syndromes. N Engl J Med