Extracorporeal membrane oxygenation utilization in burn patients with severe acute respiratory distress syndrome

Dear Editor,

Extracorporeal membrane oxygenation (ECMO) shows great promise as a rescue therapy for burn patients with inhalation injury or other forms of acute respiratory distress syndrome (ARDS). While the mainstay of traditional ARDS management in the non-burn patient includes lung protective ventilation, conservative fluid management, prone positioning and neuromuscular blockade [1], these strategies may not be feasible in the burn patient who requires ongoing fluid resuscitation and frequent operative procedures [2], [3]. ECMO support may be especially useful in mitigating these limitations and rescuing burn patients in whom ARDS progresses despite maximal traditional management. ECMO has been successfully used in patients with very large burns [4], with reported mean survival rates of 43–47% [5].

To better describe the use of ECMO in an American Burn Association-verified center, we reviewed all adult patients with a diagnosis of burn injury admitted to our Burn Unit over an approximately five-year period. This study was approved by the University of Chicago Institutional Review Board (IRB 22–1040). Between April 1, 2017 and June 30, 2022, 12 burn patients were placed on ECMO due to hypoxemic respiratory failure refractory to standard supportive care ( Table 1). The average burn size was 37.3% TBSA and 83.3% were diagnosed with a concomitant inhalation injury. On average patients were 40 years old (range 20–69), and 75% were male. Of note, no patients were ventilated in the prone position prior to ECMO initiation due the location of their burn/graft sites.

Table 1.

Patient & burn characteristics.

Patient	Age (Year)	TBSA Injury (%)	Inhalational Injury	Revised Baux	APACHE II	PaO2 nadir
1	45	50	Yes	112	25	46
2	47	35	Yes	99	a	a
3	20	45	Yes	82	a	a
4	34	40	No	74	21	40
5	69	10	Yes	96	24	72
6	21	20	Yes	58	35	61
7	56	16	Yes	89	22	60
8	35	9	Yes	61	28	66
9	52	55	Yes	124	22	82
10	35	58	Yes	110	26	48
11	23	74	Yes	114	19	74
12	44	35	No	112	20	59
Mean	40	37.3	–	91.5	24.2	59.6

^aData unavailable, patient placed on ECMO at referring center prior to arrival.

In our cohort, nine patients received venovenous (VV) ECMO, two received veno-venoarterial (V-VA) ECMO, and one received venoarterial (VA) ECMO ( Table 2). All patients met criteria for severe ARDS and had Murray scores> 3, suggesting that the severity of ARDS at the time of ECMO initiation was consistent with that used in the non-burn ECMO literature [6]. The average duration of ECMO was 258 h, or 10.75 days (range 112.5 – 692.5 h). Patients underwent a median of one debridement and/or grafting procedure prior to ECMO initiation (IRQ=1.75) and a median of 1.5 debridement and/or grafting procedures while on ECMO (IQR=3). A median of 17.9 liters of blood was transfused while on ECMO (IQR=47.7).

Table 2.

ECMO characteristics & outcomes.

Patient	ECMO Type	ECMO Run (Hr)	OR Trips on ECMO	pRBC Transfusion on ECMO (mL)	Survival
1	VV	692.5	6	40596	Yes
2	VVA	a	1	4800	No
3	VV	a	1	17932	Yes
4	VV	567.5	3	28348	Yes
5	VVA	121.75	0	2229	No
6	VV	262.75	0	5007	Yes
7	VV	253	2	3813	Yes
8	VV	138.5	1	3860	Yes
9	VV	157	3	13478	No
10	VV	365	10	58086	No
11	VA	299.5	9	31059	Yes
12	VV	112.5	0	1200	Yes
Mean	–	258	3	17534	–

^aData unavailable, patient placed on ECMO at referring center prior to arrival.

In total, eight patients (66.7%) survived to hospital discharge. Of the nine patients on VV ECMO, seven (77.7%) survived and of the three on either VA or V-VA ECMO, one survived (33.3%). The average estimated in-hospital mortality among our patients based on APACHE II scoring was 40.0% [7] and the revised Baux score predicted a 40.2% mortality rate [8]. Among non-survivors, causes of death included transition to comfort care (n = 2), thrombotic occlusion of the ECMO circuit (n = 1), and cardiac arrest (n = 1). Our survival rate is favorable when compared to larger published series [5], [9]. We believe that careful patient selection, early identification of patients who fail traditional ARDS management, and timely communication with our multidisciplinary ECMO team contributed to our outcomes.

Our data suggest that, in selected patients, ECMO for respiratory support in adult patients with burn injury is feasible. Our lower-than-expected mortality relative to existing data suggest that ECMO may improve survival in burn patients even when the duration of ECMO support was long and multiple debridement procedures were required. Our findings are limited by their retrospective nature, heterogeneity of patients and practice patterns, and variable availability of ECMO expertise. Future work should focus on establishing larger registries and multicenter studies to clarify indications and outcomes in burn patients with refractory ARDS.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Conflicts of interest

None.