COVID-19 and Polymyxin B Hemoperfusion – Data from the EUPHAS-2 registry

Polymyxin B hemoperfusion experience from critically ill COVID-19 patients with endotoxic shock has been published recently in Artificial Organs.

The study refers to an observational study including 12 COVID-19 patients who developed a secondary infection leading to septic shock status and whose data were collected in the EUPHAS-2 registry from February to May 2020.

The importance of secondary bacterial infections – endotoxemia in critically ill COVID-19 patients and the role of Polymyxin B hemoperfusion as complementary therapy have been described really well in an interesting Webinar held during the last ESICM LIVES 2020.


Aim of study

The  aim  of  this  study  was  to  evaluate  the  role  of extracorporeal  endotoxin  removal  by  Polymyxin  B  Hemoperfusion  (PMX-HP),  in  the  treatment  of  patients  with COVID-19  and  secondary  bacterial  infection.



A subgroup analysis of a multicenter, multinational, prospective, observational web-based database (EUPHAS-2) was conducted. The EUPHAS 2 aims to collect data on the daily use of PMX-HP in critically ill patients with septic shock, in order to identify specific patient populations that may benefit more from this therapy and to understand if specific approaches may be more beneficial. Sharing knowledge and creating an international network of PMX-HP users.

Currently the most frequent protocol for PMX-HP application consists of two sessions which are run every 24 hours and last 120 minutes each.



The study included 12 patients (61.5 years [54.8-68], N=3 (25%) female) with SARS-CoV-2  infection confirmed by real-time reverse transcriptase-polymerase chain reaction (RT-PCR) from nasal/oral swab, admitted to the ICU between February and May 2020, collected in 3 Italian Hospitals and 1 Spanish, and who were affected by septic shock due to secondary infection – endotoxemia and received PMX-HP.

Median SAPS II score at admission was 57.5 [52.8-75.1] and APACHE II was 23.0 [21.5-26.3].

Septic shock (sepsis – vasopressors use – blood lactate levels > 2mmol/l) was diagnosed in 9 patients (75%), Gram-negative bacteria was identified in most of the microbiological cultures (N=17, 65%), followed by Gram-positive bacteria in (N=4, 15%), fungi (N=3, 12%) and no growth (N= 2, 8%).

Baseline Endotoxin Activity Level was 0.78 [0.70-0.92].

SOFA score progressively improved over the next 120 hours following PMX-HP (median [IQR] SOFA score was 14 [12-15] at T0 and 10 [6-11] at T120, p = 0.001,) and was associated with median endotoxin activity assay (EAA) decrease from 0.78 [0.70-0.92] at T0 to 0.60 [0.44-0.72] at T120 (p = 0.245).

In addition, a direct correlation was observed between SOFA score reduction and EAA level trend over time.

Lung Injury Score (LIS) decreased from 2.88 [2.5-3.31] at T0 to 2.5 [1.5-2.67] at T120 (p = 0.02), and septic shock markers over the same period improved with VDI from 28.6 [24.4-39.5] at T0 to 0 [0-4.23] at T120, p < 0.001; and Lactates from 3.4 [2.33-4.3] at T0 to 1.6 [1-2] at T120, p<0.001)

Nine out of 12 patients (75%) required CRRT because of AKI.



In this series of consecutive critically ill COVID-19  patients with endotoxic shock, PMX-HP was associated with organ function recovery, hemodynamic and shock markers improvement and reduction of EAA levels.

And No PMX-HP related complications were observed.