ORIGINAL ARTICLE
Recurrent kidney injury in critically ill surgical patients is common and associated with worse outcomes Donald G. Harris, MD, Grace Koo, Michelle P. McCrone, Thomas M. Scalea, MD, William C. Chiu, MD, Jose J. Diaz, MD, and Matthew E. Lissauer, MD, Baltimore, Maryland Acute kidney injury (AKI) is common in critically ill surgery patients. Patients who recover are at risk for recurrence, but recurrent kidney injury (RKI) is not well studied. METHODS: This was a retrospective 12-month cohort study of adults consecutively admitted to a noncardiac, non-trauma surgical intensive care unit. Patients were identified from a prospective critical care database, and kidney injury events were diagnosed and graded by RIFLE criteria. Patients who recovered from AKI were analyzed, and the primary end point was RKI (defined as kidney injury occurring after recovery from an index AKI event). Outcomes were inpatient and 1-year mortality, inpatient lengths of stay, and discharge creatinine. RESULTS: Of 624 patients, 296 (47%) had AKI and 216 (73%) recovered. Of these, 68 (31%) developed RKI. AKI in progress on hospital admission was associated with recurrence, but otherwise RKI and non-RKI patients had similar demographics, comorbidities, and inpatient clinical factors. Recurrence was associated with significantly higher inpatient and 12-month mortality, greater resource use, and worse discharge renal function. CONCLUSION: RKI is common among critically ill surgical patients who recover from an index episode. Recurrence is a clinically significant event and is associated with worse renal and patient outcomes. Future studies should further define this process. (J Trauma Acute Care Surg. 2014;76: 1397Y1401. Copyright * 2014 by Lippincott Williams & Wilkins) LEVEL OF EVIDENCE: Prognostic and epidemiologic study, level III. KEY WORDS: Acute kidney injury; recurrence; critical care; epidemiology. BACKGROUND:
A
cute kidney injury (AKI) is among the most common forms of organ dysfunction in critically ill patients.1Y3 It is an early predictor of multiorgan dysfunction and is associated with increased short- and long-term mortality and risk for progression to chronic kidney disease (CKD).4Y19 Patients who recover from AKI are at risk for recurrent kidney injury (RKI) during the same hospitalization. Indeed, preventing recurrence is an important component of managing AKI.20Y22 Animal and clinical studies indicate that RKI is an important phenomenon. In murine models, AKI predisposes the animals to secondary insults, subjecting the recovering kidneys to more severe injury and dysfunction with RKI.23,24 Suggesting a unique pathophysiologic basis for recurrence, AKI induces impaired vascular autoregulation and altered immune responses, which may potentiate the effects of recurrent injury.23Y25 RKI may also contribute to worse long-term outcomes: in a 10-year cohort study of diabetic patients by Thakar et al.,16 30% of the
Submitted: January 21, 2014, Revised: February 25, 2014, Accepted: February 26, 2014. From the Division of General Surgery (D.G.H., G.K., M.P.M.), Department of Surgery, and R Adams Cowley Shock Trauma Center (T.M.S., W.C.C., J.J.D., M.E.L.), University of Maryland School of Medicine. Baltimore, Maryland. This study was presented at the American College of Surgeons Clinical Congress (surgical forum), October 2013, in Washington, District of Columbia (3-month/ preliminary data), and at the Society for Critical Care Medicine Critical Care Congress (poster presentation), January 2014, in San Francisco, California (12-month/final data). Supplemental digital content is available for this article. Direct URL citations appear in the printed text, and links to the digital files are provided in the HTML text of this article on the journal’s Web site (www.jtrauma.com). Address for reprints: Donald G. Harris, MD, Division of General Surgery, University of Maryland Medical Center, 22 S Greene St, Baltimore, MD 21201-1590; email: [email protected]. DOI: 10.1097/TA.0000000000000241
patients who developed AKI had multiple episodes, which were associated with increased risk for progression to CKD. AKI is well studied in intensive care unit populations, where its clinical significance is most apparent.4,5,13,18,26 However, there remain to be limited data on RKI despite its effects on renal recovery and patient mortality. Importantly, RKI has not been evaluated in critically ill patients, who potentially have the greatest risk for recurrence. The aims of this study were to determine the incidence, severity, and outcomes of RKI and identify its risk factors in a cohort of critically ill surgical patients. We hypothesized that among patients who recover from AKI, RKI during the index hospitalization is common and associated with increased mortality and worse renal function at hospital discharge.
PATIENTS AND METHODS This study was approved by the University of Maryland School of Medicine Institutional Review Board and performed at the University of Maryland Medical Center, a quaternary academic medical center. The surgical intensive care unit (SICU) is closed and is the primary critical care unit for general, vascular, thoracic, abdominal transplant, orthopedic, otolaryngology, and oral-maxillofacial surgery patients and is a secondary unit for neurosurgery patients. Separate, dedicated intensive care units admit trauma and cardiac surgery patients.27 Consecutive adult SICU admissions from January 1 through December 31, 2012, were identified from a prospectively collected Acute Physiology and Chronic Health Evaluation (APACHE) IV database (Cerner, Kansas City).28Y31 Data were collected from the APACHE database and via individual chart review. Longterm vital status was determined from the Social Security Death
J Trauma Acute Care Surg Volume 76, Number 6
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
1397
J Trauma Acute Care Surg Volume 76, Number 6
Harris et al.
Index. Exclusion criteria included fewer than two inpatient creatinine values, preexisting dialysis dependence, resolved kidney injury from a previous hospitalization, and nephrectomy or organ transplant. Patients who developed and recovered from AKI and were thus at risk for recurrence were included for analysis. Baseline serum creatinine was determined to be the lowest of (1) any value within 1 year before admission, (2) hospital or SICU admission creatinine, or (3) the Modification of Diet in Renal Disease equation solved for creatinine assuming a glomerular filtration rate of 75 mL per minute per 1.73 m2.4,32 For patients with CKD, only the first two of these values were considered to avoid an artificially low baseline creatinine, while including the third, calculated value ensured that patients admitted with AKI in progress were not given an artificially high baseline.4 With the use of baseline creatinine, inpatient AKI was diagnosed and graded by RIFLE creatinine criteria.33 Renal recovery was defined as at least two serial creatinine values below a patient’s AKI-Risk threshold (i.e., G1.5 baseline creatinine), sustained as a stable or decreasing trend for more than 24 hours without renal replacement therapy (RRT). RKI was any subsequent renal event during the index admission and was similarly diagnosed and graded. Peak creatinine, kidney injury exposure (total time with renal dysfunction meeting AKI definition, regardless of recovery), event duration (time from event onset to recovery), RRT use, renal recovery rates, and best recovery creatinine were determined for AKI and RKI episodes. RKI was the primary end point. Patients with AKI who recovered without further renal injury were classified as AKI, and those with recurrence during the same hospitalization were grouped as RKI. The AKI and RKI groups were compared to determine risk factors and outcomes associated with recurrence. Demographics, comorbidities (hypertension, diabetes, and CKD), illness severity, and index AKI severity were evaluated as potential risk factors for RKI. Primary outcomes were inpatient and 1-year mortality (both overall and censored for inpatient deaths), SICU and hospital lengths of stay, and discharge renal function. Secondary outcomes were analyzed between all AKI and RKI events and included kidney injury event severities, renal recovery rates, and RRT use. Categorical data were analyzed by two-sided Pearson W2 or Fisher’s exact tests as appropriate. Normally distributed continuous data were analyzed using unpaired Student’s t test and is reported as mean (SD), while other continuous data were compared by Mann-Whitney U-test and is reported as median (interquartile range [IQR]). To assess the independent effect of RKI on outcomes, multivariate logistic regression analysis of potential risk factors for inpatient and 1-year mortality was performed (Supplemental Table I, http://links.lww.com/TA/A410). p values less than 0.05 were accepted as significant.
RESULTS During the study period, 886 unique patients were admitted to the SICU (Fig. 1). A total of 262 patients were excluded: 3 with single creatinine values, 40 on chronic dialysis, 39 with previous AKI, 167 transplant recipients, and 25 patients status postnephrectomy. For the 624 included patients, the mean (SD) age was 59 (15) years, 369 (59%) were male, and the mean (SD) APACHE III score was 54 (28). Preadmission creatinine 1398
Figure 1. Patient flowchart.
values were available for 218 patients (35%) and used to define the baseline renal function for 114 patients (18%). Hospital and SICU admission creatinine values were used as the baseline for 252 patients (40%) and 130 patients (21%), respectively, while a calculated value was used for 128 patients (21%). A total of 296 (47%) developed AKI, and 216 (73%) recovered and were at risk for recurrence. Of the group that recovered, 68 patients (31%) subsequently developed RKI (RKI group), while 148 (69%) did not (AKI group). Median cumulative kidney injury exposure for AKI and RKI groups were 3 days and 10 days (IQRs, 2Y6 days and 6Y19 days, respectively; p G 0.0001). For RKI patients, the mean (SD) time between AKI and RKI diagnoses was 17 (12) days, and the mean (SD) period from AKI recovery to RKI onset was 12 (10) days. Admission and inpatient clinical characteristics generally were not associated with the development of RKI (Table 1). In particular, demographics, comorbidities, baseline renal function, and illness severity did not predict RKI in those who recovered from AKI. Similarly, index AKI peak creatinine, duration, and RRT use were not significantly different between the RKI and AKI groups (Table 1, Fig. 2), suggesting that the initial episode severity does not drive recurrence. However, presence of index AKI in progress at hospital admission was associated with subsequent RKI (odds ratio, 1.9; 95% confidence interval [CI], 1.0 Y3.6). RKI patients also had higher AKI recovery creatinine values when normalized to baseline, which despite returning to below the RIFLE-Risk threshold may indicate that the degree or phase of renal recovery may contribute to recurrence. RKI was associated with significantly increased shortand long-term mortality (Table 2). RKI patients had higher inpatient (21% vs. 9%, p = 0.01) and overall 1-year mortality (46% vs. 24%, p = 0.002). In addition, for patients discharged alive from the hospital, RKI was associated with increased mortality within 1-year (31% vs. 16%, p = 0.01), resulting in an overall odds ratio for death within 1-year of discharge of 2.6 (95% CI, 1.4Y5.1). After adjusting for potential risk factors for short- and long-term death, RKI remained independently associated with increased odds for inpatient (1.1; 95% CI, 1.0Y1.2) * 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
J Trauma Acute Care Surg Volume 76, Number 6
Harris et al.
TABLE 1. Demographic, Clinical, and AKI Characteristics of Patients With and Without RKI
Demographics Age, mean (SD), y Male, n (%) Black, n (%) CKD, n (%) Hypertension, n (%) Diabetes mellitus, n (%) Baseline creatinine, mean (SD), mg/dL Inpatient factors APACHE III score APACHE III surgical admission,* n (%) Surgery for principle diagnosis, n (%) Emergency admission,** n (%) AKI characteristics Present on admission, n (%) Peak creatinine, mean (SD), mg /dL RIFLE grade, n (%) Risk, n (%) Injury, n (%) Failure, n (%) RRT use, n (%) RRT days, mean (SD) Duration, median (IQR), d Best recovery: baseline creatinine†
AKI Only, n = 148
RKI, n = 68
p
61.5 (13.9 88 (59) 53 (36) 7 (5) 72 (49) 51 (34) 0.84 (0.37)
63.5 (13.3 37 (54) 23 (34) 6 (9) 37 (54) 25 (37) 0.86 (0.41)
0.32 0.49 0.78 0.24 0.43 0.74 0.70
61.5 (25.8 88 (59) 126 (85) 96 (64)
63.1 (23.9 41 (60) 61 (90) 46 (68)
0.66 0.91 0.36 0.69
55 (37) 1.95 (1.33)
36 (53) 1.75 (0.91)
0.03 0.26
89 (60) 36 (24) 23 (16) 10 (7) 10 (8) 3 (2Y6) 1.0 (0.2)
43 (63) 18 (26) 9 (13) 7 (10) 8 (5) 4 (2Y8) 0.9 (0.3)
0.66 0.74 0.66 0.37 0.69 0.44 0.003
*Admission to SICU directly from an operating or recovery room. **Admission to hospital for non-elective diagnosis. †Best recovery creatinine is the lowest creatinine achieved after AKI recovery.
and 1-year mortality (1.2; 95% CI 1.1Y1.3; Supplemental Table I, http://links.lww.com/TA/A410). In comparison, although RKI patients had lower inpatient mortality compared with patients who did not recover from the initial AKI episode (21% vs. 37%, p = 0.03), censored 1-year mortality rates were similar (31% vs. 26%, p = 0.54). In addition, RKI was associated with longer hospital stay and worse discharge renal function. Secondary analyses of AKI and RKI episodes indicated that recurrence is as severe as the index events (Table 3, Fig. 3). RKI
TABLE 2. Outcomes of Patients With and Without RKI Outcome
AKI Only, n = 148
RKI, n = 68
p
Inpatient mortality, n (%) Overall 1-y mortality, n (%) Postdischarge mortality,* n (%) SICU LOS, median (IQR), d Hospital LOS, median (IQR), d Discharge creatinine, mean (SD), mg/dL
13 (9) 31 (46) 21 (16) 9 (11) 23 (17) 0.86 (0.40)
14 (21) 36 (24) 17 (31) 13 (15) 37 (31) 1.34 (1.01)
0.01 0.002 0.01 0.04 G0.0001 G0.0001
*For patients discharged alive from the hospital. LOS, length of stay.
event severity, duration, and RRTuse were similar to AKI. AKI and RKI events had similar renal recovery rates (73% vs. 65%, p = 0.17) and kidney injuryYassociated mortality (11% vs. 15%, p = 0.41).
DISCUSSION This is an important initial investigation of RKI in a cohort of critically ill patients. The principal findings demonstrate that RKI is common among patients who recover from AKI and is associated with worse outcomes and greater resource use than those who recover from AKI and do not recur. Unfortunately, traditional risk factors failed to predict differences between patients with AKI only and those who develop recurrence. What is not clear from this study is if recurrence reflects the natural history of AKI, complications of caring for critically ill patients, or if RKI is simply another AKI event from a second insult. RKI is a clinically significant phenomenon. RKI is common among critically ill patients, occurring in nearly one third of those who recover from an index AKI event. RKI was as severe as AKI and was associated with additional loss of renal function and episode-related mortality. Given the worse outcomes associated with recurrence, it is likely that morbidity and mortality are cumulative in patients with RKI. It is possible that morbidity and mortality from AKI are ‘‘dose dependent’’ and related to number TABLE 3. Comparison of AKI and RKI Episodes Outcome Peak creatinine, mean (SD), mg/dL RIFLE grade, n (%) Risk Injury Failure RRT use, n (%) RRT days mean (SD) Median kidney injury exposure,* median (IQR), d Renal recovery, n (%) Kidney injury duration,** median (IQR), d Best recovery: baseline creatinine† Death during episode, n (%)
Figure 2. Comparison of index AKI events between patients with and without RKI. Differences between peak creatinine, rates of onset, and event durations were not significant.
AKI Events, n = 296
RKI Events, n = 68
p
2.17 (1.55)
1.96 (1.48)
0.31
155 (52) 69 (23) 76 (26) 38 (13) 11 (12) 4 (2Y10)
41 (60) 17 (25) 12 (18) 7 (10) 9 (5) 4 (2Y12)
0.24 0.77 0.16 0.57 0.66 0.87
216 (73) 3 (2Y6)
44 (65) 4 (2Y9)
0.17 0.10
0.9 (0.2) 33 (11)
1.0 (0.3) 10 (15)
0.16 0.41
*Time with kidney injury for all affected patients. **Time from event onset to recovery. †Lowest creatinine achieved after AKI recovery.
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
1399
J Trauma Acute Care Surg Volume 76, Number 6
Harris et al.
Figure 3. Comparison of AKI and RKI. Differences between peak creatinine, rates of onset, and event durations were not significant.
and duration of episodes. This should be evaluated by future studies. Although this study did not evaluate the effect of recurrence on long-term renal function, RKI patients had higher discharge creatinine values, reflecting at least short term persisting renal dysfunction. The current findings are consistent with previous studies of repetitive kidney injury. The incidence of RKI among at-risk patients in this cohort was similar to the rate of multiple events reported by Thakar et al.16 in a long-term study of diabetic patients. Furthermore, our results showing that RKI contributes to increased mortality, resource use, and persistent renal dysfunction are consistent with studies showing progressively worse outcomes with greater exposure to kidney injury.4Y6,15,16,34 In this cohort, RKI events were as severe as AKI, resulting in greater cumulative exposure to renal injury and dysfunction, which may account for the significantly worse outcomes associated with recurrence. Although these findings indicate that RKI is clinically important, long-term renal outcomes are lacking. Traditional kidney injury risk factors did not predict recurrence, and only the presence of AKI on hospital admission and higher recovery creatinine values were associated with RKI. Although the latter difference was relatively small, because slight changes in creatinine are associated with worse outcomes,13,17,34,35 this finding may reflect a minor but clinically significant difference in the effect of AKI recovery on recurrence risk. However, AKI event severities were not significantly different between the AKI and RKI groups. As such, among patients who recover from AKI, those at risk for recurrence may not be readily identifiable. One potential explanation is that patients with greater risk for RKI, such as more comorbidities or higher illness severity, do not typically recover and therefore are not exposed to risk of recurrence. Other variables not taken into account in this study may identify patients at risk for RKI development. Few studies have investigated recurrent or repetitive kidney injury, and to our knowledge, this is the first to describe RKI during the same hospital stay in either surgical or critically ill patients. Given the association between RKI and worse renal and patient outcomes, recurrence may be an important outcome for future AKI studies to measure. Similarly, since kidney injury is largely preventable and therapy supportive to limit further 1400
renal insults,20Y22,36 RKI may be a useful clinical, research, or quality marker of the efficacy of AKI management. This study has several important limitations. Because surgical patients have unique clinical conditions and risk factors for kidney injury19,37 and perhaps increased risk for recurrence from perioperative care, these findings may not be representative of RKI in other patient populations. In classifying kidney injury as a singular process, we may have missed differences among common etiologies such as sepsis, malperfusion, and nephrotoxic agents that may entail different risks for recurrence or may be responsible for the recurrence. Future work will assess these variables. Although the use of creatinine alone to define and characterize kidney injury events facilitated precise study of a large cohort, it may not have been accurate for patients who met RIFLE criteria by urine output but not creatinine criteria. The limited availability of postdischarge creatinine values prevented us from analyzing the effect of RKI on long-term renal function. However, by accurately assessing 1-year mortality using the Social Security Death Index, this served as a clinically meaningful long-term outcome. Finally, as with all retrospective database studies, there are concerns for observation bias. Despite these weaknesses, the study does provide an important and novel evaluation of RKI in critically ill surgical patients. In conclusion, this novel investigation of RKI gives important early insight into an important complication among patients who recover from AKI. RKI is common among critically ill patients, affecting approximately one third of those at risk and is associated with increased mortality, greater renal dysfunction, and higher resource use. The presence of AKI in progress at admission is associated with greater risk for recurrence; these patients should be recognized as being at increased risk for worse renal and overall outcomes and managed accordingly. Future studies should be aimed at better defining modifiable risk factors for recurrence and evaluate the long-term effect of RKI on renal function. AUTHORSHIP D.G.H., J.J.D., and M.E.L. conceived and planned the study. D.G.H., G.K., M.P.M., and M.E.L. performed the data collection and analysis. D.G.H., T.M.S., W.C.C., J.J.D., and M.E.L. contributed to the data interpretation. D.G.H., G.K., M.P.M., and M.E.L. drafted the manuscript, and T.M.S., W.C.C., and J.J.D. provided critical review.
DISCLOSURE The authors declare no conflicts of interest.
REFERENCES 1. Gaieski DF, Edwards JM, Kallan MJ, Carr BG. Benchmarking the incidence and mortality of severe sepsis in the United States. Crit Care Med. 2013;41:1167Y1174. 2. Vincent J, De Mendonc¸a A, Cantraine F, et al. Use of the SOFA score to assess the incidence of organ dysfunction/failure in intensive care units: results of a multicenter, prospective study. Crit Care Med. 1998;26: 1793Y1800. 3. Martin GS, Mannino DM, Eaton S, Moss M. The epidemiology of sepsis in the United States from 1979 through 2000. N Engl J Med. 2003;348: 1546Y1554. 4. Hoste EAJ, 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:R73.
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
J Trauma Acute Care Surg Volume 76, Number 6
5. Cruz DN, Bolgan I, Perazella MA, et al. North East Italian prospective hospital renal outcome survey on acute kidney injury (NEiPHROS-AKI): targeting the problem with the RIFLE criteria. Clin J Am Soc Nephrol. 2007;2:418Y425. 6. Hoste EAJ, Schurgers M. Epidemiology of acute kidney injury: how big is the problem? Crit Care Med. 2008;36:S146. 7. Schiffl H. Renal recovery from acute tubular necrosis requiring renal replacement therapy: a prospective study in critically ill patients. Nephrol Dial Transplant. 2006;21:1248Y1252. 8. Wald R, Quinn RR, Luo J, et al. Chronic dialysis and death among survivors of acute kidney injury requiring dialysis. JAMA. 2009;302:1179Y1185. 9. Ponte B, Felipe C, Muriel A, Tenorio MT, Lian˜o F. Long-term functional evolution after an acute kidney injury: a 10-year study. Nephrol Dial Transplant. 2008;23:3859Y3866. 10. Metnitz PG, Krenn CG, Steltzer H, et al. Effect of acute renal failure requiring renal replacement therapy on outcome in critically ill patients*. Crit Care Med. 2002;30:2051Y2058. 11. Hobson CE, Yavas S, Segal MS, et al. Acute kidney injury is associated with increased long-term mortality after cardiothoracic surgery. Circulation. 2009;119:2444Y2453. 12. Bihorac A, Baslanti TO, Cuenca AG, et al. Acute kidney injury is associated with early cytokine changes after trauma. J Trauma Acute Care Surg. 2013;74:1005Y1013. 13. Goldberg A, Kogan E, Hammerman H, Markiewicz W, Aronson D. The impact of transient and persistent acute kidney injury on long-term outcomes after acute myocardial infarction. Kidney Int. 2009;76:900Y906. 14. Lo LJ, Go AS, Chertow GM, et al. Dialysis-requiring acute renal failure increases the risk of progressive chronic kidney disease. Kidney Int. 2009;76:893Y899. 15. Schiffl H, Fischer R. Five-year outcomes of severe acute kidney injury requiring renal replacement therapy. Nephrol Dial Transplant. 2008;23: 2235Y2241. 16. Thakar CV, Christianson A, Himmelfarb J, Leonard AC. Acute kidney injury episodes and chronic kidney disease risk in diabetes mellitus. Clin J Am Soc Nephrol. 2011;6:2567Y2572. 17. Goldenberg I, Chonchol M, Guetta V. Reversible acute kidney injury following contrast exposure and the risk of long-term mortality. Am J Nephrol. 2008;29:136Y144. 18. Wohlauer MV, Sauaia A, Moore EE, Burlew CC, Banerjee A, Johnson J. Acute kidney injury and posttrauma multiple organ failure: the canary in the coal mine. J Trauma Acute Care Surg. 2012;72:373Y380. 19. White LE, Hassoun HT, Bihorac A, et al. Acute kidney injury is surprisingly common and a powerful predictor of mortality in surgical sepsis. J Trauma Acute Care Surg. 2013;75:432Y438. 20. Bellomo R, Kellum JA, Ronco C. Acute kidney injury. Lancet. 2012; 380:756Y766. 21. Prowle JR, Echeverri JE, Ligabo EV, Ronco C, Bellomo R. Fluid balance and acute kidney injury. Nat Rev Nephrol. 2009;6:107Y115.
Harris et al.
22. Tolwani A. Continuous renal-replacement therapy for acute kidney injury. N Engl J Med. 2012;367:2505Y2514. 23. Portella VG, Silva-Filho JL, Landgraf SS, et al. Sepsis-surviving mice are more susceptible to a secondary kidney insult*. Crit Care Med. 2013;41:1056Y1068. 24. Kelleher SP, Robinette JB, Miller F, Conger JD. Effect of hemorrhagic reduction in blood pressure on recovery from acute renal failure. Kidney Int. 1987;31:725Y730. 25. Conger J. Hemodynamic factors in acute renal failure. Adv Ren Replace Ther. 1997;4:25. 26. Brinkman S, de Jonge E, Abu-Hanna A, Arbous MS, de Lange DW, de Keizer NF. Mortality after hospital discharge in ICU patients*. Crit Care Med. 2013;41:1229Y1236. 27. Lissauer ME, Galvagno SM Jr, Rock P, et al. Increased ICU resource needs for an academic emergency general surgery service. Crit Care Med. 2014;42:910Y917. 28. Knaus WA, Wagner DP, Zimmerman JE, Draper EA. Variations in mortality and length of stay in intensive care units. Ann Intern Med. 1993;118: 753Y761. 29. Knaus WA, Wagner D, Draper EA, et al. The APACHE III prognostic system. Risk prediction of hospital mortality for critically ill hospitalized adults. CHEST. 1991;100:1619Y1636. 30. Zimmerman JE, Kramer AA, McNair DS, Malila FM. Acute Physiology and Chronic Health Evaluation (APACHE) IV: hospital mortality assessment for today’s critically ill patients*. Crit Care Med. 2006;34:1297Y1310. 31. Zimmerman JE, Wagner DP, Draper EA, Wright L, Alzola C, Knaus WA. Evaluation of acute physiology and chronic health evaluation III predictions of hospital mortality in an independent database. Crit Care Med. 1998;26:1317Y1326. 32. Levey AS, Coresh J, Balk E, et al. National Kidney Foundation practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Ann Intern Med. 2003;139:137Y147. 33. Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P. Acute renal failure-definition, outcome measures, animal models, fluid therapy and information technology needs: the Second International Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group. Crit Care. 2004;8:R204YR212. 34. Chertow GM, Burdick E, Honour M, Bonventre JV, Bates DW. Acute kidney injury, mortality, length of stay, and costs in hospitalized patients. J Am Soc Nephrol. 2005;16:3365Y3370. 35. Lassnigg A, Schmidlin D, Mouhieddine M, et al. Minimal changes of serum creatinine predict prognosis in patients after cardiothoracic surgery: a prospective cohort study. J Am Soc Nephrol. 2004;15:1597Y1605. 36. Li PKT, Burdmann EA, Mehta RL. Acute kidney injury: global health alert. Kidney Int. 2013;83:372Y376. 37. Abelha F, Botelho M, Fernandes V, Barros H. Determinants of postoperative acute kidney injury. Crit Care. 2009;13:R79.
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
1401
Recurrent kidney injury in critically ill surgical patients is common and associated with worse outcomes Donald G. Harris, MD, Grace Koo, Michelle P. McCrone, Thomas M. Scalea, MD, William C. Chiu, MD, Jose J. Diaz, MD, and Matthew E. Lissauer, MD, Baltimore, Maryland Acute kidney injury (AKI) is common in critically ill surgery patients. Patients who recover are at risk for recurrence, but recurrent kidney injury (RKI) is not well studied. METHODS: This was a retrospective 12-month cohort study of adults consecutively admitted to a noncardiac, non-trauma surgical intensive care unit. Patients were identified from a prospective critical care database, and kidney injury events were diagnosed and graded by RIFLE criteria. Patients who recovered from AKI were analyzed, and the primary end point was RKI (defined as kidney injury occurring after recovery from an index AKI event). Outcomes were inpatient and 1-year mortality, inpatient lengths of stay, and discharge creatinine. RESULTS: Of 624 patients, 296 (47%) had AKI and 216 (73%) recovered. Of these, 68 (31%) developed RKI. AKI in progress on hospital admission was associated with recurrence, but otherwise RKI and non-RKI patients had similar demographics, comorbidities, and inpatient clinical factors. Recurrence was associated with significantly higher inpatient and 12-month mortality, greater resource use, and worse discharge renal function. CONCLUSION: RKI is common among critically ill surgical patients who recover from an index episode. Recurrence is a clinically significant event and is associated with worse renal and patient outcomes. Future studies should further define this process. (J Trauma Acute Care Surg. 2014;76: 1397Y1401. Copyright * 2014 by Lippincott Williams & Wilkins) LEVEL OF EVIDENCE: Prognostic and epidemiologic study, level III. KEY WORDS: Acute kidney injury; recurrence; critical care; epidemiology. BACKGROUND:
A
cute kidney injury (AKI) is among the most common forms of organ dysfunction in critically ill patients.1Y3 It is an early predictor of multiorgan dysfunction and is associated with increased short- and long-term mortality and risk for progression to chronic kidney disease (CKD).4Y19 Patients who recover from AKI are at risk for recurrent kidney injury (RKI) during the same hospitalization. Indeed, preventing recurrence is an important component of managing AKI.20Y22 Animal and clinical studies indicate that RKI is an important phenomenon. In murine models, AKI predisposes the animals to secondary insults, subjecting the recovering kidneys to more severe injury and dysfunction with RKI.23,24 Suggesting a unique pathophysiologic basis for recurrence, AKI induces impaired vascular autoregulation and altered immune responses, which may potentiate the effects of recurrent injury.23Y25 RKI may also contribute to worse long-term outcomes: in a 10-year cohort study of diabetic patients by Thakar et al.,16 30% of the
Submitted: January 21, 2014, Revised: February 25, 2014, Accepted: February 26, 2014. From the Division of General Surgery (D.G.H., G.K., M.P.M.), Department of Surgery, and R Adams Cowley Shock Trauma Center (T.M.S., W.C.C., J.J.D., M.E.L.), University of Maryland School of Medicine. Baltimore, Maryland. This study was presented at the American College of Surgeons Clinical Congress (surgical forum), October 2013, in Washington, District of Columbia (3-month/ preliminary data), and at the Society for Critical Care Medicine Critical Care Congress (poster presentation), January 2014, in San Francisco, California (12-month/final data). Supplemental digital content is available for this article. Direct URL citations appear in the printed text, and links to the digital files are provided in the HTML text of this article on the journal’s Web site (www.jtrauma.com). Address for reprints: Donald G. Harris, MD, Division of General Surgery, University of Maryland Medical Center, 22 S Greene St, Baltimore, MD 21201-1590; email: [email protected]. DOI: 10.1097/TA.0000000000000241
patients who developed AKI had multiple episodes, which were associated with increased risk for progression to CKD. AKI is well studied in intensive care unit populations, where its clinical significance is most apparent.4,5,13,18,26 However, there remain to be limited data on RKI despite its effects on renal recovery and patient mortality. Importantly, RKI has not been evaluated in critically ill patients, who potentially have the greatest risk for recurrence. The aims of this study were to determine the incidence, severity, and outcomes of RKI and identify its risk factors in a cohort of critically ill surgical patients. We hypothesized that among patients who recover from AKI, RKI during the index hospitalization is common and associated with increased mortality and worse renal function at hospital discharge.
PATIENTS AND METHODS This study was approved by the University of Maryland School of Medicine Institutional Review Board and performed at the University of Maryland Medical Center, a quaternary academic medical center. The surgical intensive care unit (SICU) is closed and is the primary critical care unit for general, vascular, thoracic, abdominal transplant, orthopedic, otolaryngology, and oral-maxillofacial surgery patients and is a secondary unit for neurosurgery patients. Separate, dedicated intensive care units admit trauma and cardiac surgery patients.27 Consecutive adult SICU admissions from January 1 through December 31, 2012, were identified from a prospectively collected Acute Physiology and Chronic Health Evaluation (APACHE) IV database (Cerner, Kansas City).28Y31 Data were collected from the APACHE database and via individual chart review. Longterm vital status was determined from the Social Security Death
J Trauma Acute Care Surg Volume 76, Number 6
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
1397
J Trauma Acute Care Surg Volume 76, Number 6
Harris et al.
Index. Exclusion criteria included fewer than two inpatient creatinine values, preexisting dialysis dependence, resolved kidney injury from a previous hospitalization, and nephrectomy or organ transplant. Patients who developed and recovered from AKI and were thus at risk for recurrence were included for analysis. Baseline serum creatinine was determined to be the lowest of (1) any value within 1 year before admission, (2) hospital or SICU admission creatinine, or (3) the Modification of Diet in Renal Disease equation solved for creatinine assuming a glomerular filtration rate of 75 mL per minute per 1.73 m2.4,32 For patients with CKD, only the first two of these values were considered to avoid an artificially low baseline creatinine, while including the third, calculated value ensured that patients admitted with AKI in progress were not given an artificially high baseline.4 With the use of baseline creatinine, inpatient AKI was diagnosed and graded by RIFLE creatinine criteria.33 Renal recovery was defined as at least two serial creatinine values below a patient’s AKI-Risk threshold (i.e., G1.5 baseline creatinine), sustained as a stable or decreasing trend for more than 24 hours without renal replacement therapy (RRT). RKI was any subsequent renal event during the index admission and was similarly diagnosed and graded. Peak creatinine, kidney injury exposure (total time with renal dysfunction meeting AKI definition, regardless of recovery), event duration (time from event onset to recovery), RRT use, renal recovery rates, and best recovery creatinine were determined for AKI and RKI episodes. RKI was the primary end point. Patients with AKI who recovered without further renal injury were classified as AKI, and those with recurrence during the same hospitalization were grouped as RKI. The AKI and RKI groups were compared to determine risk factors and outcomes associated with recurrence. Demographics, comorbidities (hypertension, diabetes, and CKD), illness severity, and index AKI severity were evaluated as potential risk factors for RKI. Primary outcomes were inpatient and 1-year mortality (both overall and censored for inpatient deaths), SICU and hospital lengths of stay, and discharge renal function. Secondary outcomes were analyzed between all AKI and RKI events and included kidney injury event severities, renal recovery rates, and RRT use. Categorical data were analyzed by two-sided Pearson W2 or Fisher’s exact tests as appropriate. Normally distributed continuous data were analyzed using unpaired Student’s t test and is reported as mean (SD), while other continuous data were compared by Mann-Whitney U-test and is reported as median (interquartile range [IQR]). To assess the independent effect of RKI on outcomes, multivariate logistic regression analysis of potential risk factors for inpatient and 1-year mortality was performed (Supplemental Table I, http://links.lww.com/TA/A410). p values less than 0.05 were accepted as significant.
RESULTS During the study period, 886 unique patients were admitted to the SICU (Fig. 1). A total of 262 patients were excluded: 3 with single creatinine values, 40 on chronic dialysis, 39 with previous AKI, 167 transplant recipients, and 25 patients status postnephrectomy. For the 624 included patients, the mean (SD) age was 59 (15) years, 369 (59%) were male, and the mean (SD) APACHE III score was 54 (28). Preadmission creatinine 1398
Figure 1. Patient flowchart.
values were available for 218 patients (35%) and used to define the baseline renal function for 114 patients (18%). Hospital and SICU admission creatinine values were used as the baseline for 252 patients (40%) and 130 patients (21%), respectively, while a calculated value was used for 128 patients (21%). A total of 296 (47%) developed AKI, and 216 (73%) recovered and were at risk for recurrence. Of the group that recovered, 68 patients (31%) subsequently developed RKI (RKI group), while 148 (69%) did not (AKI group). Median cumulative kidney injury exposure for AKI and RKI groups were 3 days and 10 days (IQRs, 2Y6 days and 6Y19 days, respectively; p G 0.0001). For RKI patients, the mean (SD) time between AKI and RKI diagnoses was 17 (12) days, and the mean (SD) period from AKI recovery to RKI onset was 12 (10) days. Admission and inpatient clinical characteristics generally were not associated with the development of RKI (Table 1). In particular, demographics, comorbidities, baseline renal function, and illness severity did not predict RKI in those who recovered from AKI. Similarly, index AKI peak creatinine, duration, and RRT use were not significantly different between the RKI and AKI groups (Table 1, Fig. 2), suggesting that the initial episode severity does not drive recurrence. However, presence of index AKI in progress at hospital admission was associated with subsequent RKI (odds ratio, 1.9; 95% confidence interval [CI], 1.0 Y3.6). RKI patients also had higher AKI recovery creatinine values when normalized to baseline, which despite returning to below the RIFLE-Risk threshold may indicate that the degree or phase of renal recovery may contribute to recurrence. RKI was associated with significantly increased shortand long-term mortality (Table 2). RKI patients had higher inpatient (21% vs. 9%, p = 0.01) and overall 1-year mortality (46% vs. 24%, p = 0.002). In addition, for patients discharged alive from the hospital, RKI was associated with increased mortality within 1-year (31% vs. 16%, p = 0.01), resulting in an overall odds ratio for death within 1-year of discharge of 2.6 (95% CI, 1.4Y5.1). After adjusting for potential risk factors for short- and long-term death, RKI remained independently associated with increased odds for inpatient (1.1; 95% CI, 1.0Y1.2) * 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
J Trauma Acute Care Surg Volume 76, Number 6
Harris et al.
TABLE 1. Demographic, Clinical, and AKI Characteristics of Patients With and Without RKI
Demographics Age, mean (SD), y Male, n (%) Black, n (%) CKD, n (%) Hypertension, n (%) Diabetes mellitus, n (%) Baseline creatinine, mean (SD), mg/dL Inpatient factors APACHE III score APACHE III surgical admission,* n (%) Surgery for principle diagnosis, n (%) Emergency admission,** n (%) AKI characteristics Present on admission, n (%) Peak creatinine, mean (SD), mg /dL RIFLE grade, n (%) Risk, n (%) Injury, n (%) Failure, n (%) RRT use, n (%) RRT days, mean (SD) Duration, median (IQR), d Best recovery: baseline creatinine†
AKI Only, n = 148
RKI, n = 68
p
61.5 (13.9 88 (59) 53 (36) 7 (5) 72 (49) 51 (34) 0.84 (0.37)
63.5 (13.3 37 (54) 23 (34) 6 (9) 37 (54) 25 (37) 0.86 (0.41)
0.32 0.49 0.78 0.24 0.43 0.74 0.70
61.5 (25.8 88 (59) 126 (85) 96 (64)
63.1 (23.9 41 (60) 61 (90) 46 (68)
0.66 0.91 0.36 0.69
55 (37) 1.95 (1.33)
36 (53) 1.75 (0.91)
0.03 0.26
89 (60) 36 (24) 23 (16) 10 (7) 10 (8) 3 (2Y6) 1.0 (0.2)
43 (63) 18 (26) 9 (13) 7 (10) 8 (5) 4 (2Y8) 0.9 (0.3)
0.66 0.74 0.66 0.37 0.69 0.44 0.003
*Admission to SICU directly from an operating or recovery room. **Admission to hospital for non-elective diagnosis. †Best recovery creatinine is the lowest creatinine achieved after AKI recovery.
and 1-year mortality (1.2; 95% CI 1.1Y1.3; Supplemental Table I, http://links.lww.com/TA/A410). In comparison, although RKI patients had lower inpatient mortality compared with patients who did not recover from the initial AKI episode (21% vs. 37%, p = 0.03), censored 1-year mortality rates were similar (31% vs. 26%, p = 0.54). In addition, RKI was associated with longer hospital stay and worse discharge renal function. Secondary analyses of AKI and RKI episodes indicated that recurrence is as severe as the index events (Table 3, Fig. 3). RKI
TABLE 2. Outcomes of Patients With and Without RKI Outcome
AKI Only, n = 148
RKI, n = 68
p
Inpatient mortality, n (%) Overall 1-y mortality, n (%) Postdischarge mortality,* n (%) SICU LOS, median (IQR), d Hospital LOS, median (IQR), d Discharge creatinine, mean (SD), mg/dL
13 (9) 31 (46) 21 (16) 9 (11) 23 (17) 0.86 (0.40)
14 (21) 36 (24) 17 (31) 13 (15) 37 (31) 1.34 (1.01)
0.01 0.002 0.01 0.04 G0.0001 G0.0001
*For patients discharged alive from the hospital. LOS, length of stay.
event severity, duration, and RRTuse were similar to AKI. AKI and RKI events had similar renal recovery rates (73% vs. 65%, p = 0.17) and kidney injuryYassociated mortality (11% vs. 15%, p = 0.41).
DISCUSSION This is an important initial investigation of RKI in a cohort of critically ill patients. The principal findings demonstrate that RKI is common among patients who recover from AKI and is associated with worse outcomes and greater resource use than those who recover from AKI and do not recur. Unfortunately, traditional risk factors failed to predict differences between patients with AKI only and those who develop recurrence. What is not clear from this study is if recurrence reflects the natural history of AKI, complications of caring for critically ill patients, or if RKI is simply another AKI event from a second insult. RKI is a clinically significant phenomenon. RKI is common among critically ill patients, occurring in nearly one third of those who recover from an index AKI event. RKI was as severe as AKI and was associated with additional loss of renal function and episode-related mortality. Given the worse outcomes associated with recurrence, it is likely that morbidity and mortality are cumulative in patients with RKI. It is possible that morbidity and mortality from AKI are ‘‘dose dependent’’ and related to number TABLE 3. Comparison of AKI and RKI Episodes Outcome Peak creatinine, mean (SD), mg/dL RIFLE grade, n (%) Risk Injury Failure RRT use, n (%) RRT days mean (SD) Median kidney injury exposure,* median (IQR), d Renal recovery, n (%) Kidney injury duration,** median (IQR), d Best recovery: baseline creatinine† Death during episode, n (%)
Figure 2. Comparison of index AKI events between patients with and without RKI. Differences between peak creatinine, rates of onset, and event durations were not significant.
AKI Events, n = 296
RKI Events, n = 68
p
2.17 (1.55)
1.96 (1.48)
0.31
155 (52) 69 (23) 76 (26) 38 (13) 11 (12) 4 (2Y10)
41 (60) 17 (25) 12 (18) 7 (10) 9 (5) 4 (2Y12)
0.24 0.77 0.16 0.57 0.66 0.87
216 (73) 3 (2Y6)
44 (65) 4 (2Y9)
0.17 0.10
0.9 (0.2) 33 (11)
1.0 (0.3) 10 (15)
0.16 0.41
*Time with kidney injury for all affected patients. **Time from event onset to recovery. †Lowest creatinine achieved after AKI recovery.
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
1399
J Trauma Acute Care Surg Volume 76, Number 6
Harris et al.
Figure 3. Comparison of AKI and RKI. Differences between peak creatinine, rates of onset, and event durations were not significant.
and duration of episodes. This should be evaluated by future studies. Although this study did not evaluate the effect of recurrence on long-term renal function, RKI patients had higher discharge creatinine values, reflecting at least short term persisting renal dysfunction. The current findings are consistent with previous studies of repetitive kidney injury. The incidence of RKI among at-risk patients in this cohort was similar to the rate of multiple events reported by Thakar et al.16 in a long-term study of diabetic patients. Furthermore, our results showing that RKI contributes to increased mortality, resource use, and persistent renal dysfunction are consistent with studies showing progressively worse outcomes with greater exposure to kidney injury.4Y6,15,16,34 In this cohort, RKI events were as severe as AKI, resulting in greater cumulative exposure to renal injury and dysfunction, which may account for the significantly worse outcomes associated with recurrence. Although these findings indicate that RKI is clinically important, long-term renal outcomes are lacking. Traditional kidney injury risk factors did not predict recurrence, and only the presence of AKI on hospital admission and higher recovery creatinine values were associated with RKI. Although the latter difference was relatively small, because slight changes in creatinine are associated with worse outcomes,13,17,34,35 this finding may reflect a minor but clinically significant difference in the effect of AKI recovery on recurrence risk. However, AKI event severities were not significantly different between the AKI and RKI groups. As such, among patients who recover from AKI, those at risk for recurrence may not be readily identifiable. One potential explanation is that patients with greater risk for RKI, such as more comorbidities or higher illness severity, do not typically recover and therefore are not exposed to risk of recurrence. Other variables not taken into account in this study may identify patients at risk for RKI development. Few studies have investigated recurrent or repetitive kidney injury, and to our knowledge, this is the first to describe RKI during the same hospital stay in either surgical or critically ill patients. Given the association between RKI and worse renal and patient outcomes, recurrence may be an important outcome for future AKI studies to measure. Similarly, since kidney injury is largely preventable and therapy supportive to limit further 1400
renal insults,20Y22,36 RKI may be a useful clinical, research, or quality marker of the efficacy of AKI management. This study has several important limitations. Because surgical patients have unique clinical conditions and risk factors for kidney injury19,37 and perhaps increased risk for recurrence from perioperative care, these findings may not be representative of RKI in other patient populations. In classifying kidney injury as a singular process, we may have missed differences among common etiologies such as sepsis, malperfusion, and nephrotoxic agents that may entail different risks for recurrence or may be responsible for the recurrence. Future work will assess these variables. Although the use of creatinine alone to define and characterize kidney injury events facilitated precise study of a large cohort, it may not have been accurate for patients who met RIFLE criteria by urine output but not creatinine criteria. The limited availability of postdischarge creatinine values prevented us from analyzing the effect of RKI on long-term renal function. However, by accurately assessing 1-year mortality using the Social Security Death Index, this served as a clinically meaningful long-term outcome. Finally, as with all retrospective database studies, there are concerns for observation bias. Despite these weaknesses, the study does provide an important and novel evaluation of RKI in critically ill surgical patients. In conclusion, this novel investigation of RKI gives important early insight into an important complication among patients who recover from AKI. RKI is common among critically ill patients, affecting approximately one third of those at risk and is associated with increased mortality, greater renal dysfunction, and higher resource use. The presence of AKI in progress at admission is associated with greater risk for recurrence; these patients should be recognized as being at increased risk for worse renal and overall outcomes and managed accordingly. Future studies should be aimed at better defining modifiable risk factors for recurrence and evaluate the long-term effect of RKI on renal function. AUTHORSHIP D.G.H., J.J.D., and M.E.L. conceived and planned the study. D.G.H., G.K., M.P.M., and M.E.L. performed the data collection and analysis. D.G.H., T.M.S., W.C.C., J.J.D., and M.E.L. contributed to the data interpretation. D.G.H., G.K., M.P.M., and M.E.L. drafted the manuscript, and T.M.S., W.C.C., and J.J.D. provided critical review.
DISCLOSURE The authors declare no conflicts of interest.
REFERENCES 1. Gaieski DF, Edwards JM, Kallan MJ, Carr BG. Benchmarking the incidence and mortality of severe sepsis in the United States. Crit Care Med. 2013;41:1167Y1174. 2. Vincent J, De Mendonc¸a A, Cantraine F, et al. Use of the SOFA score to assess the incidence of organ dysfunction/failure in intensive care units: results of a multicenter, prospective study. Crit Care Med. 1998;26: 1793Y1800. 3. Martin GS, Mannino DM, Eaton S, Moss M. The epidemiology of sepsis in the United States from 1979 through 2000. N Engl J Med. 2003;348: 1546Y1554. 4. Hoste EAJ, 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:R73.
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
J Trauma Acute Care Surg Volume 76, Number 6
5. Cruz DN, Bolgan I, Perazella MA, et al. North East Italian prospective hospital renal outcome survey on acute kidney injury (NEiPHROS-AKI): targeting the problem with the RIFLE criteria. Clin J Am Soc Nephrol. 2007;2:418Y425. 6. Hoste EAJ, Schurgers M. Epidemiology of acute kidney injury: how big is the problem? Crit Care Med. 2008;36:S146. 7. Schiffl H. Renal recovery from acute tubular necrosis requiring renal replacement therapy: a prospective study in critically ill patients. Nephrol Dial Transplant. 2006;21:1248Y1252. 8. Wald R, Quinn RR, Luo J, et al. Chronic dialysis and death among survivors of acute kidney injury requiring dialysis. JAMA. 2009;302:1179Y1185. 9. Ponte B, Felipe C, Muriel A, Tenorio MT, Lian˜o F. Long-term functional evolution after an acute kidney injury: a 10-year study. Nephrol Dial Transplant. 2008;23:3859Y3866. 10. Metnitz PG, Krenn CG, Steltzer H, et al. Effect of acute renal failure requiring renal replacement therapy on outcome in critically ill patients*. Crit Care Med. 2002;30:2051Y2058. 11. Hobson CE, Yavas S, Segal MS, et al. Acute kidney injury is associated with increased long-term mortality after cardiothoracic surgery. Circulation. 2009;119:2444Y2453. 12. Bihorac A, Baslanti TO, Cuenca AG, et al. Acute kidney injury is associated with early cytokine changes after trauma. J Trauma Acute Care Surg. 2013;74:1005Y1013. 13. Goldberg A, Kogan E, Hammerman H, Markiewicz W, Aronson D. The impact of transient and persistent acute kidney injury on long-term outcomes after acute myocardial infarction. Kidney Int. 2009;76:900Y906. 14. Lo LJ, Go AS, Chertow GM, et al. Dialysis-requiring acute renal failure increases the risk of progressive chronic kidney disease. Kidney Int. 2009;76:893Y899. 15. Schiffl H, Fischer R. Five-year outcomes of severe acute kidney injury requiring renal replacement therapy. Nephrol Dial Transplant. 2008;23: 2235Y2241. 16. Thakar CV, Christianson A, Himmelfarb J, Leonard AC. Acute kidney injury episodes and chronic kidney disease risk in diabetes mellitus. Clin J Am Soc Nephrol. 2011;6:2567Y2572. 17. Goldenberg I, Chonchol M, Guetta V. Reversible acute kidney injury following contrast exposure and the risk of long-term mortality. Am J Nephrol. 2008;29:136Y144. 18. Wohlauer MV, Sauaia A, Moore EE, Burlew CC, Banerjee A, Johnson J. Acute kidney injury and posttrauma multiple organ failure: the canary in the coal mine. J Trauma Acute Care Surg. 2012;72:373Y380. 19. White LE, Hassoun HT, Bihorac A, et al. Acute kidney injury is surprisingly common and a powerful predictor of mortality in surgical sepsis. J Trauma Acute Care Surg. 2013;75:432Y438. 20. Bellomo R, Kellum JA, Ronco C. Acute kidney injury. Lancet. 2012; 380:756Y766. 21. Prowle JR, Echeverri JE, Ligabo EV, Ronco C, Bellomo R. Fluid balance and acute kidney injury. Nat Rev Nephrol. 2009;6:107Y115.
Harris et al.
22. Tolwani A. Continuous renal-replacement therapy for acute kidney injury. N Engl J Med. 2012;367:2505Y2514. 23. Portella VG, Silva-Filho JL, Landgraf SS, et al. Sepsis-surviving mice are more susceptible to a secondary kidney insult*. Crit Care Med. 2013;41:1056Y1068. 24. Kelleher SP, Robinette JB, Miller F, Conger JD. Effect of hemorrhagic reduction in blood pressure on recovery from acute renal failure. Kidney Int. 1987;31:725Y730. 25. Conger J. Hemodynamic factors in acute renal failure. Adv Ren Replace Ther. 1997;4:25. 26. Brinkman S, de Jonge E, Abu-Hanna A, Arbous MS, de Lange DW, de Keizer NF. Mortality after hospital discharge in ICU patients*. Crit Care Med. 2013;41:1229Y1236. 27. Lissauer ME, Galvagno SM Jr, Rock P, et al. Increased ICU resource needs for an academic emergency general surgery service. Crit Care Med. 2014;42:910Y917. 28. Knaus WA, Wagner DP, Zimmerman JE, Draper EA. Variations in mortality and length of stay in intensive care units. Ann Intern Med. 1993;118: 753Y761. 29. Knaus WA, Wagner D, Draper EA, et al. The APACHE III prognostic system. Risk prediction of hospital mortality for critically ill hospitalized adults. CHEST. 1991;100:1619Y1636. 30. Zimmerman JE, Kramer AA, McNair DS, Malila FM. Acute Physiology and Chronic Health Evaluation (APACHE) IV: hospital mortality assessment for today’s critically ill patients*. Crit Care Med. 2006;34:1297Y1310. 31. Zimmerman JE, Wagner DP, Draper EA, Wright L, Alzola C, Knaus WA. Evaluation of acute physiology and chronic health evaluation III predictions of hospital mortality in an independent database. Crit Care Med. 1998;26:1317Y1326. 32. Levey AS, Coresh J, Balk E, et al. National Kidney Foundation practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Ann Intern Med. 2003;139:137Y147. 33. Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P. Acute renal failure-definition, outcome measures, animal models, fluid therapy and information technology needs: the Second International Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group. Crit Care. 2004;8:R204YR212. 34. Chertow GM, Burdick E, Honour M, Bonventre JV, Bates DW. Acute kidney injury, mortality, length of stay, and costs in hospitalized patients. J Am Soc Nephrol. 2005;16:3365Y3370. 35. Lassnigg A, Schmidlin D, Mouhieddine M, et al. Minimal changes of serum creatinine predict prognosis in patients after cardiothoracic surgery: a prospective cohort study. J Am Soc Nephrol. 2004;15:1597Y1605. 36. Li PKT, Burdmann EA, Mehta RL. Acute kidney injury: global health alert. Kidney Int. 2013;83:372Y376. 37. Abelha F, Botelho M, Fernandes V, Barros H. Determinants of postoperative acute kidney injury. Crit Care. 2009;13:R79.
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
1401
