ORGAN SYSTEM FAILURES PREDICTION MODEL IN INTENSIVE CARE PATIENTS WITH ACUTE RENAL FAILURE TREATED WITH DIALYSIS

Abstract

Objective: To evaluate the organ system failures hospital mortality predictions in critically ill patients with acute renal failure requiring dialysis. Design: Prospective, cohort study. Setting: Intensive care units in a tertiary care university hospital in Taiwan. Patients: A total of 112 patients admitted to the intensive care units with acute renal failure who required dialysis from January 1999 through December 1999. Interventions: Collection of information necessary to compute the number of failed organs. Measurements and results: Of the 112 patients studied, 75 were men and 37 were women. The mean age of survivors and non-survivors was 58.59 ± 19.91 years and 58.76 ± 19.62 years. The overall mortality rate was 67%. There were no significant differences between survivors and non-survivors in terms of age, gender, or indication for dialysis. The cause of death in the majority of patients was related to organ system failure during the 24 hours immediately preceding the initiation of acute hemodialysis, and carry mortality rates exceeding 83% with the coexistence of four or more failed organs. The area under the organ system failures prediction model receiver operating characteristic curve equaled 0.772 ± 0.046. Conclusion: We conclude that mortality rate for acute renal failure in intensive care unit patients continues to be high. Organ system failures prediction model performed well and simple in its ability to identify patients who die in hospital. Mortality rate increases as number of failed organ increases.

• Acute renal failure
• Critical care
• Multiple organ system failure
• Prognosis

INTRODUCTION

It is well established that acute renal failure in the critically ill patient carries a high mortality rate despite advancing technology. This may be because patients are older, have a greater number of premorbid conditions, or more frequently develop multi-system organ failure. Moreover, patients with acute renal failure typically have a complicated hospital course because of the frequent development of complications in other organ systems [[1]].

The European Dialysis and Transplant Association [[2]] reported an average mortality of 60% for patients dialyzed in intensive care units. Patients requiring dialysis in an intensive care unit have even higher mortality than patients dialyzed in a renal unit [[3]]. Because acute renal failure is a heterogeneous syndrome, occurring in a wide range of patients with diverse disease etiologies, predicting outcome after the onset of acute renal failure is difficult. Several studies [4-6] have shown that the number of organ system failures correlates with mortality in intensive care unit patients.

We therefore prospectively studied the mortality rate in intensive care unit patients who developed acute renal failure requiring dialysis and identified the number of failed organs, present at the time the first patient was first dialized, which is predictive of patient survival.

PATIENTS AND METHODS

Patient Information and Treatment

The registry of all patients with acute renal failure who required dialysis in the medical, general surgical, cardiac, cardiac surgical, trauma, and bone marrow transplant intensive care units of Chang Gung memorial hospital in a 1-year period (January 1999 - December 1999) were reviewed. One hundred and twelve patients were identified. A prospective analysis of the available clinical and laboratory data was performed. Sex, age, the number of organ system failures on the first day of dialysis, and outcome were recorded.

The indications for dialysis included: volume overload with pulmonary edema inadequately controlled with diuretic therapy, hyperkalemia refractory to conservative measures, the need for hyperalimentation with insufficient urinary output, or a sign or symptom, such as encephalopathy, for which uremia could not be ruled out as a precipitating cause. We utilized continuous renal replacement therapy techniques in the hemodynamically unstable patients (systolic blood pressure <90 mmHg at the time of dialysis initiation). There were no strict laboratory criteria employed in the decision to initiate treatment. All cases of acute renal failure requiring dialysis were included, with the exception of acute renal failure after renal transplantation. The study included acute renal failure attributed to acute tubular injury, acute glomerulonephritis, hepatorenal syndrome, and urinary tract obstruction. One hundred and twelve patients received renal replacement therapy: 99 were treated with intermittent hemodialysis (IHD), and 13 with continuous renal replacement therapy (CRRT). All patients who received IHD or CRRT had dialysis angioaccess established by means of double lumen venous catheters placed in a femoral vein. Blood flow through the extracorporeal circuit using standard arterial-venous tubing was maintained via a blood roller pump at a flow rate of 200 mL/min for IHD, and 120 mL/min for CRRT. The patients were dialyzed on a KF-201 dialyzer (polymer coupling) for IHD, and a 0.60 m² polyacrylonitrile hollow-fiber hemofilter (HOSPAL AN69HF) was used for CRRT. A venous pressure monitor and a bubble detector were included in the circuit. A bicarbonate-based dialysate with a sodium concentration of 140 meq/L was used in all patients.

Definitions

The criteria for organ system failure was defined as follows [6-7]: acute renal failure was defined as a serum creatinine level exceeding 3.2 mg/dL or a 2-fold creatinine rise in chronic renal failure, after correcting prerenal causes and mechanical obstruction, or the acute need of renal replacement therapy [8-9]. Chronic renal failure was defined as a known serum creatinine level exceeding 2.3 mg/dL. Criteria for failure of cardiovascular, pulmonary, neurological, hematological, hepatic, and gastrointestinal systems were described previously [9-10] and are summarized in Table 1. For neurological failure, coma due to drug overdose and diabetes mellitus must be excluded. The number of organ system failure included acute renal failure itself.

Table 1. Criteria for Organ System Failure

Organ System	Criteria
Neurological	Glasgow coma scale ≤6 (in absence of sedation at any one point in day)
Cardiovascular	Mean arterial pressure ≤50 mmHg
	Heart rate ≤50 beats/min
	Need for volume loading and/or vasoactive drugs to maintain systolic blood pressure above 100 mmHg
	Occurrence of ventricular tachycardia/fibrillation
	Cardiac arrest
	Acute myocardial infarction
Pulmonary	Respiratory rate ≤5/min or ≤50/min
	Mechanical ventilation for 3 or more days
	Fraction of inspired oxygen (FiO2)>0.4 and/or postive end-expiratory pressure >5 cm H2O
Gastrointestinal	Stress ulcer necessitating transfusion of more than 2 units of blood per 24 h
	Hemorrhagic pancreatitis
	Acalculous cholecystitis
	Necrotizing enterocolitis
	Bowel perforation
Hepatic	Clinical jaundice or total bilrubin level ≥3 mg/dL in the absence of hemolysis
	Serum glutamic-pyruvic transaminase > twice normal
	Hepatic encephalopathy
Hematological	Hematocrit ≤20%
	Leukocyte count ≤0.3×10^9/L
	Thrombocyte count ≤50×10^9/L
	Disseminated intravascular coagulation

Statistical Analysis

Continuous variables are summarized by means, standard deviations, and comparisons, evaluated by the Student's two-tailed t test for unequal variance. Univariate analysis of nonparametric factors between survivors and non-survivors was accomplished by using the Chi-square test. We used the Chi-square test of trend to analyze mortality rates associated with organ system failures among all patients. P values less than 0.05 were considered statistically significant. Discrimination, the ability of a model to distinguish between patients who survive and patients who die, can be assessed using receiver operating characteristic (ROC) curve analysis. ROC curves show the relationship between sensitivity (correct identification of those who die) and 1.0 minus the specificity (or the incorrect identification of those who survive). If all possible pairs of patients where one patient survived and the other died were identified, the area under the ROC curve [11-12] can be interpreted as the proportion of the pairs where the model can correctly identify the patient who died (by assigning a higher score to that patient). If the area under the ROC curve is 0.5, the model has no discriminatory power, and if the area is 1.0, the model discriminates perfectly. These tests were computed with SPSS 9.0 for Windows.

RESULTS

Subject Characteristics

From January 1999 to December 1999, one hundred and twelve patients received dialytic therapy for acute renal failure in the intensive care units. Seventy-five (67%) were men and 37 (33%) were women. Overall, the in-hospital mortality for the entire group was 67%. The ages of the survivors and the non-survivors were not significantly different statistically (mean ± SD: 58.59 ± 19.91 years vs. 58.76 ± 19.62 years, 95% confidence interval [CI]: −7.96 to 8.01, p = 0.967). No gender distinction was noted: 29.3% (22/75) of the men survived, compared to 35.1% (13/37) of the women (χ² = 0.388, p = 0.533). For all patients, the median number of failed organ on the first day of dialysis was 4 (range, 3 to 4.75; 25th to 75th percentiles). Table 2 shows the patient characteristics of both survivors and non-survivors. There were no significant differences in the number of days from time of intensive care unit admission to dialysis initiation between the two groups (9.22 ± 18.74 days vs. 9.68 ± 14.71 days, 95% CI: −5.96 to 6.89, p = 0.887).

Table 2. Patient Characteristics

	Survivors	Non-survivors	p-Value
Outcome	37	75
Age	58.59±19.91	58.76±19.62	0.967
Sex (M:F)	22 : 13	53 : 24	0.533
Interval ICU to H/D (days)	9.22±18.74	9.68±14.71	0.887
Number of organ failure	3.11±0.94	4.23±1.06	<0.001

Abbrev: M = male; F = female; ICU = intensive care unit; H/D = hemodialysis.

Renal Replacement Therapy

All patients received some form of dialytic intervention for acute renal failure, based on the usual clinical criteria. A total of 99 patients received IHD, and CRRT was performed in 13 patients. The mortality rate was lower among patients that underwent IHD than those undergoing CRRT (66.7% vs. 69.2%), although the difference did not reach statistical significance (χ² = 0.002, p = 0.968).

Mortality and Organ System Failures

The mean number of organ system failures was significantly higher in non-survivors than in survivors, 4.23 ± 1.06 vs. 3.11 ± 0.94, respectively (95% CI: 0.71 to 1.53, p<0.001). The relations between organ system failures and mortality depended on organ system failure number. There was a progressive and significantly increase in mortality associated with organ system failures among all patients (χ² for trend, χ² = 28.492, p<0.001, Fig. 1).

Figure 1. Mortality in organ system failure patients with acute renal failure requiring dialysis (χ² for trend p < 0.001).

The model ROC curve shows the true-positive and false-positive rates on the vertical and horizontal axes, respectively (Fig. 2). The area under the organ system failures prediction model ROC curve equaled 0.772 ± 0.046 (mean ± SEM) (95% CI: 0.682 to 0.862), which suggests good model discrimination.

Figure 2. Receiver operating charateristic plot of hospital mortality predictions using the organ failure systems for 112 critically ill patients with acute renal failure requiring dialysis. The 45°diagonal line represents chance performance (i.e., tossing a coin). The area under under the curve is 0.772.

DISCUSSION

There has been no significant reduction in mortality of acute renal failure over the last 40 years despite refinements in dialysis technique and improvements in the management of critically ill patients. Acute renal failure in critically-ill patients is often part of a process of progressive multiple organ failure and is frequently associated with sepsis. As dialysis can replace renal function almost indefinitely, such patients die with renal failure rather than as a consequence of it. The mortality rate of 67% in this study compares favorably with most other studies of high-risk groups [[4]], [13-18].

The number of failed organs seemed to influence the risk of dying revealing in Figure 1. However, the real survival for any given number of failed organs may be better than other studies [[14]], [[19]]. Maher and colleagues reported mortality to be 33% for two failed organs, 54% for three failed organs, 88% for four failed organs, and 100% for five-organ failure [[19]]. Study of Cosentino and colleagues showed mortality rates of 0, 35%, 71.4%, 71.5%, 87.9%, and 95.8% for two to seven failed organ systems (including ARF) respectively [[14]]. In our studies revealed mortality rates of 0, 18.2%, 53.3%, 83.3%, 89.5%, 100%, and 100% for one to seven organ failures. The above studies evaluated critically ill patients with acute renal failure in intensive care units requiring dialysis. We have found that the prognosis of critically-ill patients is closely related to the number of separate organ failures. Mortality rates increased to a large extent when more than three failed organs were noted in our studies.

The interpretation of the area under the ROC curve is quite simple. If the entire sample were divided into patients who lived and patients who died and each patient who lived was paired with each patient who died, there would be n_L×n_D such pairs (where n_L is the number of patients who lived and n_D is the number of patient who died). The area under the ROC curve is the proportion of the total number of pairs in which the model resulted in a higher probability for the patient who died than the patient who lived. Clearly, if this area is in the neighborhood of 0.50, the model is performing no better than a coin toss. Developers of models are typically not satisfied unless the ROC area of a model exceeds 0.70 [[20]]. The area under the ROC curve observed in our population is 0.772 for number of failed organs obtained within 24 hours of initiation of dialysis (Fig. 2). The results from the ROC analysis showed that organ system failures prediction model performed well in its ability to identify patients who die in the hospital.

In summary, 67% of critically ill patients with acute renal failure requiring dialysis died in the hospital. Our studies showed that the risk of patients with acute renal failure in the intensive care units is increased when any concomitant extrarenal organ system fails. Furthermore, organ system failure prediction is a good model discrimination. Mortality rate increases as the number of failed organs, on the first day of dialysis increases.