Acute renal failure

Published on 27/02/2015 by admin

Last modified 22/04/2025

Print this page

This article have been viewed 1880 times

Chapter 40 Acute renal failure

Acute renal failure (ARF) remains a major diagnostic and therapeutic challenge for the critical care physician. The term ‘ ARF’ describes a syndrome characterised by a rapid (hours to days) decrease in the kidney’s ability to eliminate waste products. Such loss of function is clinically manifested by the accumulation of end-products of nitrogen metabolism such as urea and creatinine. Other typical clinical manifestations include decreased urine output (not always present), accumulation of non-volatile acids and an increased potassium and phosphate concentration.

Depending on the criteria used to define its presence, ARF has been reported to occur in 15–20% of intensive care unit (ICU) patients.¹ Recently a consensus definition and classification for ARF have been developed and validated in hospital and ICU patients.²^–⁴ This definition, which goes by the acronym of RIFLE, divides renal dysfunction into the categories of risk, injury and failure (Figure 40.1) and is likely to be the dominant approach to defining ARF in ICUs for the next 5–10 years. Using this classification, the incidence of some degree of renal dysfunction has reported to be up to 67% in a recent study of > 5000 ICU patients.³ In hospital patients, the development of failure (in the RIFLE classification) increases the odds ratio of death 10 times.⁴

Figure 40.1 RIFLE (risk injury failure loss end-stage) classification scheme for acute renal failure. The classification system includes separate criteria for creatinine and urine output. The criteria, which lead to the worst possible classification, should be used. Note that RIFLE-F (F = failure) is present even if the increase in serum creatinine concentration (S_Creat) is less than threefold as long as the new S_Creat > 4.0 mg/dl (350 μmol/l) in the setting of an acute increase of at least 0.5 mg/dl (44 μmol/l). The designation RIFLE-F_C should be used in this case to denote ‘ acute-on-chronic’ disease. Similarly, when RIFLE-F classification is reached by urine output criteria only, a designation of RIFLE-F_O should be used to denote oliguria. The shape of the figure denotes the fact that more patients (high sensitivity) will be included in the mild category, including some who do not actually have ‘ renal failure’ (less specificity). In contrast, at the bottom, the criteria are strict and therefore specific, but some patients with renal dysfunction might be missed. GFR, glomerular filtration rate; ARF, acute renal failure; UO, urine output.

ASSESSMENT OF RENAL FUNCTION

Renal function is complex (control of calcium and phosphate, acid–base balance, water balance, tonicity, erythropoiesis, disposal of some cytokines, lactate removal). In the clinical context, however, monitoring of renal function is reduced to the indirect assessment of glomerular filtration rate (GFR) by the measurement of urea and creatinine in blood. These waste products are insensitive markers of GFR and are heavily modified by nutrition, the use of steroids, the presence of gastrointestinal blood, muscle mass, age, gender or muscle injury. Furthermore, they start becoming abnormal only when more than 50% of GFR is lost, they do not reflect dynamic changes in GFR and are grossly modified by aggressive fluid resuscitation. The use of creatinine clearance (2- or 4-hour collections) or of calculated clearance by means of formulae might increase the accuracy of GFR estimation, but rarely changes clinical management. The use of more sophisticated radionuclide-based tests is cumbersome in the ICU and only useful for research purposes.

DIAGNOSIS AND CLINICAL CLASSIFICATION

The most practically useful approach to the aetiological diagnosis of ARF is to divide its causes according to the probable source of renal injury: prerenal, renal (parenchymal) and postrenal.

PRERENAL RENAL FAILURE

This form of ARF is by far the most common in ICU. The term indicates that the kidney malfunctions predominantly because of systemic factors, which decrease GFR. For example, GFR may be decreased if renal blood flow is diminished by decreased cardiac output, hypotension or raised intra-abdominal pressure. Raised intra-abdominal pressure can be suspected on clinical grounds and confirmed by measuring bladder pressure with a urinary catheter. A pressure of > 25–30 mmHg above the pubis should prompt consideration of decompression. If the systemic cause of renal failure is rapidly removed or corrected, renal function improves and relatively rapidly returns to near-normal levels. However, if intervention is delayed or unsuccessful, renal injury becomes established and several days or weeks are then necessary for recovery. Several tests (measurement of urinary sodium, fractional excretion of sodium and other derived indices) have been promoted to help clinicians identify the development of such ‘ established’ ARF. Unfortunately, their accuracy is doubtful.⁵ The clinical utility of these tests in ICU patients who receive vasopressors, massive fluid resuscitation and loop diuretic infusions is low. Furthermore, it is important to observe that prerenal ARF and established ARF are part of a continuum, and their separation has limited clinical implications. The treatment is the same – treatment of the cause while promptly resuscitating the patient using invasive haemodynamic monitoring to guide therapy.

PARENCHYMAL RENAL FAILURE

This term is used to define a syndrome where the principal source of damage is within the kidney and where typical structural changes can be seen on microscopy. Disorders which affect the glomerulus or the tubule can be responsible (Table 40.1).

Table 40.1 Causes of parenchymal acute renal failure

Glomerulonephritis

Vasculitis

Interstitial nephritis

Malignant hypertension

Pyelonephritis

Bilateral cortical necrosis

Amyloidosis

Malignancy

Nephrotoxins

Among these, nephrotoxins are particularly important, especially in hospital patients. The most common nephrotoxic drugs affecting ICU patients are listed in Table 40.2. Many cases of drug-induced ARF rapidly improve upon removal of the offending agent. Accordingly, a careful history of drug administration is mandatory in all patients with ARF. In some cases of parenchymal ARF, a correct working diagnosis can be obtained from history, physical examination and radiological and laboratory investigations. In such patients one can proceed to a therapeutic trial without the need to resort to renal biopsy. However, if immunosuppressive therapy is considered, renal biopsy is recommended. Renal biopsy in ventilated patients under ultrasound guidance does not carry additional risks compared to standard conditions.

Table 40.2 Drugs which may cause acute renal failure in the intensive care unit

Radiocontrast agents

Aminoglycosides

Amphotericin

Non-steroidal anti-inflammatory drugs

β-lactam antibiotics (interstitial nephropathy)

Sulphonamides

Acyclovir

Methotrexate

Cisplatin

Cyclosporin A

FK-506 (tacrolimus)

More than a third of patients who develop ARF in ICU have chronic renal dysfunction due to factors such as age-related changes, long-standing hypertension, diabetes or atheromatous disease of the renal vessels. Such chronic renal dysfunction may be manifest by a raised serum creatinine. However, this is not always the case. Often, what may seem to the clinician to be a relatively trivial insult, which does not fully explain the onset of ARF in a normal patient, is sufficient to unmask lack of renal functional reserve in another.

HEPATORENAL SYNDROME

This condition is a form of ARF, which occurs in the setting of severe liver dysfunction in the absence of other known causes of ARF. Typically, it presents as progressive oliguria with a very low urinary sodium concentration (< 10 mmol/l). Its pathogenesis is not well understood. It is important to note, however, that, in patients with severe liver disease, other causes of ARF are much more common. They include sepsis, paracentesis-induced hypovolaemia, raised intra-abdominal pressure due to tense ascites, diuretic-induced hypovolaemia, lactulose-induced hypovolaemia, alcoholic cardiomyopathy, and any combination of these. The avoidance of hypovolaemia by albumin administration in patients with spontaneous bacterial peritonitis has been shown to decrease the incidence of renal failure in a recent randomised controlled trial (RCT).⁶ These causes must be looked for and promptly treated. Recent small studies suggest that a vasopressin derivative called terlipressin may improve GFR in this condition.⁷

RHABDOMYOLYSIS-ASSOCIATED ARF

This condition accounts for close to 5–10%⁸ of cases of ARF in ICU depending on the setting. Its pathogenesis involves prerenal, renal and postrenal factors. It is now typically seen following major trauma, drug overdose with narcotics, vascular embolism, and in response to a variety of agents which can induce major muscle injury. The principles of treatment are based on retrospective data, small series and multivariate logistic regression analysis, because no RCTs have been conducted. They include prompt and aggressive fluid resuscitation, elimination of causative agents, correction of compartment syndromes, the alkalinisation of urine (pH > 6.5) and the maintenance of polyuria (> 300 ml/h). The role of mannitol is controversial.

POSTRENAL RENAL FAILURE

Obstruction to urine outflow is the most common cause of functional renal impairment in the community,⁹ but is uncommon in the ICU. Typical causes of obstructive ARF include bladder neck obstruction from an enlarged prostate, ureteric obstruction from pelvic tumours or retroperitoneal fibrosis, papillary necrosis or large calculi. The clinical presentation of obstruction may be acute or acute-on-chronic in patients with long-standing renal calculi. It may not always be associated with oliguria. If obstruction is suspected, ultrasonography can be easily performed at the bedside. However, not all cases of acute obstruction have an abnormal ultrasound and, in many cases, obstruction occurs in conjunction with other renal insults (e.g. staghorn calculi and severe sepsis of renal origin). Assessment of the role of each factor and overall management should be conducted in conjunction with a urologist. Finally the sudden and unexpected development of anuria in an ICU patient should always suggest obstruction of the urinary catheter as the cause. Appropriate flushing or changing of the catheter should be implemented in this setting.

PATHOGENESIS OF ACUTE RENAL FAILURE

The pathogenesis of obstructive ARF involves several humoral responses as well as mechanical factors. The pathogenesis of parenchymal renal failure is typically immunological. It varies from vasculitis to interstitial nephropathy and involves an extraordinary complexity of immunological mechanisms. The pathogenesis of prerenal ARF is of greater direct relevance to the intensivist. Several mechanisms have been reported to play a major role in the development of renal injury:

1 ischaemia of outer medulla with activation of the tubuloglomerular feedback ¹⁰

2 tubular obstruction from casts of exfoliated cells ¹¹

3 interstitial oedema secondary to back-diffusion of fluid ¹¹

4 humorally mediated afferent arteriolar renal vasconstriction ¹¹

5 inflammatory response to cell injury and local release of mediators ¹¹

6 disruption of normal cellular adhesion to the basement membrane ¹¹

7 radical oxygen species-induced apoptosis ¹¹

8 phospholipase A₂-induced cell membrane injury ¹²

9 mitogen-activated protein kinase-induced renal injury ¹³

In septic patients with hyperdynamic circulations, ARF may occur in the setting of increased renal blood flow, with the loss of GFR being secondary to decreased intraglomerular pressure induced by efferent arteriolar vasodilatation.¹⁴

No studies exist to tell us which of the above mechanism/s are pathogenetically most important in ICU patients with or without sepsis and when they might be active in the course of a patient’s illness.

THE CLINICAL PICTURE

The most common clinical picture seen in ICU is that of a patient who has sustained or is experiencing a major systemic insult (trauma, sepsis, myocardial infarction, severe haemorrhage, cardiogenic shock, major surgery and the like). When the patient arrives in the ICU, resuscitation is typically well under way or surgery may have just been completed. Despite such efforts, the patient is already anuric or profoundly oliguric, the serum creatinine is rising and a metabolic acidosis is developing. Potassium and phosphate levels may be rapidly rising as well. Accompanying multiple-organ dysfunction (mechanical ventilation and need for vasoactive drugs) is common. Fluid resuscitation is typically undertaken in the ICU under the guidance of invasive hemodynamic monitoring. Vasoactive drugs are often used to restore mean arterial pressure (MAP) to ‘ acceptable’ levels (typically > 65–70 mmHg). The patient may improve over time and urine output may return with or without the assistance of diuretic agents. If urine output does not return, however, renal replacement therapy needs to be considered.

If the cause of ARF has been removed, and the patient has become physiologically stable, slow recovery occurs (from 4–5 days to 3–4 weeks). In some cases, urine output can be above normal for several days. If the cause of ARF has not been adequately remedied, the patient remains gravely ill, the kidneys do not recover and death from multiorgan failure may occur.

PREVENTING ACUTE RENAL FAILURE

The fundamental principle of ARF prevention is to treat its cause. If prerenal factors contribute, these must be identified and haemodynamic resuscitation quickly instituted.

RESUSCITATION

Intravascular volume must be maintained or rapidly restored, and this is often best done using invasive hemodynamic monitoring (central venous catheter, arterial cannula and pulmonary artery catheter or pulse contour cardiac output catheters in some cases). Oxygenation must be maintained. An adequate haemoglobin concentration (at least > 70 g/l) must be maintained or immediately restored. Once intravascular volume has been restored, some patients remain hypotensive (MAP < 70 mmHg). In these patients, autoregulation of renal blood flow may be lost. Restoration of MAP to near-normal levels may increase GFR.^15,¹⁶ Such elevations in MAP, however, require the addition of vasopressor drugs.^15,¹⁶ In patients with hypertension or renovascular disease, a MAP of 75–80 mmHg may still be inadequate. The nephroprotective role of additional fluid therapy in a patient with a normal or increased cardiac output and blood pressure is questionable. Despite these resuscitation measures renal failure may still develop if cardiac output is inadequate. This may require a variety of interventions from the use of inotropic drugs to the application of ventricular assist devices.

NEPHROPROTECTIVE DRUGS

Following haemodynamic resuscitation and removal of nephrotoxins, it is unclear whether the use of additional pharmacological measures is of further benefit to the kidneys.

RADIOCONTRAST NEPHROPATHY

In patients receiving radiocontrast an RCT suggested that saline infusion to maintain intravascular fluid expansion is superior to the addition of mannitol or furosemide. Several RCTs of similar patients demonstrated a beneficial effect of N-acetylcysteine treatment before and after radiocontrast administration, as confirmed in a recent meta-analysis.²³ Since these preventive interventions have minimal toxicity, they should be considered whenever a patient is scheduled for the administration of intravenous radiocontrast. Their effectiveness in ICU patients who have already been fluid-resuscitated remains unknown.

DIAGNOSTIC INVESTIGATIONS

An aetiological diagnosis of ARF must always be established. Such diagnosis may be obvious on clinical grounds. However, in many patients, it is best to consider all possibilities and exclude common treatable causes by simple investigations. Such investigations include the examination of urinary sediment and exclusion of a urinary tract infection (in most, if not all, patients), the exclusion of obstruction when appropriate (some patients) and the careful exclusion of nephrotoxins (all patients).

In specific situations, other investigations are necessary to establish the diagnosis, such as creatine kinase and free myoglobin for possible rhabdomyolysis. A chest radiograph, a blood film, the measurement of non-specific inflammatory markers and the measurement of specific antibodies (antiglomerular basement membrane, antineutrophil cytoplasm, anti-DNA, antismooth-muscle) are useful screening tests to help support the diagnosis of vasculitis or of certain types of collagen disease or glomerulonephritis. If thrombotic thrombocytopenic purpura is suspected, the additional measurement of lactic dehydrogenase, haptoglobin, unconjugated bilirubin and free haemoglobin is needed. In some patients, specific findings (cryoglobulins, Bence Jones proteins) are almost diagnostic. In a few rare patients, the clinical picture, laboratory investigations and radiological investigations are not sufficient to make a causative diagnosis with sufficient certainty. In such patients a renal biopsy might become necessary.

MANAGEMENT OF ESTABLISHED ACUTE RENAL FAILURE

The principles of management of established ARF are the treatment or removal of its cause and the maintenance of physiological homeostasis while recovery takes place. Complications such as encephalopathy, pericarditis, myopathy, neuropathy, electrolyte disturbances or other major electrolyte, fluid or metabolic derangement should never occur in a modern ICU. Their prevention may include several measures, which vary in complexity from fluid restriction to the initiation of extracorporeal renal replacement therapy.

Nutritional support must be started early and must contain adequate calories (30–35 kcal/kg per day) as a mixture of carbohydrates and lipids. Adequate protein (at least 1–2 g/kg per day) must be administered. There is no evidence that specific renal nutritional solutions are useful. Vitamins and trace elements should be administered at least according to their recommended daily allowance. The role of newer immunonutritional solution remains controversial. The enteral route is preferred to the use of parenteral nutrition.

Hyperkalaemia (> 6 mmol/l) must be promptly treated with insulin and dextrose administration, the infusion of bicarbonate if acidosis is present, the administration of nebulised salbutamol, or all of the above together. If the ‘ true’ serum potassium is > 7 mmol/l or electrocardiographic signs of hyperkalaemia appear, calcium chloride (10 ml of 10% solution intravenously) should also be administered. The above measures are temporising actions while renal replacement therapy is being set up. The presence of hyperkalaemia is a major indication for the immediate institution of renal replacement therapy.

Metabolic acidosis is almost always present but rarely requires treatment per se. Anaemia requires correction to maintain a haemoglobin of at least > 70 g/l. More aggressive transfusion needs individual patient assessment.²⁴ Drug therapy must be adjusted to take into account the effect of the decreased clearances associated with loss of renal function. Stress ulcer prophylaxis is advisable and should be based on H₂-receptor antagonists or proton pump inhibitors in selected cases. Assiduous attention should be paid to the prevention of infection.

Fluid overload can be prevented by the use of loop diuretics in polyuric patients. However, if the patient is oliguric, the only way to avoid fluid overload is to institute renal replacement therapy at an early stage (see Chapter 41). Marked azotaemia ([urea] > 40 mmol/l or [creatinine] > 400 μmol/l) is undesirable and should probably be treated with renal replacement therapy unless recovery is imminent or already under way and a return toward normal values is expected within 24–48 hours. It is recognized, however, that no RCTs exist to define the ideal time for intervention with artificial renal support.