Abstract
Acute renal failure (ARF) is a common complication in hospitalized patients, but little is known about the epidemiology of ARF in China. In this study, we performed a prospective examination of the cause, prognosis, and risk factors associated with ARF at a hospital in Shanghai, China. We considered all ARF patients who were admitted to our hospital from December 2003 to December 2006. Among the 320 ARF patients, 135 (42.2%) were over the age of 60. Sepsis, heart failure, and nephrotoxic drug use were the leading causes of ARF. The overall mortality rate was 31.9%, and mortality rate was significantly higher among the elderly. Logistic regression indicated that heart failure, respiratory failure, and malignant cancer were risk factors independently associated with poor prognosis. In this Shanghai hospital, there was a high incidence and mortality rate of patients hospitalized with ARF. The prognosis of patients who underwent renal replacement therapy was better than those who were treated more conservatively.
INTRODUCTION
Acute renal failure (ARF) is a serious medical condition that results from damage of the kidneys. At the clinical level, it is characterized by a sudden and sustained decline in glomerular filtration rate and an abrupt and sustained rise in serum creatinine (Scr) and blood urea nitrogen (BUN). ARF may also be accompanied oliguria or anuria, metabolic disturbances, and body fluid imbalance. Despite continuing advances in the medical treatment of kidney diseases and ARF, ARF-related morbidity and mortality rates remain high.
Previous studies have shown that ARF accounts for 1% of all hospital admissions, is a complication in more than 7% of all admissions, and typically occurs in patients with underlying chronic kidney disease.[Citation1,Citation[2]] For patients with ARF that is severe enough to require dialysis, the mortality rate is about 50%.[Citation3–5] Among patients admitted to intensive care units (ICUs), those with ARF have a two- to three-fold greater mortality rate. The reasons ARF may be so common is likely to include factors such as the increasing number of elderly patients, increased prevalence of chronic kidney disease (CKD), and improved survival rates for patients with ischemic heart disease and diabetes mellitus.[Citation6,Citation[7]]
There are only a few studies of ARF in patients at Chinese hospitals, and most of these studies have been retrospective.[Citation8–10] We performed a prospective observational study for which the primary objective was to determine the incidence rate of ARF and the characteristics of patients with ARF. The secondary objectives were to identify risk factors and describe the short-term clinical outcomes of patients hospitalized with ARF. Information on the overall incidence, risk factors, and patient outcomes will be useful in the development of preventive and therapeutic strategies for treatment of Chinese and other patients with ARF.
SUBJECTS AND METHODS
Study Population
This was a prospective observational study conducted at the Sixth People's Hospital (Shanghai Jiao Tong University, Shanghai, China), a tertiary teaching hospital with 1,600 beds. From December 2003 to December 2006, there were 108,744 patients admitted to this hospital, 320 of whom were enrolled in this study. We included patients whose Scr was 176 to 264 μmol/L, who stayed in the hospital for at least 48 h, and who were at least 16 years old. We excluded patients with end-stage renal failure, which was diagnosed if the patient was on hemodialysis (HD) or peritoneal dialysis (PD) for more than one month. For the renal failure to be considered “acute,” it must have developed within 72 h. For patients with more than one admission during the study period, only the first admission was registered. This study was approved by the ethics committee of the Shanghai Jiao Tong University.
ARF was defined as a sudden rise in Scr to more than 176 μmol/L in patients with previously normal renal function. For patients admitted to the hospital with a history of chronic renal failure (Scr<264 μmol/L), ARF was defined as an increase of at least 50% from the Scr that was obtained previously.[Citation11] For diagnosis of acute on chronic kidney disease (A-on-C), we used previously described criteria.[Citation12,Citation[13]] Multiple organ dysfunction syndrome (MODS) was diagnosed if there was dysfunction of two or more organ systems.[Citation14]
Baseline data collected for all patients admitted to our hospital include age, gender, and diagnosis. The following clinical data were collected in all of our 320 patients: hospital department of admission, date of admission and discharge, primary disease, baseline Scr, peak Scr, date of ARF onset, date of renal recovery, etiology, outcome, and treatment. We also recorded the presence of pre-existing kidney disease, diabetes mellitus, use of nephrotoxic drugs, surgery before development of ARF, hypotension, use of a contrast agent, oliguria, and other complications (such as organ failure). We performed follow-ups until patients were discharged or died.
Trained research nurses and fellows collected data daily in specially designed case report forms. To avoid the inadvertent exclusion of cases, we performed a diagnosis certification process at the end of the study. This certification process confirmed that no cases had been missed.
Methods
We classified each case of ARF according to cause (i.e., pre-renal, intra-renal [including tubular-interstitial disease and glomerular disease], or post-renal), patient age (i.e., less than 60 years old, 60–79 years old, and more than 80 years old), department of hospital stay (i.e., internal medicine group or surgical group), and renal function at time of discharge. The classes of renal function at discharge were complete recovery (i.e., patients who ceased dialysis, who no longer had clinical symptoms and signs, whose renal function returned to normal, or whose Scr returned to the initial leve; before ARF was diagnosed), improved (i.e., patients who were dependent on renal replacement therapy, whose renal function did not return to normal, or whose Scr returned to the level before ARF was diagnosed), and deceased (patients who died in the hospital). The incidence rate of ARF was defined as the number of ARF episodes per 100 cases.
Statistical Analysis
Statistical analysis was performed using SAS version 6.12 (SAS Institute Inc., Cary, North Carolina, USA). Quantitative variables are reported as means ± SDs and qualitative variables as percentages. For the univariate analysis of risk factors, we used the Student's t-test or the Wilcoxon test for the quantitative variables and Pearson's χ2 test for the qualitative variables. For logistic regression, we used the forward selection method for multivariate analysis of risk factors, with the level of significance of p < 0.05.
RESULTS
From December 2003 to December 2006, there were 108,744 patients admitted to the Sixth People's Hospital, 320 of whom (0.294%) had ARF. shows the demographic and clinical features of these 320 patients. Mean age was 56.6 ± 18.1 yrs (range, 16 to 97 yrs) and 62% were male. There were 96, 105, and 119 cases admitted in 2004, 2005, and 2006 respectively; this increase over time was not significant (p > 0.05). The mean length of hospital stay was 23.8 ( ± 20.5) days. The baseline and peak levels of Scr were 346.2 μmol/L ( ± 209.5) and 491.3 μmol/L (± 332.7), respectively (see ).
Table 1 Clinical characteristics of ARF in hospitalized patients (n = 320)
A total of 125 patients had pre-renal ARF, with heart failure (22.4%), fluid deletion (21.6%), bleeding (20%), liver cirrhosis (8%), and nephrotic syndrome (5.6%) as the major causes. There were 157 patients with intra-renal ARF, with glomerular-vascular diseases (19.7%) and tubular-interstitial diseases (80.3%) as the major causes. Among patients with intra-renal ARF, this condition was drug-induced in 25 patients, associated with infections in 39 patients, and related to cancer in 15 patients. There were 38 patients with post-renal ARF, with urolithiasis, prostate diseases and cancer as the major causes.
There were 171 patients in the “internal medicine group,” with 22.8% of these cases caused by infectious diseases, 18.1% caused by glomerular-vascular disease, 16.4% caused by heart failure, and 14.6% induced by drugs. There were 149 patients in the “surgical group”, 44.3% of whom had various types of postoperative ARF and 25.5% of whom had obstructive uropathy.
A total of 185 patients were less than 60 years old, 135 patients were 60–79 years old, and 40 patients were 80 or more years old. There were significant differences in etiology among the age groups (see ).
Table 2 Cause among different age groups
During the three years of our study, 102 of the 320 ARF patients died, corresponding to an overall mortality rate of 31.9%. There were 117 patients in the “improved” group (including 16 self-discharge cases) and 101 patients in the “complete recovery” group. Among the 102 patients who died, 42 (33.6%) had pre-renal ARF, 53 (33.8%) had intra-renal ARF, and 7 (18.4%) had post-renal ARF. These differences in mortality were not significant (p = 0.173). However, there was a significant difference in mortality among the three age groups (p = 0.022; see ).
Figure 1. Comparison of mortality among different age groups.
There were 145 patients with “simple” ARF, and the mortality rate in this group was 14.5% (21/145). The average length of hospital stay in simple ARF group was significantly less than that of the MODS group [21.6 (± 19.3) vs. 25.7 (± 24.5) days, p = 0.027)]. Moreover, in the MODS group, the mortality rate increased as the number of organ failures increased (p < 0.01; see ).
Table 3 The relationship between MODS and prognosis
There were 95 patients in the renal replacement therapy (RRT) group, and the mortality rate in this group was 23.2% (22/95). There were 225 patients in the conservative treatment group (non-RRT group), and the mortality in this group was 35.6% (80/225). There was significant difference in mortality between these two groups (p = 0.031). Patients in the RRT group included those on intermittent hemodialysis (IHD), slow low-efficiency hemodialysis (SLED), continuous venovenous hemofiltration (CVVH) and continuous venovenous hemofiltration dialysis (CVVHD), and plasma exchange.
Among the 75 patients with acute-on-chronic kidney disease (A-on-C), accounting for 23.4% of all ARF patients, 45 were older than 60 years. The main causes of A-on-C were sepsis (25 cases), drugs (19 cases), and congestive heart failure (10 cases). A total of 42 patients had pre-existing chronic renal failure (basal creatinine, 132–264 μmol/L). Among our A-on-C patients, 25 experienced “complete recovery,” 28 attained the status of “improved” (7 of whom needed RRT), and 22 patients died. The mortality rate in this group was 29.3%. A total of 19 of the 42 patients with pre-existing chronic renal failure died, corresponding to a much higher mortality rate than those with normal renal function (p = 0.001). Of the 75 A-on-C patients, 39 patients experienced oliguria, 16 of whom died. The mortality rate was significantly higher for patients with oliguria than those without oliguria (p = 0.021). A total of 18 elderly patients died, corresponding to a higher mortality rate than the youth group (p = 0.013).
There were 116 patients with hospital-acquired acute renal failure (HA-ARF), corresponding to 36.3% of all ARF cases. Among these 116 patients, we classified 45 (38.8%) as drug-induced, 66 (56.9%) as postoperative ARF, and 5 (4.3%) as contrast agent nephropathy. Nearly half of the nephrotoxic drugs were antibiotics (e.g., cephalosporins and aminoglycosides, especially cefradine). Most cases of postoperative ARF were caused by cardiovascular surgery (25.8%). Among all 116 HA-ARF patients, 39 died (mortality rate, 33.6%), 42 had complete recovery, and 35 improved. compares the risk factors between patients who died and those who survived.
Table 4 The comparison of risk factors of patients with HA-ARF between death and survival groups
In univariate analysis, we found that hypotension, oliguria, respiratory failure, mechanical ventilation, heart failure, hepatic failure, cancer, high catabolism, and pre-existing renal failure were significantly associated with the poor prognosis of patients with ARF. Logistic regression analysis showed that respiratory failure, heart failure, and malignant cancer were the significant independent risk factors (see ).
Table 5 The comparison of factors of patients with ARF between death and survival groups
DISCUSSION
ARF is a common and serious clinical problem, and almost all medical departments must deal with patients who have ARF. The development of ARF in the hospital is associated with increased mortality, increased length of hospital stay, and increased use of healthcare resources and costs.[Citation15]
In this prospective study, we comprehensively characterized patients who were hospitalized with ARF at a hospital in Shanghai, China. Over a three-year period, the overall incidence of ARF was 0.294%, lower than that reported in other studies.[Citation16,Citation[17]] This difference may be explained by differences in how ARF is defined. The very strict definition that we used may have underestimated number of cases. A consensus definition of ARF has been proposed by the working group Acute Dialysis Quality Initiative. This classification, known as RIFLE (Risk, Injury, Failure, Loss, and End-Stage Renal Failure), considers the different stages of renal failure according to Scr and urine output.[Citation18] This classification considers the different levels of renal failure to correspond to different clinical outcomes. This classification was not yet available when our study was conducted. The different ARF incidence rates may also be explained by population differences. In our study, we did not include children. There was a trend for an increased incidence of ARF over the three years of this study, although this difference was not statistically significant.
This study showed that sepsis, heart failure, use of nephrotoxic drugs, cancer, and surgery were the leading causes of ARF. Among patients in the internal medicine group, acute tubular necrosis was induced by sepsis, and nephrotoxic drug use was the most common causes of ARF; by contrast, post-operative ARF was the majority in the surgical group.
Among our patients, intra-renal failure was the most common ARF classification, with pre-renal and post-renal ARF being less common. We also found that intra-renal ARF declined with increasing age, while post-renal and pre-renal ARF increased with increasing age, although none of these changes was statistically significant. It should also be noted that the mortality of patients with post-renal ARF was lower (18.4%). If a renal obstruction is corrected in a timely manner, renal function might return to normal for most patients.
Although RRT use has increased in recent decades, ARF-related mortality rate has not decreased, and remains in the range of 30 to 80%.[Citation19] This might be because current cases of ARF have more complex etiologies or are due to more serious illnesses, making treatment more difficult.[Citation19] Elderly patients have a higher risk of developing ARF.[Citation20,Citation[21]] In our study, the proportion of elderly patients with ARF was 42.2%, higher than that reported by other studies.[Citation9,Citation[22]] As expected, we found that advanced age was a risk factor for development of ARF. The overall ARF-related mortality rate was 31.9%, but was 50% for patients older than 80 years. We also found that the length of hospital stay for patients with MODS ARF was longer than those with non-MODS ARF, and that among MODS patients, the more organs that failed, the worse the prognosis. These results are similar to those of previous research.[Citation23] The mortality of patients who received conservative treatment was 35.6%, while that of patients who received RRT was 23.2%. Thus, early initiation of dialysis seems important for patients with severe ARF, especially the elderly.
The study about A-on-C was relatively few in the hospital patients. In the PICARD cohort study, one-third of patients had CKD stage IV or above.[Citation24] Similarly, a study from a hospital-affiliated Peking University showed that 35.5% of patients with ARF had CKD.[Citation12] In our study, A‐on-C patients accounted for 23.4% of all ARF cases, and the main causes were sepsis, nephrotoxic drug use, and heart failure. Among our A-on-C patients, 42 had preexisting renal dysfunction, suggesting that this condition may be a risk factor.[Citation25] Unexpectedly, we found that the mortality rate from A-on-C was 29.3%, lower than the overall ARF mortality rate (31.9%), although this was not statistically significant. The reasons for this seemingly paradoxical finding may include confounding by malnutrition (and lower SCr values due to lower muscle mass), and unrecorded differences in disease severity among persons with and without CKD who subsequently developed ARF. In other words, a slight kidney injury or the presence of fewer associated complications may cause ARF in patients with underlying CKD.[Citation26] In a population-based surveillance study of ARF, 63% of all patients with ARF who required dialysis one year following admission had preexisting CKD.[Citation27]
Hospital-acquired acute renal failure (HA-ARF) is considered an iatrogenic disorder. Some researchers reported that HA-ARF has increased in recent years.[Citation2] We found that 36.3% of our ARF patients had HA-ARF. Several recent large epidemiological studies showed that nephrotoxic drugs contributed to 19–25% of all cases of severe ARF among critically ill patients.[Citation5,Citation[28]] In our study, cephalosporins (rather than aminoglycosides) were the most commonly administered nephrotoxic drugs. Surgery may be another important iatrogenic cause of HA-ARF. Our data showed that 66 of 320 cases (20.6%) of ARF were postoperative. The incidence and prognosis of ARF differ for patients who receive different types of surgery. Patients who underwent cardiovascular surgery have higher incidence of ARF and face a poorer prognosis from ARF.[Citation29] With the increasing use of contrast agents, contrast-induced nephropathy has also become a significant cause of HA-ARF. Previous research[Citation30,Citation[31]] reported that pre-existing renal dysfunction, diabetes, volume-depletion, types and dosage of contrast agent, advanced age, congestive heart failure (NYHA class III and IV), hepatic cirrhosis, multiple myeloma, and history of metformin use were all risk factors associated with contrast-induced nephropathy.
In this study, we analyzed 20 risk factors associated with ARF. Univariate analysis found that hypotension, respiratory failure, use of mechanical ventilation, heart failure, hepatic failure, oliguria, high catabolism, and pre-existing renal dysfunction were all associated with mortality. This is similar to a previous analyses of conditions associated with ARF.[Citation32] Analysis by logistic regression indicated that respiratory failure, heart failure, and cancer were independent risk factors associated with mortality. It is well-known that cancer patients are prone to renal dysfunction due to their disease or treatment. For example, Lavilla et al.[Citation33] found that the Multiorgan Dysfunction Index (MFI) is a useful tool to assess ARF-specific mortality risk in cancer patients with ARF, and that a low Karnofsky Performance Status Score (KPSS) is associated with greater tubular damage and worse prognosis, including higher mortality and greater risk of multiple organ failure. Therefore, adequate preventive measures, early diagnosis, and careful management are crucial for the treatment of cancer patients with ARF.
This study has several limitations. First, the definition of ARF that we used did not rely on multiple renal function variables, such as urine output. Second, this is a single-center study. Third, this study only included inpatients and did not include patients from the emergency room or outpatients. Fourth, we were only able to assess short-term outcomes because the duration of patient follow-up (i.e., three years) was somewhat short. However, this study also has several strengths. First, it is a prospective study that was conducted on all ARF patients admitted to our hospital within a defined timeframe. Second, it is much more comprehensive than most other studies of ARF in China. Finally, our study provides recent epidemiologic data on ARF that considers the entire hospital, not just the ICU.
In summary, ARF is a common and critical medical condition. It is often necessary for medical staff of various departments of a hospital to work together to prevent the onset of ARF. For critically ill patients, it is important to initiate RRT in a timely manner, especially for patients with MODS. This study will help to guide the development and implementation of treatments that can decrease the incidence of ARF and improve the prognosis of patients who have ARF.
DECLARATION OF INTEREST
The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
ACKNOWLEDGMENTS
This study was a key topic of Shanghai Health Bureau (2003ZD01) and was supported by a grant from the Shanghai Scientific and Natural Committee (No. 05ZR14086). We thank the staff of the Medical Records Room and Biochemical Room for their assistance, and we thank our patients and their families for their participation.
Related Research Data
REFERENCES
- Kaufman J, Dhakal M, Patel B, et al. Community-acquired acute renal failure. Am J Kidney Dis. 1991; 17: 191–198
- Nash K, Hafeez A, Hou S. Hospital-acquired renal insufficiency. Am J Kidney Dis. 2002; 39: 930–936
- Clermont G, Acker CG, Angus DC, et al. Renal failure in the ICU: Comparison of the impact of acute renal failure and end-stage renal disease on ICU outcomes. Kidney Int. 2002; 62: 986–996
- 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: 2051–2058
- Mehta RL, Pascual MT, Soroko S, et al. Spectrum of acute renal failure in the intensive care unit: The PICARD experience. Kidney Int. 2004; 66: 1613–1621
- Muntner P, Coresh J, Powe NR, et al. The contribution of increased diabetes prevalence and improved myocardial infarction and stroke survival to the increase in treated end-stage renal disease. J Am Soc Nephrol. 2003; 14: 1568–1577
- Ronco C, Zanella M, Brendolan A, et al. Management of severe acute renal failure in critically ill patients: An international survey in 345 centres. Nephrol Dial Transplant. 2001; 16: 230–237
- Yue W, Zhuan C, Minhua F. Retrospective analysis on Chinese patients diagnosed with acute renal failure hospitalized during the last decade (1994–2003). Am J Nephrol. 2005; 25: 514–519
- Zhang T, Mei CL. Clinical study of 100 cases with acute renal failure. Chin J Nephrol. 2005; 12: 732–736
- Yue W, Zhuan C, Minhua F, et al. Hospital-acquired and community-acquired acute renal failure in hospitalized Chinese: A ten-year review. Ren Fail. 2007; 29: 163–168
- Liano F, Pascual J. Epidemiology of acute renal failure: A prospective, multi-center, community-based study. Madrid Acute Renal Failure Study Group. Kidney Int. 1996; 50: 811–818
- Zhang L, Wang M, Wang H. Acute renal failure in chronic kidney disease—clinical and pathological analysis of 104 cases. Chin J Nephrol. 2005; 63: 346–350
- Brezis M, Rosen S, Epstein FH, . Acute renal failure. The kidney4th, BM Brenner, FC Rector, et al. Saunders, Philadelphia 1991; 993–1061
- Bone RC, Balk RA, Cerra FB. The ACCP/SCCM Consensus Conference Committee. Definition for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Chest. 1992; 101(6)1644–1655
- Chertow GM, Burdick E, Honour M, et al. Acute kidney injury, mortality, length of stay, and costs in hospitalized patients. J Am Soc Nephrol. 2005; 16: 3365–3370
- Uchino S, Bellomo R, Goldsmith D, et al. An assessment of the RIFLE criteria for acute renal failure in hospitalized patients. Crit Care Med. 2006; 34: 1913–1917
- Liangos O, Wald R, O'Bell J, et al. Epidemiology and outcomes of acute renal failure in hospitalized patients: A national survey. Clin J Am Soc Nephrol. 2006; 1: 43–51
- Ronco C, Kellum JA, Mehta R. The Acute Dialysis Quality Initiative: The New York conference. Adv Ren Replace Ther. 2002; 9: 248–251
- Chertow GM. On the design and analysis of multicenter trials in acute renal failure. Am J Kidney Dis. 1997; 30: S96–S101
- de Mendonca A, Vincent JL, Suter PM, et al. Acute renal failure in the ICU: Risk factors and outcome evaluated by the SOFA score. Intensive Care Med. 2000; 26: 915–921
- Brivet FG, Kleinknecht DJ, Loirat P, et al. Acute renal failure in intensive care units—causes, outcome, and prognostic factors of hospital mortality: A prospective, multicenter study. French Study Group on Acute Renal Failure. Crit Care Med. 1996; 24: 192–198
- Zhang W, Chen N, Ren H, et al. The epidemiologic investigation in patients with acute renal failure. Nephrol Dial Transplant. 2002; 11: 323–327, [in Chinese]
- Schwilk B, Wiedeck H, Stein B, et al. Epidemiology of acute renal failure and outcome of haemodiafiltration in intensive care. Intensive Care Med. 1997; 23: 1204–1211
- Chertow GM, Soroko SH, Paganini EP, et al. Mortality after acute renal failure: Models for prognostic stratification and risk adjustment. Kidney Int. 2006; 70: 1120–1126
- Bisenbach G, Janko O, Schmekal B. Etiology and prognosis of “internal medicine”acute renal failure in 1981-1990 and 1991-2000—an analysis of cases in a single center. Wien Med Wochenschr. 2003; 153: 530–533
- Waikar SS, Liu KD, Chertow GM, et al. The incidence and prognostic significance of acute kidney injury. Curr Opin Nephrology. 2007; 16(3)227–236
- Bagshaw SM, Laupland KB, Doig CJ, et al. Prognosis for long-term survival and renal recovery in critically ill patients with severe acute renal failure: A population-based study. Crit Care. 2005; 9: R700–R709
- Uchino S, Kellum JA, Bellomo R, et al. Acute renal failure in critically ill patients: A multinational, multicenter study. JAMA. 2005; 17: 813–818
- Suen WS, Mok CK, Chiu SW, et al. Risk factors for development of acute renal failure (ARF) requiring dialysis in patients undergoing cardiac surgery. Angiology. 1998; 49: 789–800
- Asif A, Preston RA, Roth D. Adiocontrast-induced nephropathy. Am J Therapeutics. 2003; 10: 137–147
- Lindholt JS. Radiocontrast-induced nephropathy. Eur J Vasc Edovasc Surg. 2003; 25: 296–304
- Yang XN, Zhang X, Hou FF, et al. Analysis of risk factors that influence in-hospital mortality and renal outcome in patients with acute renal failure. Chin J Nephrol. 2001; 17: 287–289
- Lavilla J, Perez-Gracia JL, Gurpide A, et al. Prospective assessment and validation of renal prognostic scores in oncology patients with acute renal failure. ASCO Annual Meeting Proceedings (Post-Meeting Edition). J Clin Oncol. 2004; 22(14s): 8098