Severe acute arsine poisoning treated by plasma exchange

Abstract

Introduction. Exposure to arsine gas can cause fatal hemolysis and multiorgan damage. Whole blood exchange transfusion and hemodialysis have been recommended to treat severe acute arsine poisoning, but are associated with significant complications and sub-optimal outcomes. Plasma exchange is another method of blood purification technique but there are no data on its use in acute arsine poisoning. This retrospective study evaluated the clinical and effects and arsenic clearance from the use of plasma exchange treatment of patients with acute arsine poisoning. Methods. Data from patients with severe acute arsine poisoning, treated with plasma exchange from December 2000 to December 2005 were collected and analyzed. Measured laboratory factors, performed before and after plasma exchange treatment included routine biochemistry and hematology tests as well as arsenic concentrations in blood, urine, and discarded plasma. Results. During the study period, 12 patients with severe acute arsine poisoning were treated with plasma exchange. Plasma exchange was performed one or two times on each patient, during which the replacement fluid was fresh frozen plasma (total volume ranged from 1400 to 4000 mL). The range of concentrations of arsenic in discarded plasma was 27.7 to 88.7 mg/L and the range of total arsenic removed by plasma exchange was 55.4 to 177.4 mg. Plasma exchange appears to rapidly terminate arsine-induced hemolysis and favorably modify damage to the kidneys and other organs. Laboratory factors that showed significant association with treatment response were creatine kinase, lactate dehydrogenase, blood urea nitrogen, total bilirubin, and heart-related enzymes. All patients recovered from the poisoning and were in good condition at a 2 to 3 months follow-up. Conclusions. Plasma exchange appears to be an effective treatment intervention for patients with severe acute arsine poisoning. It is suggested that it be used as early as possible.

Keywords:

• Severe acute arsine poisoning
• Plasma exchange

Introduction

Arsine is a colorless gas with a mild, garlic-like odor, which can cause acute fatal hemolysis and serious damage to kidney, liver and other organs (1–5). Most cases of the poisoning occur when arsine is unintentionally generated as a by-product of a chemical reaction involving an acid and a metal containing an arsenic impurity. When water or acid comes in contact with these ores or metals, arsine gas, as a waste, may be released in hazardous concentrations (6).

Acute exposure to arsine is fatal in up to 25% of reported human cases (7). In China, the fatalities occurred in about 37.8% (28/74) of the human arsine poisoning cases reported from 1995 to 1999 (8), but may be as high as 55.6% (8/14) even when considerable medical interventions are carried out (9). Acute arsine poisoning is seldom reported in developed countries nowadays but it remains a serious problem in China where most human exposures occur after accidental formation from arsenicals mixed with acid or water (8,9).

Without specific antidotes for arsine, treatment consists of respiratory, vascular and renal support. Treatment of severe acute arsine poisoning (SAAP), according to the National Institute for Occupational Safety and Health (NIOSH) guidelines (10), should include an immediate blood exchange transfusion to replace hemolysed red blood cells and remove arsenic and the hemoglobin-arsine complexes, or hemodialysis (HD) to minimize kidney damage; however, only sporadic cases of SAAP treated with whole blood exchange transfusion have been reported (3,4,11). Whole blood exchange transfusion has its side effects (12) and is inconsistent with the current opinion on component blood transfusion (13). Currently, there are many blood purification methods, including plasma exchange (PE), HD, peritoneal dialysis, and continuous veno-venous hemodialysis (CVVH). There are no studies of the use of PE or other blood purification methods in the treatment of SAAP.

PE is an extracorporeal technique that removes toxins, metabolites, inflammatory factors and toxic mediators from the plasma (14–17) and it is hypothesized that removal of these factors can be therapeutic in certain situations. We hypothesized that PE would effectively remove the toxins and intravascular erythrocyte fragments, and metabolites of arsine that accumulate in blood following arsine exposure cause the secondary damage to kidney and other organs. From December 2000 to December 2005, we treated SAAP patients with PE.

Patients and methods

Patients

The proposal for our retrospective study was approved by our hospital's review board. All patients with SAAP admitted to our hospital from December 2000 to December 2005 and treated by the PE method were included. The diagnostic criteria of SAAP are based on the Diagnostic Criteria of Occupational Acute Arsine Poisoning (GBZ44–2002) published by the Health Ministry of the People's Republic of China on April 8, 2002 (18) and all the following criteria need to be met: (1) A history of exposure to arsine gas and early onset of headache, nausea and dark red or brown discolouration of the urine; (2) Evidence of clinical toxicity, including abdominal pain, hematuria, oliguria or anuria, jaundice, anemia, cyanosis or disturbance of consciousness; (3) Clinical picture consistent with severe intravascular hemolysis and laboratory results demonstrating intravascular erythrocyte fragments, anisocytosis and poikilocytosis, or reticulocytosis, accompanied by an apparent decrease in peripheral blood hemoglobin (Hb) concentration or a strong positive test for occult blood; (4) Evidence of a secondary toxic nephropathy, including elevated serum creatinine concentrations; and (5) confirmation of the diagnosis be at least three physicians who specialize in occupational diseases and clinical toxicology.

Exclusion criteria

Patients with SAAP who were treated with other blood purification methods, such as HD, were excluded. Mild acute arsine poisoning patients (defined as having only symptoms of headache, nausea, abdominal pain and red urine, mild intravascular hemolysis without oliguria or anuria, and complete recovery in 1 to 2 weeks) were not included in the study.

Treatment with PE

Informed consent was obtained from each patient prior to PE treatment. The patients admitted to our hospital with SAAP within 48 hours of exposure to arsine and who were experiencing severe hemolysis were treated with PE as soon as possible. A femoral or internal jugular vein was used for PE access. Operations were almost simultaneously performed on the patients, when massive numbers of patients were enrolled, by means of Accura or BM₂₅ (Baxter Healthcare Ltd USA). During the PE process, the flow rate was 100–200 mL/minute using a total of 1400 to 2000 mL (30 to 35mL/kg body weight) of fresh frozen plasma (FFP) and 0.9% saline. The plasmapheresis device was a Plasmaflux P₂S (Fresenius, Germany). Heparin was used for anticoagulation: a loading dose of 4000 units was followed by maintenance doses of 250 to 400 units/hour depending on the activated partial thromboplastin time (APTT, INR) determined by Whole Blood Microcoagulation System Hemochron® Jr. Signature (ITC, USA).

Supportive treatment

Supportive care of respiratory, vascular, and renal function was used according to the state of illness. Vital signs and urine volumes were monitored in all the patients. All patients received reduced glutathione 2.4–3.6 g/day as a continuous IV infusion, dexamethasone 10–20 mg/day IV for the first three days, and 5% sodium bicarbonate 100–250 ml/day as a continuous IV infusion. Two patients with Hb less than 40g/L after PE were given 2 units of packed red cells in the subsequent 24 hours. Non-invasive positive pressure ventilation (NIPP) was used in two patients with acute respiratory failure (Po₂ 30–50 mmHg and normal Pco₂) by means of Esprit™ Ventilator (Respironics, Inc. USA) with the following settings: IPAP 8∼12 cm H₂O, EPAP 4∼6 cm H₂O, O₂ concentration 30–40%.

Determination of arsenic in blood and urine

Before and after the 12 hours of PE, samples of venous blood and urine, as well as the discarded plasma from the PE treatment (after being well-mixed) were collected. The concentrations of arsenic blood, discarded plasma, and urine were measured following the Hydride Generation Atomic Fluorescence Spectrometric Method, using AFS-230E Atomic Fluorescence Photometer (Beijing Haiguang Inc. China).

Methods of observation and measurement

Before PE (0h) and at 12h, 24h, 72h, 1w (1 week), and 2w after PE, blood samples were collected to measure cardiac enzymes, creatine kinase (CK), lactate dehydrogenase (LDH), alkaline phosphatase (ALP), alanine aminotransferase (ALT), aspartate aminotransferase (AST), α-hydroxybutyric dehydrogenase (HBDH), total bilirubin (TBIL), unconjugated bilirubin (UBIL), conjugated bilirubin (CBIL), blood urea nitrogen (BUN), and serum creatinine (Cr); a Synchron LX-20 (Backman-Coulter, Lmt. USA) was used for the analyses. At the same time, routine blood and urine routine tests were performed. In 24–72 hour period post-PE, an ultrasonic examination of the abdomen and a chest x-ray were performed. An analysis of blood gas was performed, too, as needed.

Statistics

All statistical analyses were performed using SAS (vision 9.0, SAS institute, Cary, NC). All data were expressed as the mean ± standard deviation. Statistical analyses were calculated with paired t tests. An asterisk (*) was used to denote values that were statistically different from control (p < 0.05) and double asterisks (**) were used when they were statistically different from control (p < 0.01). The correlation analysis was evaluated by means of linear regression model.

Results

There were 41 cases of exposure to arsine during the study period of December 2000 to December 2005. Twenty-seven patients with mild acute arsine poisoning completely recovered in 1–2 weeks of treatment by medications. Fourteen patients who were referred to our hospital fulfilled the diagnostic criteria for SAAP. Two patients treated with HD were excluded; one died immediately after HD and the other spent months recovering from kidney damage. Twelve patients with SAAP treated with PE were included in this retrospective study.

The 12 study patients had a mean age of 47.92 ± 11.87 years (range 23 to 67 years). There were six women, including one who was 5 months pregnant. One patient worked in a metal refinery and was exposed to arsine when he sprayed water on a large quantity of molten metal dross to reduce the temperature. A worker sprinkled water on mineral materials containing arsenic impurities and the mineral materials were then packed and stored in the materials in his house. Thirty-six individuals were exposed to arsine in the house of whom one died and ten were diagnosed with SAAP. The twelfth patient was a gold mine worker who used sulphuric acid and mineral material containing arsenic impurity to refine gold.

All 12 patients were admitted to our hospital within 6 to 9 hours after exposure. Characteristics of the patients are in Table 1. The exposure durations ranged from 15 to 25 minutes and the latency of onset ranged from 30 minutes to 2 hours. Two patients developed multi-organ failure (one patient with acute kidney, liver and heart failure, acute pulmonary edema, and hematemesis; and the other with acute kidney and liver failure and adult respiratory distress syndrome; both patients had large amounts of pleural and peritoneal fluids on ultrasound). Two patients developed hepatomegaly, altered kidney morphology and pleural effusions on ultrasound; and another two patients had hepatomegaly and pleural effusions. All the aforesaid six patients had costovertebral angle tenderness. The remaining six patients had no morphological changes of liver or kidney, and pleural or abdominal fluid was not observed.

Table 1. The characteristics of the patients with SAAP

No.	Sex	Age (year)	Exposure duration (min)	Latency (hour)	Method of Treatment	Times of treatment	Doses of plasma	As in discarded plasma (mg/L)	Total Asdiscarded (mg)	Outcome
1	M	63	20	1.5	PE	1	2000 ml	61.1	122.2	R
2	M	56	20	1.5	PE	1	2000 ml	34.5	69.0	R
3	F	35	15	2	PE	1	2000 ml	83.5	167.0	R
4	F	52	15	2	PE	1	2000 ml	40.7	81.4	R
5	F	67	15	1.5	PE	1	2000 ml	27.7	55.4	R
6	M	50	20	2	PE	1	2000 ml	36.7	73.4	R
7	M	51	15	2	PE	1	2000 ml	61.2	122.4	R
8	F	46	25	1	PE	1	2000 ml	88.7	177.4	R
9	F	40	25	1	PE	2	4000 ml	Fir 60.6 S 13.6	148.4	R
10	F	46	20	2	PE	1	2000 ml	69.1	138.2	R
11	M	23	10	2	PE	1	2000 ml	41.5	83.0	R
12	M	42	15	2	PE	1	1400 ml	55.3	110.6	R

M: male; F: female; As: arsenic; R: recover; PE: plasma exchange; Fir: first time; S: second time.

Eleven patients received PE once, with a total of 1400–2000 mL FFP (1 case 1400 mL, 10 cases 2000 mL), and 1 patient received PE twice for a total of 4000 mL FFP over two days post-exposure. All plasma was provided by the Beijing Red Cross Blood Center and was screened again by the Department of Blood Transfusion of our hospital according to recommendations from the transfusion service.

After 2 to 4 hours of PE, the dark red urine of all the patients became light-colored, the urine of 11 out of the 12 patients turned clear from dark red or brown, and the urine of another one from dark red to mild pink after the first PE, and then to clear after the second PE treatment. Hemolysis of the all patients ended in 24 hours of treatment by clinical judgments, suggesting that PE can rapidly terminate SAAP-induced hemolysis. However, routine urine tests demonstrated that microscopic hematuria might last more than 10 to 14 days in some patients.

Nine of the 12 patients went from oliguria to diuresis within 24 hours post-PE. In one patient, oliguria lasted 13 days before gradually returning to normal. Two patients who received PE shortly after being exposed did not develop oliguria.

Changes of arsenic concentrations in blood and urine

The concentrations of arsenic in both blood and urine decreased significantly at 12h after PE treatment. The mean blood arsenic concentration post-PE (15.86 ± 8.28 mg/L) was significantly lower than pre-PE values (57.84 ± 22.12mg/L, p = 0.0006) and the average urine arsenic concentration post-PE (0.307 ± 0.14 mg/L) significantly lower than pre-PE values (1.092 ± 0.32 mg/L, p = 0.0003). The mean arsenic concentration in discarded plasma was 55.05 ± 19.40 mg/L, with absolute amounts of removed arsenic ranging from 55.4 to 177.4 mg. There was a linear correlation between the concentrations of arsenic in blood and urine (r = 0.718, p = 0.019); however, there were no significant correlations between the concentrations of arsenic in blood or urine and CK, LDH, ALP, ALT, AST, HBDH, TBIL, UBIL, CBIL, BUN, Cr values, including exposure durations (all p > 0.05).

Changes in hemoglobin (Hb) levels and red blood cell counts (RBC) levels (Table 2)

Hemoglobin levels initially tended to decrease quickly in all patients at 12h post-PE (P = 0.001). There were statistically significant changes in hemoglobin levels in 12h, 24h compared to pre-PE values (P < 0.01); however, hemoglobin levels stabilized after 24h post-PE. Like the changes in hemoglobin levels, RBC dropped significantly at 12h of post-PE compared to pre-PE values (P = 0.006) and stabilized after 24h post-PE. In response to the decreases in Hb and RBC, reticulocyte counts increased 3.48% to 11.2% and then slowly decreased as Hb values increased. No significant changes were observed in platelet indices between the pre- and post-treatment values.

Table 2. Changes of blood routine and icteric index (χ ± s)

Time	n	Hb (g/L)	RBC (×10^12/L)	WBC (×10^9/L)	TBIL (umol/L)	UBIL (umol/L)
0	12	113 .09 ± 21.81	3.79 ± 0.43	16.52 ± 5.02	5.20 ± 1.88	4.94 ± 1.90
12h	12	90.181 ± 24.60**	2.95 ± 0.67**	21.04 ± 4.98	4.86 ± 1.96	4.58 ± 2.02
24h	12	74.00 ± 16.68**	2.42 ± 0.54**	18.61 ± 4.47	2.70 ± 1.81**	2.39 ± 1.69**
72h®	10	76.62 ± 15.80**	2.48 ± 0.45**	11.41 ± 2.71**	1.21 ± 0.39**	0.87 ± 0.35**
1w	10	78.40 ± 19.99**	2.51 ± 0.56**	11.11 ± 6.34**	0.92 ± 0.34**	0.72 ± 0.28**
2w	10	80.12 ± 14.35**	2.63 ± 0.61**	8.65 ± 3.08**	0.68 ± 0.36**	0.52 ± 0.19**

The group of pre-PE (0) as the control, *P < 0.05, **P < 0.01.

72h®: two cases were not included because of being infused with 2U red cells.

An initial increase in WBC levels was observed from a pre-PE level of 16.52 ± 5.02 ×10⁹/L to a 12h post-PE value of 21.04 ± 4.98 ×10⁹/L (p = 0.0025), which may be explained by a corresponding increase in the severity of the poisoning. Afterwards, WBC levels decreased slowly until they returned to normal two weeks post-PE. The changes in WBC levels were due to changes in neutrophil counts.

Changes in TBIL and UBIL levels (Table 2)

After PE, the mean TBIL levels gradually decreased such that at 24h post-PE mean TBIL levels were significantly lower than pre-PE values (p = 0.002). Changes in TBIL values were due to changes in UBIL levels. CBIL levels remained constant and in the normal range.

Changes in BUN and Cr levels (Table 3)

Compared to pre-PE values, there was a very slow decrease in BUN levels with no significant changes observed at 12h and 24h post-PE (p = 0.760, p = 0.528, respectively) but at 72h post-PE, the mean BUN value (6.13 ± 4.15 mmol/L) had decreased significantly (p = 0.0022). The Cr levels increased significantly at 24h post-PE in comparison to that of pre-PE (p = 0.012) and remained high for more than two weeks (consistent with the recovery from acute renal failure).

Table 3. Changes of renal, liver function, and the heart-related enzymes on pre- and post-PE (χ ± s)

Time	n	BUN (mmol/L)	Cr (μmol/L)	ALP (IU/L)	AST (IU/L)	CK (IU/L)	LDH (IU/L)
0	12	10.52 ± 2.21	79.63 ± 35.43	62.27 ± 16.04	148.27 ± 84.21	210.45 ± 59.15	2118.73 ± 1393.92
12h	12	10.94 ± 5.43	88.42 ± 44.24	48.45 ± 11.65**	138.27 ± 66.95	247.45 ± 62.34	1510.64 ± 675.05
24h	12	9.22 ± 5.94	106.11 ± 53.04	45.50 ± 7.44**	71.10 ± 36.00	162.20 ± 46.44	1042.30 ± 436.20
72h	12	6.13 ± 4.15**	132.67 ± 50.73**	73.27 ± 17.99	74.18 ± 24.94	51.64 ± 50.16**	615.00 ± 257.39**
1w	10	5.21 ± 2.72**	141.45 ± 48.45**	83.38 ± 17.69	42.75 ± 22.86**	49.57 ± 44.23**	354.14 ± 220.03**
2w	10	5.44 ± 1.62**	112.62 ± 51.23	80.25 ± 21.47	45.25 ± 17.34**	38.46 ± 34.32**	289.26 ± 180.12**

The group of pre-PE (0) is the control, *P < 0.05, **P < 0.01.

Others

There was a slow decline in both the average AST levels and the average LDH levels post-PE. However, no significant changes in the average ALT levels remained within normal limits.

Side effects of treatment and follow-up

In the process of PE, two patients experienced pruritis and rash. All side effects were relieved after treatment with dexamethasone (5 to 10 mg) and had no influence on the therapeutic process. Hemolytic reactions and transfusion reactions were not noted.

All patients treated with PE developed a rapid decrease in Hb concentration due to the dilution by plasma administration. This must be taken seriously and pertinent measures, including the transfusion of packed red blood cells, may be needed.

All the patients treated with PE were successfully treated and discharged from hospital after two to three weeks. Residual symptoms included lethary, weakness, nausea and anorexia. Two to three months after discharge from hospital, all of the patients returned for re-examination which showed normal kidney and liver function and several cases with lethargy, weakness, and pruritis. Unfortunately, blood and urine arsenic concentrations were not measured due to financial constraints.

Discussion

Our study results show that PE may be an effective treatment modality for the patients with SAAP: the concentrations of arsenic in both blood and urine significantly decreased after the PE treatment, the concentrations of arsenic in discarded plasma were 27.7 to 88.7 mg/L and the arsenic discarded by PE totaled 55.4 to 177.4 mg, all the patients treated with PE survived (a welcome contrast to the previous reports of high mortality), PE terminated hemolysis resulting from SAAP within 24 hours, and PE may prevent or ameliorate secondary organ damage from arsine. Additionally, after PE treatment, the enzymes such as CK, LDH, ALP, AST and HBDH decreased rapidly.

Blood exchange plays an important role in the treatment of SAAP and exchange transfusion has been used to support the oxygen-carrying ability of the blood and remove free hemoglobin, arsine and arsenic dihydride residues (1,2,11). Gosselin et al. (19) reported four cases of acute unintentional arsine poisoning with acute intravascular hemolysis developing within a few hours. A total of 8.2 to 10.2 L of blood was exchanged through a continuous perfusion pump at the rate of 1 L/hour. Two patients resumed urine flow during the transfusion, but the other two required repeated HD. Also it is reported that the patients poisoned by mixing a large amount of molten metal dross with wet tin ore (due to rain) were treated with blood exchange transfusion and survived (4).

However, there are obvious side effects to exchange transfusion such as hypocalcemia, metabolic acidosis, seizures, sepsis, necrotizing enterocolitis, cardiac arrest, and death (12). Blood exchange transfusion does not fit with current component blood transfusion opinions (13). So, much more effective methods are needed for SAAP. Our study shows that PE may be an effective approach to the rescue of patients with SAAP. And more importantly, PE may decrease the mortality of SAAP as shown above.

The frequency of PE, the amount of exchanged plasma, and supportive treatments depend clinical judgment until the condition has stabilized and the patients show signs of recovery (e.g., the dark red urine becomes light-colored and Hb concentrations stabilize, suggesting the termination of hemolysis). In our study, 2000 mL (30–35 mL/kg) of FFP (20), (amounting to 40% of total body plasma) was appropriate in the most cases. If needed, another PE is suggested as soon as possible. We suggest that the patients who present with SAAP within 48 hours of exposure and who are undergoing hemolysis should be treated with PE as early as possible.

Uldall reported that it took months for patient with SAAP treated with peritoneal dialysis to recover (21). Among the 14 patients with SAAP and received treatment with HD, peritoneal dialysis or hemoperfusion, eight patients died and only six successfully recovered (9). SAAP patients treated by HD alone either fail to survive or their kidneys take months to recover. The advantages of PE over HD might be the following: PE can non-specifically scavenge exogenous toxins, cytokines and endotoxins, especially those bound to protein (14); but HD removes low molecular weight substances such as BUN and Cr but elevate cytokines (22,23). Another possible reason is that the arsenic is not hemodialysable (7) because it is bound to protein and arsine (arsenic)-hemoglobin-complexes (19,24).

The mechanism of arsine-induced hemolysis is still not clear. Experiments on red blood cells exposed to arsine gas showed that arsine depleted the reduced glutathione content of the red blood cells, resulting in an oxidation of sulfhydryl groups in hemoglobin and red cell membranes, producing membrane instability with rapid intravascular hemolysis (25). Later investigations have proposed a non-oxidant mechanism by suggesting that the target site for arsine is the sodium-potassium pump on the membrane (7,26). Recently, it has been reported that the interaction between arsine and hemoglobin results in an increase in heme release which may contribute to the hemolytic mechanism of arsine (27).

Our study suggests that PE can rapidly terminate the hemolysis resulting from SAAP, even though acute SAAP-induced hemolysis lasts less than 96 h (5). At 2 to 4 hours post-PE, the hematuria/hemoglobinuria in 11 out of 12 patients cleared. After the second PE, the haematuria/hemoglobinuria of another patient ended quickly too, as suggested by stabilization of Hb concentrations. More studies are needed to identify the mechanism by which PE prevents or corrects the hemolysis prevention by PE.

Similarly, the mechanisms of damage of arsine to kidney, liver, heart and brain are not completely understood. It is generally held that the kidney failure in arsine intoxication (due to the effects of free haemoglobin, degradation products and massive hemoglobinuria) may lead to anuria, and if untreated, to death (2,7,10). Toxic pulmonary edema and acute circulatory failure have also been observed, and reported to be the cause of death in arsine poisoning (28,29). Animal experiments and patient autopsies show that arsenic remains in many organs, including the liver, lungs, brain and kidneys (7,30–32). Two patients in our study developed acute respiratory failure. We suppose that the damage to multiple organs might be similar to kidney damage (compromised delivery of blood and oxygen secondary to hemolysis) as the autopsy reports describe a large part of necrosis due to ischemia and hypoxia (32), also due to the direct toxicity of arsine and its metabolic products (33).

Several limitations to our study deserve to be mentioned: no concentrations of arsenic in blood or urine were measured several months after discharge when the patients returned hospital for re-examination. Arsine poisoning can cause long-term damage, such as peripheral sensory-motor neuropathy (34), and the use of a chelating agent such as dimercaptosuccinic ccid (DMSA or succimer) may be considered. However, DMSA may be contraindicated in patients with renal impairment in the period of poisoning (5,11). Moreover, in the follow-up of all the patients, no obvious symptoms and signs about chronic arsenic poisoning were observed. Whether or not it is due to the effects of PE needs further study. Another limitation is lack of the control group, which may limit the generality of our study findings to other patients with SAAP; however, this type of study may be very difficult perform for ethical reasons since both in our experience and in the literature data, patients with SAAP treated by HD or other blood purification methods recovered slowly with a high mortality even up to 55.6% (7–9).

Conclusions

The mechanism underlying the therapeutic success of PE remains speculative, but in part it may be because PE effectively removes the toxins, intravascular erythrocyte fragments and metabolites of arsine, all of which accumulate in blood and cause severe secondary damage to the kidney and other organs. The outcomes of the cases in this report show that, perhaps, PE may have a favorable impact on the clinical course due to its clearance of toxins and the erythrocytes fragments. PE may be an effective method of treating patients with SAAP, taking into account the reduction of the time of recovery and the mortality. If PE is to be used to treat patients with SAAP, it is suggested that treatment be initiated as early as possible.

Conflict of interest statement

Song YG, Wang DX, Li HL, Hao FT, Ma J, and Xa YJ have no financial relationships with any commercial entity that has an interest in the subject of this article.

Acknowledgements

The authors deeply thank the Beijing Municipal Government; the Health Bureau of the Beijing Municipality for their great help when massive patients with poisoning appeared; the Beijing Red-Cross Blood Center for their rapid supply of fresh frozen plasma; and the doctors and nurses from the Department of Emergency, Hematology, Blood Transfusion, Intensive Care Units, and Clinical Toxicology for their hard work in the rescue of the patients.