Does early renal replacement therapy reduce morbidity or mortality in patients with crush injury?

Date First Published:

July 17, 2026

Last Updated:

July 17, 2026

Report by:

Dr Elizabeth Freeman, Senior Registrar, Anaesthetics (Plymouth Hospitals NHS Trust, Derriford Hospital, Plymouth)

Search checked by:

Benjamin Earle Wright, Senior Registrar, Emergency Medicine

Three-Part Question:

In [patients with suspected crush injury] [does early renal replacement therapy] [improve morbidity and mortality]?

Clinical Scenario:

You are the ICU registrar working in a over night in a Major Trauma Centre. A 32 year old man, who had been caving with friends, has been brought in to the Emergency Department, after sustaining a crush injury to both his legs when a large boulder in the cave system they were exploring dislodged and trapped him. Due to the remote location of the incident and challenges of the environment he remained trapped for a prolonged period before being extricated and brought to hospital. You are aware that crush syndrome can cause Acute Renal Failure and wonder whether if commencing Renal Replacement Therapy early on in this patient's care would be beneficial.

Search Strategy:

Please see link below for comprehensive search strategy:
http://tiny.cc/CrushBestBETsLitSearch

Search Details:

A search was conducted on 2nd and 6th December 2022 and updated on 19th November 2024 to include any relevant articles published in the intervening period. The following databases were searched:
- Medline on EBSCO platform
- CINAHL on EBSCO platform
- EMBASE on Ovid platform

Limiters:
- English language
- Last 10 years
- Conference abstracts excluded

Outcome:

71 abstracts screened, with 26 identified for full article screening.

No systematic review, clinical trial, or prospective cohort study was found comparing the effect of early vs delayed renal replacement therapy on morbidity and mortality in suspected crush injury. Results from 2 retrospective cohort studies and 5 case series deemed to be of most relevance are summarised below.

Relevant Paper(s):

Study Title	Patient Group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
The Marmara earthquake: epidemiological analysis of the victims with nephrological problems Sever MS et al. September 2001 Turkey	639 patients admitted with acute renal failure to 35 hospitals after the Marmara earthquake in 1999.	Retrospective cohort analysis	Clinical and laboratory findings	477 (74.6%) of patients received renal replacement therapy (RRT), 147 recovered without RRT and 15 died before RRT was started (factors linked to acute renal failure such as hyperkalaemia and circulatory overload reported not to have played a major role in this group). Overall mortality was 15.2% (17.2% in the RRT group, 9.3% in the non RRT group). On multivariate analysis, only DIC and ARDS/respiratory failure showed a significant association with mortality.	Criteria for commencing RRT not reported. Time to initiation of RRT not reported.
			Surgical interventions
			Frequency and duration of RRT
Clinical features and outcome of crush syndrome caused by the Chi-Chi earthquake Huang KC, Lee TS, Lin YM, Shu KH April 2002 Taiwan	95 patients with crush syndrome (defined as peak CK >1000 u/L), admitted to 8 major hospitals in the area surrounding the Chi-Chi earthquake in 1999, within the first two weeks.	Retrospective cohort analysis	Laboratory data	44 of the 95 patients were reported to have acute renal failure, and 30 of these received dialysis. All 30 regained normal renal function. Incidence of acute oligo-/anuria, renal failure and need for haemodialysis were increased significantly in the group with peak CK >50,000. 8 patients (8.4%) died.	Mortality in the acute renal failure group not reported. Time to initiation of dialysis not reported.
			Time trapped
			Time to hospital
			Injury sites
			Total fluid volume in first 48 hours
			Oliguria
			Dialysis
			Fasciotomy
			Morbidity
			Mortality
			Discharge
			Transfer
The outcome of patients presenting with crush syndrome after the Marmara earthquake Kazancioğlu R et al. January 2001 Turkey	60 patients admitted to units of the Istanbul School of Medicine following the Marmara earthquake (1999), fitting criteria for diagnosis of crush syndrome.	Case series	Laboratory parameters	40 patients required RRT (peritoneal dialysis in 1, haemodialysis in 37, haemofiltration in 2). The clinical and laboratory data of those who required RRT (n=40) and those who did not (n=20) were compared. Time spent under the rubble, admission blood pressure, serum potassium, phosphate, uric acid creatinine phosphokinase and haematocrit differed between the two groups (p<0.05). Of note, the RRT group had shorter mean time under the rubble and higher mean admission BP. Mortality in the RRT group was 23%, and in the non RRT group 20%. Overall mortality was 21.6%, with all remaining patients reported to regain normal renal function	Criteria for starting RRT not reported. Time to initiation of RRT not reported.
			Trauma sites
			Fasciotomies
			Requirement for RRT
			Type of RRT
			Complications
			Cause of mortality
Survival analysis of the factors affecting in mortality in injured patients requiring dialysis due to acute renal failure during the Marmara earthquake: survivors vs non-survivors Ersoy A et al. May 2003 Turkey	60 patients admitted to Uladag University Medical School hospital following the Marmara earthquake (1999) who underwent RRT.	Case series	Mortality (time point not specified).	Patients divided into survivors (Group A, n=39) and non-survivors (Group B, n=21). Dialysis was started in group A at a mean time of 2.8 +/- 0.2 days (range 1-8) and in group B at 3.7 +/- 0.6 days (range 1-12) post earthquake, with a p value >0.05. The parameters found to be statistically significant predictors of mortality were female gender, multiple trauma, serum peak CK >20,000 U/l and systolic hypotension on admission. Overall mortality rate in dialysed patients was 35%.	No analysis of data from patients who did not undergo RRT.
Crush syndrome: saving more lives in disasters: lessons learned from the early-response phase in Haiti Bartal C et al. April 2011 Haiti	8 patients with acute oligoanuric renal failure treated at the Israel Defence Forces medical Corps Field Hospital in Haiti, 2010 (RRT not available).	Case series	Mortality following 48 hours of conservative treatment.	8 patients were admitted with oligo-/anuric renal failure having had no early prophylactic treatment. RRT was not available. Of these, 2 recovered within 48 hours, 4 patients (50%) died, and 2 were transferred to another hospital for palliative care with signs of pulmonary oedema. Clinical characteristics and laboratory parameters were compared between the two groups but statistical analysis was not performed.	Small case series. No statistical analysis performed.
Characteristics of crush syndrome caused by prolonged limb compression longer than 24 h in the Sichuan earthquake Chunguang Z et al. August 2010 China	9 patients with severe crush syndrome admitted to the Intensive Care Unit at the West China Hospital of Sichuan University following the Sichuan earthquake (2008).	Case series	Laboratory findings	All 9 patients whose data was analysed survived. All patients received unspecified volumes of intravenous fluid prior to transfer to the hospital studied. Duration of burial, time to administration of fluid and time from injury to ICU admission were reported for each case but not analysed with respect to other parameters/outcome measures.	Small case series Time to RRT not reported No statistical analysis performed
			Interventions
			Complications
			Mortality (time point not specified).
Crush syndrome patients after the Marmara earthquake Demirkiran O et al. May 2003 Turkey	18 patients admitted to the intensive care unit of a University Hospital following the Marmara earthquake in 1999.	Case series	Patient characteristics (age, sex, APACHE II score, time to rescue, time to admission, length of stay, mortality)	The mean time to rescue was 24.1 hours (range 8-45) and time from first hospital admission to transfer to ICU was 16.35 days (range 45 minutes – 72 hours). 13 patients developed renal failure, 6 underwent continuous renal replacement therapy (CRRT) and 7 underwent haemodialysis. 7 patients of the 25 admitted to ICU died, not all of whom received RRT.	Small case series Multiple missing data points Time to initiation of RRT not reported Requirement for RRT not correlated with time to admission to ICU
			Laboratory findings
			Interventions

Author Commentary:

No study has specifically looked at whether the timing of initiation of RRT affects morbidity and mortality in crush injury. All existing studies are retrospective analyses of data from disparate groups of patients with acute renal failure following crush injury. Most of these are patients evacuated after earthquakes, where timing of evacuation and access to initial treatment and resuscitation varies significantly. Access to RRT is also variable, often requiring further evacuation and inter-hospital transfer. Data is often incomplete and the timing of initiation of RRT is infrequently reported. The clinical thresholds for starting RRT in these patient populations are not specified, and are also likely to be variable and resource dependant. In the existing studies that include data on incidence of acute renal failure, use of RRT and mortality, conclusions on causality cannot be drawn due to presence of multiple confounding factors.

Existing review articles are based on expert opinion. It has been recommended that RRT in crush injury is initiated according to standard indications (oliguria/anuria, volume overload or severe uraemia/hyperkalaemia/acidosis), with consideration of prophylactic RRT in those at high risk of hyperkalaemia [3].

The existing literature does suggest that not all patients who develop acute renal failure secondary to crush injury require RRT, and early, individualised intravenous fluid therapy can prevent AKI and avoid the need for RRT [4]. Given the risks associated with starting RRT (eg. large bore access, coagulopathy, haemodynamic instability) and the challenges around resource management in mass casualty incidents associated with crush injury, early RRT is unlikely to be clinically justifiable without evidence of benefit.

Of note, early initiation of RRT in critically ill patients with acute kidney injury (non-crush specific) has not demonstrated a survival benefit, and remains controversial [1][2]. A well-designed randomised trial examining the effect of pre-defined ‘early’ and ‘standard’ initiation of RRT on morbidity and mortality is required in both crush and non-crush related acute kidney injury.

Bottom Line:

There is inadequate evidence to address the question of whether early RRT reduces morbidity and mortality in patients with crush injury. Current guidelines are based on expert opinion and suggest initiating RRT according to standard indications.

Level of Evidence:

Level 2: Studies considered were neither 1 or 3

References:

Sever MS et al.. The Marmara earthquake: epidemiological analysis of the victims with nephrological problems
Huang KC, Lee TS, Lin YM, Shu KH. Clinical features and outcome of crush syndrome caused by the Chi-Chi earthquake
Kazancioğlu R et al.. The outcome of patients presenting with crush syndrome after the Marmara earthquake
Ersoy A et al.. Survival analysis of the factors affecting in mortality in injured patients requiring dialysis due to acute renal failure during the Marmara earthquake: survivors vs non-survivors
Bartal C et al.. Crush syndrome: saving more lives in disasters: lessons learned from the early-response phase in Haiti
Chunguang Z et al.. Characteristics of crush syndrome caused by prolonged limb compression longer than 24 h in the Sichuan earthquake
Demirkiran O et al.. Crush syndrome patients after the Marmara earthquake