Is secondary chemical exposure of hospital personnel of clinical importance?

Abstract

Introduction

There is increasing concern among hospital personnel about potential secondary exposure when treating chemically contaminated patients.

Objective

To assess which circumstances and chemicals require the use of Level C Personal Protective Equipment (chemical splash suit and air-purifying respirator), to prevent secondary contamination of hospital personnel treating a chemically contaminated patient.

Methods

The US National Library of Medicine PubMed database was searched for the years 1985 to 2020 utilizing combinations of relevant search terms. This yielded 557 papers which were reviewed by title and abstract. After excluding papers on biological or radiological agents, or those not related to hospital personnel, 38 papers on chemicals remained. After a full-text review, 13 papers without an in-depth discussion on the risk for secondary contamination were omitted, leaving 25 papers for review. The references of these papers were searched and this yielded another seven additional citations, bringing the total to 32 papers.

Incidence of secondary toxicity

Secondary toxicity in hospital personnel is rare: a large-scale inventory of 120,000 chemical incidents identified only nine cases, an occurrence of 0.0075%.

Skin contact as a secondary exposure route

Skin exposure is rare under normal hygienic working conditions, reflected by the very small number of cases reported in the literature: two cases with corrosive effects due to unprotected contact and one case of presumed skin absorption.

Inhalation as a secondary exposure route

Most case reports described secondary toxicity as a result of inhalation. The chemicals involved were irritating solid particles (capsaicin spray/CS), toxic gases formed in the stomach of patients (arsine/hydrazoic acid/phosphine) and vapours from volatile liquids (solvents).

Features of secondary toxicity

Reported symptoms after secondary inhalation were generally mild and reversible (mostly irritation of eyes and respiratory tract, nausea, headache, dizziness/light-headedness) and did not require treatment. In many cases, special circumstances increased exposure: treatment/decontamination of multiple patients, regurgitation of the chemical agent from the stomach, or inadequate room ventilation.

Use of more than standard Personal Protective Equipment

Normal hygienic precautions prevent direct skin contact from exposure to common chemical agents. When solid particle contamination is extensive, a mask and eye protection should be applied. Splash proof outer clothing (splash suit) and eye protection is preferred if (partial) wet decontamination is performed on single patients. Adequate ventilation, careful removal of clothing in case of solid particles contamination and adequate disposal of gastric content reduces exposure. Hospital staff can be rotated if symptoms occur, which can be odour-mediated. The use of more elaborate personal protective equipment with an air-purifying respirator (Level C) is only necessary in exceptional cases of contamination with highly toxic volatile chemicals (e.g., sarin). It should also be considered when decontaminating a large number of patients.

Conclusions

The risk of secondary contamination and subsequent toxicity in hospital personnel decontaminating or treating chemically contaminated patients is small. Normal hygienic precautions (gloves and water-resistant gown) will adequately protect hospital staff when treating the majority of chemically contaminated patients. More extensive protection is only necessary infrequently and there is no reason to delay critical care, even if more elaborate protection is not immediately available.

Keywords:

• Secondary exposure
• secondary contamination
• secondary toxicity
• chemically contaminated patients
• personal protective equipment (PPE)

Introduction

The Dutch Poisons Information Centre has noticed an increase in enquiries from hospital personnel concerned about the risk from secondary exposure when treating chemically contaminated patients. In some cases, disproportionate precautions were taken by Dutch hospitals out of fear for the chemical involved or when minor symptoms were experienced by hospital staff. This ranged from reluctance to admit or treat a contaminated patient to evacuation of the Emergency Department (ED). For example, an ED was evacuated after a physician reported minor respiratory distress when treating an already decontaminated patient who had been dermally exposed to a hydrofluoric acid-containing cleaning product. Firefighters equipped with airway protection took air samples in the ED, and attending personnel underwent a health check [1]. In another case, a paramedic felt unwell after attending a patient who had ingested a paraquat-containing product. Although the Dutch Poisons Information Centre informed the hospital personnel that the risk was negligible, personal protective equipment (PPE), including airway protection, was used at the hospital [1]. Despite these elaborate precautions, several healthcare workers felt unwell after treating the patient, who died 16 h post-ingestion. Additionally, although the deceased was in an airtight body bag, a morgue employee reported health complaints. Clearly, these health effects were a typical stress reaction as a result of “fear for hazardous chemicals”.

Yet enquiries to the Dutch Poisons Information Centre indicate the low incidence of toxicity following secondary exposure. In the last fifteen years only two cases were recorded in which hospital personnel experienced minor health complaints (skin irritation in one case, eye irritation in the other) which could be attributed to secondary exposure.

In this period, concern for secondary exposure was expressed in 56 enquiries of which 43 occurred in the last seven years. This increase in concern over treatment of contaminated patients, is occurring despite, or maybe even as a result of, increased focus on major incident preparedness. An anticipated increase in terrorist attacks has motivated hospitals in the Netherlands to increase preparedness for the admission of patients contaminated with chemical, biological, radiological, or nuclear (CBRN) agents. This led to the extension of emergency response plans, purchase of decontamination facilities and PPE, and exercises with large-scale incident scenarios [2]. What is often missing in these emergency response plans though, is an assessment of the actual health risk for hospital personnel and whether the use of more elaborate PPE (Level C) is necessary in a particular exposure scenario. The focus is mainly on large chemical incidents with multiple casualties while most incidents are minor with only one or a few contaminated casualties.

Extensive precautionary measures are often seen as justified by staff when receiving even a single contaminated patient on the “safety-first principle”. However, this response does not take into account the potentially negative aspects of extensive precautions at the ED: treatment of critically ill patients may be delayed or sub-optimal, and in case of evacuation, healthcare to other patients may be hampered, and financial costs may be substantial.

Three previous reviews on this topic have been published. Clarke et al. [3] described general factors determining the risk and severity of secondary contamination. Hick et al. [4] reviewed some of the published literature on secondary contamination to determine the maximum level of PPE required during hospital decontamination. They concluded that the risks from secondary exposure from contaminated patients at health care facilities was low and could be adequately addressed in most cases by Level C PPE (chemical splash suit and air-purifying respirator). Stewart-Evans et al. [5] used the examples of phosphide, cyanide and sulfite salts to consider secondary risks after ingestion and concluded that secondary contamination is relatively rare but potentially serious.

None of these papers provides an in-depth literature review of cases of secondary contamination and their management.

Objective

To assess which circumstances and chemicals require the use of Level C PPE (chemical splash suit and air-purifying respirator) to prevent secondary contamination of hospital personnel treating a chemically contaminated patient.

Methods

A detailed search of the US National Library of Medicine PubMed database was performed on titles added from 1985 to 2020 using combinations of search terms: secondary AND (contamination OR exposure OR hazard*), contaminated AND (patient* OR victim* OR casualt*), chemically AND (exposed OR contaminated), (nosocomial OR bodies) AND (dangerous OR poisoning OR intoxication OR chemical*), hospital* AND (decontamination OR HAZMAT OR terrorism* OR chemical*), protective equipment AND (decontamination OR contamination). This yielded 557 papers which were reviewed by title and abstract. After excluding papers on biological or radiological agents, or those not related to hospital personnel, 38 papers on chemicals remained. After a full-text review, 13 papers on hospital preparedness/PPE without an in-depth discussion on the risk for secondary toxicity were omitted, leaving 25 papers for review. The references of these papers were checked and this yielded another seven additional citations, bringing the total to be reviewed to 32 papers. These papers mainly consisted of case reports or case series describing secondary inhalational exposure.

Incidence of secondary toxicity

Secondary exposure to a chemical agent can occur during unprotected contact with a contaminated patient. The substance can be transferred by direct skin contact or through inhalation. The subsequent occurrence of secondary toxicity in hospital personnel due to this exposure is a very rare phenomenon.

Horton et al. [6–8] and Larson et al. [9] analysed over 120,000 chemical incidents in the United States from 1996 to 2013. Health effects were experienced by hospital staff in only nine incidents after unprotected treatment of contaminated patients. These case reports are reviewed below.

Skin contact as a secondary exposure route

Secondary skin exposure can occur due to direct chemical contact with the skin or clothing of the patient [6] or contact with their vomitus [10]. In the event of a casualty being exposed to vapour, condensation to skin or clothing can occur but it would be anticipated that a large part of this contamination will have evaporated before arrival at the hospital. This also applies to volatile liquids as it is common practice to remove heavily contaminated clothing before transport to hospital.

Based on the studies of Horton et al. [6–8] and Larson et al. [9] the risk of secondary skin exposure in a hospital setting is expected to be low in the case of common chemicals, since the exposed skin area and exposure time are expected to be limited, if affected skin is washed immediately. But more importantly, it will be prevented altogether if normal hygienic precautions, including gloves and water-resistant gown, are donned by staff when treating/decontaminating a contaminated patient.

Horton et al. [6] described an incident where a container with 3 gallons of hydrofluoric acid (concentration not reported) was disposed in a dumpster. Two sanitation employees were exposed to the agent when the container was compacted in the garbage truck. Both experienced chemical burns and one died of hydrofluoric acid inhalation. The other victim was transported to the hospital without being decontaminated. Two ED personnel not using PPE experienced respiratory and skin irritation.

Five boys spilled 12 ounces (approximately 350 mL) from a small unlabelled bottle that they had found in a field near their trailer park [6]. It was presumed to be a malathion-containing pesticide (no composition details given). Kerosene and tar were also reported as being involved in the exposure. In two children the exposure burned their skin and clothing. They were transported to the hospital without first undergoing decontamination. The father and both sons were treated for respiratory irritation, skin irritation, and gastrointestinal symptoms (nausea and vomiting). While treating the family, two nurses (not wearing PPE), experienced respiratory irritation, eye irritation, skin irritation, gastrointestinal problems, and chemical burns. The ED was then evacuated as a precaution (15 people for 6 h). Since the bottle was unlabelled and the occurrence of chemical burns does not correspond with skin exposure to an organophosphorus insecticide formulation, the actual exposure needs to be questioned. Furthermore, no cholinesterase activities were measured.

Geller et al. [10–12] described a patient contaminated with a veterinary product containing phosmet 11.6%, naphthalene 73%, xylene and surfactant. A nurse suffered from respiratory distress, profuse secretions, emesis, diaphoresis and weakness. Secondary toxicity was probably (partly) due to skin absorption of this agent due to unprotected skin contact with vomitus and respiratory secretions from the patient. This case is discussed in more detail in the paragraph on “Organophosphorus insecticide formulations”.

Inhalation as a secondary exposure route

The majority of victims of large-scale chemical incidents are exposed by inhalation only [13]. There is consensus that gases, such as chlorine and phosgene, will dissipate before the casualty arrives at hospital and not be a secondary contamination risk for hospital staff [3,5,13]. Horton et al. [6] reported an incident in a prison following the mixing of a chlorine bleach with a phosphoric acid-containing product. Twenty-six victims presented at the ED and four ED personnel experienced respiratory/eye/skin irritation, headache and dizziness (“or other CNS symptoms”). It is unclear if contamination with the liquid parent compounds was involved in some of the patients, as an alternative explanation for the secondary exposure.

Based on volatility and evaporation time constants, Georgopoulos et al. [14] assessed the secondary inhalation risk of selected chemical warfare agents and industrial chemicals: sarin, soman, sulfur mustard, tabun, GF (cyclosarin), VX, chlorine, phosgene, hydrogen cyanide, and methyl isocyanate. If an agent is highly volatile, most of it will evaporate before the patient arrives at the hospital (e.g., methylisocyanate, hydrogen cyanide). Agents with low volatility on the other hand (e.g., VX, tabun) will not pose a major inhalational hazard. Taking into account toxicological and chemical properties, the authors [14] identified sarin as a realistic hazard for hospital personnel due to its very high toxicity in combination with an “intermediate volatility”: not so high to evaporate before arrival at the hospital and not so low as not to present an inhalation hazard.

In general, for toxicity to occur to hospital staff by inhalation, the exposure would have to be substantial. The likelihood of this is low since the amount of material a patient carries to the hospital is usually limited.

Based on published case reports and case series, we distinguish the following relevant secondary inhalation hazards: solid particles, gases formed in the stomach of the patients and vapours from volatile liquids. Organophosphorus insecticide formulations are a special case since the active ingredients in such formulations are often erroneously regarded as presenting an inhalation hazard.

Solid particles

CS (o-chlorobenzylidene malononitrile; a “tear gas”) and capsaicin spray (pepper spray) are irritating solid particles in a solvent and are released in aerosol form. Particles can remain on the skin and clothing of a victim after solvent evaporation. Resuspension from a contaminated patient (e.g., when removing clothing), can lead to secondary inhalation exposure.

Horton et al. [8] described three patients exposed to CS who presented at the ED. Their clothes were bagged and quarantined. Two nurses, who were not wearing PPE, experienced mild secondary toxicity (irritation of skin, eyes and respiratory tract) as they assisted with decontamination.

In another paper, Horton et al. [6] reported respiratory and eye irritation in hospital personnel (not wearing PPE) after contact with a single patient who was sprayed in the face with capsaicin spray.

Morcom [15] reported an incident in which 24 patients presented at the ED after they had been exposed to CS in a local night club. Patients with only mild symptoms were treated outside and those with more significant reactions (chest tightness, wheezing, nausea) were taken into the trauma room. Two nurses removed and bagged clothes from more than a dozen victims. After working for 20 min they experienced sore eyes and throat, despite the use of aprons, gloves, masks and eye protection (type not specified). Symptoms were mild as work could be continued with regular breaks every 10 min. Since the PPE should have prevented secondary exposure it might not have been used properly but this is not discussed by the author.

From these published reports, it is clear that irritating solid particles constitute a very limited secondary exposure risk to healthcare personnel compared to direct exposure of these agents during riot-control or when it is directly sprayed into eyes and face.

Gases formed in the stomach

Gases can be formed in the stomach after ingestion of certain agents, due to reaction with water or gastric acid. Phosphine, for example, will be formed after ingestion of metal phosphides. Likewise, sodium azide produces hydrazoic acid and arsenic trioxide produces arsine. Furthermore, cyanide salts produce hydrogen cyanide and sulfite salts produce hydrogen sulfide.

These gases can be released through the oesophagus, or can be present in exhaled air. Exposure risk is expected to increase with the use of a muscle relaxant during intubation, since this also relaxes the oesophageal sphincter. In some of the cases described below, the use of (surgical) masks was reported, but these do not protect from gas exposure.

Phosphine

Three papers [16–18] were identified that reported the occurrence of symptoms due to secondary exposure to phosphine. Musley et al. [16] described the massive fatal ingestion of 750 g (250 pellets) of 60% aluminium phosphide. During the resuscitation process, as the laryngoscope blade was inserted into the mouth, a large quantity of material with an acrid/garlic-like odour (typical for metal phosphides) was returned from the oesophagus. The odour caused immediate pulmonary and ocular distress in the ED staff. After evacuation to the open air ambulance bay, resuscitation efforts continued for another hour before the patient died. Simultaneously, other patients were evacuated from the ED. Despite personnel using gloves, gowns and N95 masks, the treating physician reported transient bronchospasm and a burning sensation in the eyes. A nurse experienced “respiratory distress” over the next day and was briefly hospitalized. It would appear unlikely that her features were directly related to phosphine exposure.

Stewart et al. [17] (see also [19]) described another fatal case where a garlic-like smell was noticed during resuscitation of a patient with a presumed aluminium phosphide ingestion (amount unknown). When the smell spread to other areas, the air conditioning system was turned off and the patient transferred to a “side room”. Two out of five staff members experienced minor symptoms including nausea, dizziness and headache and many staff members were concerned. This led to access to the ED being blocked for 15 h, during which time seven patients were diverted to other hospitals. In addition, the ambulance used to transport the patient required decontamination and was out of use for 11 h. The body was moved to a well-ventilated area where it was placed in two normal polyvinylchloride body bags inside a plastic lined coffin by fire personnel wearing personal protective equipment and cremated. The management of the whole incident took four days.

A very similar case was reported by Nocera et al. [18] (see also [19]) Healthcare personnel attending a patient who had ingested three aluminium phosphide tablets complained of nausea [18] and headache [19] caused by a smell of garlic that permeated from the patient. The ED was evacuated. After the patient died, the body was placed in an encapsulated suit and placed in a “hazardous materials recovery bin”. On the far-reaching precautions to prevent further secondary exposure, Nocera et al. [18] reported: “The patient’s property and clothing were incinerated, and a request from the family to view the body was denied for safety reasons. After four days, the patient’s body within the bin was transported by van directly to a cemetery, immediately placed in a grave and covered using earth-moving equipment. Video footage of the burial was broadcast on national television” [18].

In response to the paper of Nocera et al. [18], Christophers et al. [20] pointed out that no air samples were collected for analysis and only nausea was reported. Furthermore, the authors [20] stated that in India ingestion of aluminium phosphide tablets is a common way to attempt suicide, with an estimated 15,000 cases a year of which two-thirds are fatal. The hospital in the city of Chandigarh, in northern India, known to one of the authors [20] receives about 50 cases a year. Hospital staff do not take any special precautions during resuscitation and treatment of these patients and yet do not develop serious clinical problems. The authors [20] concluded that the secondary exposure risk was overstated by Nocera et al. [18] and continued: “In general, apart from a few highly toxic, mainly anticholinesterase compounds that can be absorbed through the skin (e.g., sarin and tabun), there are no known poisons that will seriously endanger hospital staff routinely caring for patients in an emergency department” [20].

Leenders et al. [21] calculated phosphine concentrations in a standard intensive care unit room resulting from ingestion of two aluminium phosphide tablets. The authors [21] concluded that standard ventilation (100 m³ fresh air/person/hour) is sufficient to decrease the phosphine concentration to a harmless concentration very quickly. This is in line with another incident reported by Musley et al. [16] where gastric lavage and whole bowel irrigation were performed on a patient who ingested zinc phosphide. Eighteen hours post-admission a strong odour was noticed. The patient was transferred to a negative pressure room and phosphine sensors were installed. The only sensor registering phosphine (reading 24 ppm; STEL 1 ppm) was located under the bedsheets near the patient’s rectum. After ventilation of this secluded area, ambient phosphine was no longer detected. Secondary toxicity did not occur during the care for this patient. The calculations [21] and measurements [16] strongly indicate that without vomiting, the release of phosphine from a patient is very limited.

Hydrazoic acid

Abrams et al. [22] reported secondary exposure of hospital personnel to hydrazoic acid due to resuscitation of a patient who had ingested a fatal dose of sodium azide 15–20 g. The authors [22] reported: “During resuscitation efforts, several members of the medical team developed headaches, light-headedness, and nausea. Those affected most were the closest and/or longest in contact with the patient. A few could not participate in the resuscitation as mere presence in the same room elicited these symptoms. Ultimately, it was decided to work in shifts of five to ten minutes to avoid exposure to the azide. There was no obvious pungent smell” [22].

Downes et al. [23] described a patient who had ingested an unknown amount of sodium azide. After arrival at the ED the hazardous materials team was alerted because of concern about toxic vapours. As a result, the ED was closed and the resuscitation team was not permitted access for a period of 30 min. Only after approval, ED staff could continue treatment. The patient died 4 h after admission. The incident was followed up within three months by telephone interviews with involved healthcare personnel to assess their time in close proximity to the patient (within 1 m) and “time off work” as a clear marker of any significant physical complications. Occurrence of minor symptoms was not assessed. Ten staff members without breathing protection were in close contact with the patient for periods ranging from over 60 min (4 members), 15–60 min (3 members) and 5–15 min (3 members). Two staff were absent from work, but neither attributed their absence to the physical effects of the exposure. The authors [23] concluded that sodium azide ingestion does not cause clinically significant secondary toxicity.

This is in line with the experience of the Dutch Poisons Information Centre. In 2017, attention in the Dutch media and public debate on “suicide powders” led to an increase in intoxications with sodium azide. [24] In the period from September 2017 until June 2020 enquiries regarding 13 intoxications (12 fatal) were received (unpublished data). Occurrence of secondary toxicity was not reported in any of these. Two of these patients were treated in our hospital and no special precautions were taken during treatment at the ICU other than mask and eye protection during intubation as part of the COVID-19 protocol then operating. In two other cases, information on the risk of secondary exposure was requested out of concern. These observations support the conclusion of Downes et al. [23].

Arsine

Kinoshita et al. [25] described a patient who ingested 20 g of arsenic trioxide. At the ED he complained of abdominal pain and his breath had a mild garlic odour (known to occur with arsine). After lavage, an abdominal radiograph revealed a lump of arsenic in the stomach which could not be removed by endoscopy because it was attached to the stomach lining. A total gastrectomy was performed. The patient died on the second day of admission. The authors [25] reported the secondary exposure of 22 members of medical staff. The symptoms which occurred after gastric lavage had been performed were generally mild: “eye pain” (13), general fatigue (12), sore throat (11), headache (7), abdominal pain (2) and nausea (2) were reported most commonly. Symptoms were more severe in the physician performing lavage (reddish swollen face but without blister formation) and in personnel performing gastrectomy (corneal epithelial erosions, laryngitis, contact dermatitis). In general, staff closest to the gastric contents had more pronounced symptoms. Information on PPE use was not provided. The authors [25] attributed the secondary toxicity to “a gas causing skin and mucosal irritation”, presumed to be arsine. Since arsine does not irritate tissues [26] the cause for the secondary toxicity remains uncertain. Direct skin contact or inhalation of solid particles of arsenic trioxide could have been involved as this agent is corrosive [27] (clearly not all agent was dissolved into the stomach fluid). Alternatively, the patient might have actually ingested a metal phosphide resulting in secondary exposure to phosphine (which also has a garlic odour).

Hydrogen cyanide and hydrogen sulfide

Stewart-Evans et al. [5] specifically mentioned the risk of secondary exposure to hydrogen cyanide and hydrogen sulfide after ingestion respectively of cyanide salts and sulfite salts.

Burton [28] wrote that the “literature that relates to the hazards posed by chemical contaminants at necropsy focuses primarily on the necropsy of patients who have died from cyanide poisoning” and continues: “However, in two subsequent reported cases of necropsies on cyanide related deaths, no increase in blood cyanide concentrations was detected in the post-mortem workers. However, one pathologist experienced headache and a burning throat sensation and one technician reported light-headedness and throat discomfort” [28]. Assuming similar secondary exposure in healthcare personnel, we estimate the risk of serious toxicity to be minor.

Hydrogen sulfide has an odour of “rotten eggs” and a low odour threshold, below concentrations where inhalational toxicity (such as mucous membrane irritation) is expected, as Stewart-Evans et al. [5] also pointed out, so usually hospital personnel should be alerted before a significant concentration can build up, though as olfactory fatigue may occur smell may not provide adequate warning of hazardous concentrations.

Gases formed in the stomach: Summary

Looking at all cases involving “gases formed in the stomach”, it is remarkable that in four [16–18,23] of six cases [16–18,22,23,25] the ED was evacuated. In two incidents [17,18] the symptoms experienced by hospital staff were very minor (nausea, headache, dizziness) and in one case [16] absent. Concern due to the high toxicity of the agents probably played an important role. In the cases [16,25] with more pronounced but still mild secondary toxicity, a large amount of parent compound returned from the stomach creating a special circumstance.

Stewart-Evans et al. [5] stated that in general: “secondary exposure from a single patient’s exhaled breath or vomitus is likely to be very low”. This is supported by experience in developing countries with a high incidence of such exposures [20], by calculations [21] and measurement of phosphine concentrations [16]. We therefore conclude that gases formed in the stomach can cause mild symptoms but do not present a serious secondary toxicity risk to healthcare personnel if care is taken to quickly and adequately dispose of any returned stomach content to prevent further off-gassing.

Vapours from volatile liquids

Organophosphorus insecticide formulations

Four papers [10,29–31] reported secondary health effects in hospital personnel due to organophosphorus insecticide containing products.

Geller et al. [10–12] described a patient contaminated with a veterinary product containing phosmet 11.6%, naphthalene 73%, xylene and surfactant. The patient was not decontaminated and a chemical odour was noted in the ED. After unprotected skin contact with vomitus and respiratory secretions from the patient, and continued exposure to “chemical fumes”, a nurse suffered from respiratory distress, profuse secretions, emesis, diaphoresis and weakness. She was ventilated for 24 h and treated with atropine and pralidoxime for seven days. Two other nurses without direct skin contact but “sharing the patient’s breathing space” also experienced symptoms. In one nurse, confusion, diaphoresis, hypersalivation, nausea and abdominal cramps were treated with atropine and pralidoxime over 12 h. The other nurse experienced dyspnoea, confusion and headache. Atropine administration did not relieve dyspnoea but invoked hallucinations. Multiple other ED staff developed respiratory irritation, attributed by the authors [12] to an offensive solvent odour.

In response to this paper, Robertson [32] questioned the cause of the nurses’ secondary symptoms since no data on serum cholinesterase activity was presented. More notable in our opinion though is the fact that the patient was not decontaminated and unprotected skin contact occurred in the nurse most severely affected. Toxicity was probably (partly) due to skin absorption of phosmet together with inhalational exposure to volatile ingredients in this formulation. Phosmet itself is non-volatile (very low vapour pressure) and is a solid substance at room temperature, dissolved in organic solvents for application as an insecticide. The extent of secondary skin contamination and the duration of exposure (before washing) is not mentioned. Only inhalational exposure probably occurred in the other two nurses. At least in one nurse the symptoms corresponded with inhalation of organic solvents, supported by the fact that atropine administration did not relieve dyspnoea (and even induced toxicity).

Butera et al. [29] reported secondary exposure to a malathion-containing pesticide product in healthcare workers at the ED. No further details on composition were stated. The patient ingested a large amount of the product and died 45 min after admission, despite cardio-pulmonary resuscitation. The body was left in the ED for 2 h. Health effects reported by 14 healthcare workers included: eye irritation (11), pharyngeal pain (7), nausea (6), lacrimation (5), headache (4), cough (4) and excessive salivation (2). Eight of the healthcare workers cared for or moved the patient, others were in the same room or elsewhere on the ED. Respiratory or skin protective equipment was not worn. Plasma pseudocholinesterase activities were assessed but did not show a reduction. This indicates strongly that the features were not related to malathion exposure.

Merrit et al. [30] described a patient who was saturated from “head to toe” with a pesticide product containing the organophosphate malathion and the solvent toluene (detailed composition not stated). On arrival at the ED, the patient was apnoeic requiring intubation, hypothermic (33.6 °C) and non-responsive to verbal or painful stimuli; pupils were pinpoint and non-responsive to light, heart rate was 39 bpm and blood pressure inaudible (50 mmHg determined by ultrasound). The patient was decontaminated with water. Due to a persistent pungent smell, the patient was treated in an isolation room and the ventilation system to the ED was turned off. Due to the “overpowering fumes”, fans were placed in two doorways to the outside of the hospital and staff were rotated every 20 min. Healthcare personnel used gloves, masks and gowns. Six nurses experienced headache, nausea, vomiting, muscular twitching and fatigue and one nurse with a history of asthma experienced respiratory difficulty. All symptoms were attributed by the authors [30] to solvent exposure, though cholinesterase activities were not measured.

Stacey et al. [31] reported the clinical course of a 45-year-old man who ingested a demeton-s-methyl-containing pesticide (detailed composition unknown). The patient was hypersalivating, his heart rate was 60 bpm, his blood pressure was 110/67 mmHg, his pupils were constricted and the Glasgow Coma score was 3/15. He was intubated and ventilated by the paramedic team. No decontamination was undertaken at the scene even though the patient’s clothing was contaminated with vomitus. An hour later the patient was decontaminated at the ED by healthcare personnel wearing PPE (gown, face mask with visor, gown, gloves). The authors [31] explained that full protective suits with self-contained ventilation were used by the attending medical doctors during resuscitation but proved too cumbersome to allow medical procedures to be carried out. Several staff members started experiencing symptoms (chest tightness and light-headedness) and the ED was closed until the next morning. Twenty-five healthcare workers who had been in contact with the patient were examined. Ten reported health effects: chest tightness (7), light-headedness (1), hypersalivation (1) and giddiness (1). Two were observed for up to 4 h and eight were discharged after examination. None was judged to have been poisoned and treatment was not necessary.

In a response to this paper, Roberts et al. [33] emphasized that cholinesterase activity was not measured in the staff and the symptoms described could be consistent with solvent exposure or anxiety. They explained that the organophosphorus insecticides present in pesticide formulations have a low volatility, so these agents (if not aerosolized) have a low risk of inhalational toxicity. The authors [33] explained that they had treated >700 patients with organophosphorus insecticide poisoning over 2 years at Anuradhapura General Hospital, Sri Lanka. During this time, despite frequent medical review of these patients, only brief episodes of nausea and dyspnoea were reported by doctors. These reactions were infrequent, followed close contact with severely poisoned patients, and symptoms settled quickly with fresh air. Roberts et al. [33] concluded that inhalation of solvent from the breath or clothes of a patient is the most likely cause for these effects. “Hospital wards in Sri Lanka are largely open, which may improve ventilation and contribute to the apparent lack of secondary poisoning of hospital staff. However, we are not aware of reports of more severe toxicity in relatives or ambulance officers who transfer patients. We have not measured cholinesterase activity in staff with close contact because there has never been a clinical or other indication to do so” [33].

In a “consensus statement”, Little at al. [34] indicated that in the published case reports, cholinesterase activity was either not performed [10,31] or showed activities in the normal range [29] and symptoms could be attributed to solvent exposure. On the only serious case, reported by Geller et al. [10] (described in detail above) the authors [34] further remarked that the patient was not decontaminated and staff did not wear any skin protective clothing. Little et al. [34] continued: “Nosocomial OP [organophosphate] poisoning has not been reported from those developing countries with a high incidence of self-poisoning with OPs [organophosphates]. Health workers in those countries do not use PPE and do not perceive themselves at risk. The combined experience of all the authors includes treatment of over 100 patients with OP [organophosphate] poisoning and no similar cases have been observed and nor are we aware of any other cases manifesting either clear evidence of cholinergic toxicity or requiring intubation”. The authors [34] concluded that there is little evidence that hospital personnel are at any risk of developing secondary toxicity by treating an organophosphorus pesticide poisoned patient under normal circumstances.

Taking into account the low volatility of the organophosphorus insecticide present in pesticide formulations, the limited case reports on secondary exposure and the stark contrast with general experience in developing countries, we conclude that a patient contaminated with an organophosphorus insecticide formulation constitutes a limited secondary inhalation risk to hospital staff. In the absence of direct skin contact, symptoms can be attributed to solvent exposure which is expected to be mild and self-limiting in a hospital setting (also see paragraph on solvents below).

Solvents

Common volatile chemicals in consumer or professional products are hydrocarbon solvents (e.g., toluene, xylene). Secondary contamination can occur for example due to spilling or vomiting after ingestion.

Minor symptoms can occur as a result of secondary exposure to solvents, as is apparent from case reports on contamination with solvent containing insecticide formulations [29–31] (see above) and from incidents with metamfetamine production chemicals [6,7,35] that can include solvents (see below).

Only one paper was found that briefly described a single exposure to a solvent as such. Burgess et al. [35] reported a case of a patient presenting at the ED with a laceration. His clothes emitted a pungent solvent odour due to working with an aliphatic hydrocarbon liquid mixture. Four healthcare workers experienced mucous membrane irritation and headache. The ED was evacuated for 2 h.

Schultz et al. [36] assessed the secondary exposure hazard of solvents. Clothes on a plastic mannequin were drenched with 800 mL of either acetone or p-xylene as a “worst case” contamination. Real-time air sampling showed that in a non-ventilated room at 18 °C, breathing-zone concentrations during a 10 min decontamination did not pose a respiratory risk. The short-term exposure limit (STEL) was not exceeded. Although exposure time might be longer in a hospital setting, we conclude that occurrence of serious secondary toxicity is unlikely and more importantly that frequently occurring minor contaminations due to spilling or vomiting after ingestion of solvent-based products will not present a clinically significant risk to hospital personnel. Without excluding minor direct effects due to low ambient solvent concentrations, it should be recognized though that the pungent smell of some chemicals could cause an odour-mediated response in concentrations far below commonly accepted toxic concentrations, as Burgess et al. [13] has pointed out.

Acids

Acids are common ingredients in consumer/professional products and treatment of exposed patients is very common, generally not resulting in secondary toxicity. Extensive external contamination to highly concentrated formulations can occur in the working environment or in met - amfetamine production (described in the next paragraph). Without adequate use of PPE, a direct corrosive effect on the skin would be expected. Risk of secondary inhalation exposure would depend on the volatility of the specific acid.

Larson et al. [9] described an incident where a worker cleaned rust from the inside of a railroad tank (used to transport sulfuric acid) with a phosphoric acid containing product. Upon re-entering the tank without breathing protection the man collapsed. He was decontaminated at the ED. Eight of eleven potentially exposed ED personnel experienced nausea, vomiting, eye irritation, unspecified respiratory symptoms, headache, dizziness or other CNS symptoms. A strong odour was reported and attributed to sulfuric acid. This is remarkable since phosphoric acid and sulfuric acid have low volatilities (especially sulfuric acid), limiting the buildup of hazardous concentrations by evaporation in case of limited spills/contaminations. The exact nature of the exposure therefore needs to be questioned.

Metamfetamine production chemicals

Four secondary exposures to metamfetamine production chemicals (including solvents, acids and bases) have been reported in the literature [6,7,35].

Horton et al. [6] described two incidents. The first incident occurred when an oven exploded as two persons were using acetone, hydrochloric acid and sodium hydroxide to manufacture metamfetamine. One “cooker” presented himself at the ED without decontamination. Three healthcare personnel (not wearing PPE) experienced nausea and vomiting. The ED was evacuated for 6 h. The second incident was very similar: an explosion and fire occurred in an apartment where a “cooker” used 5 gallons of various acids and solvents (not specified further). One ED employee, not wearing PPE, experienced dizziness (“or other central nervous system symptoms”) when treating the patient who was not yet decontaminated.

Horton et al. [7] in another paper, briefly mentioned metamfetamine chemicals (not further specified) as a cause of secondary toxicity (respiratory irritation, headache) in five hospital personnel not wearing PPE.

Burgess [35] described a patient that presented at the ED with burns. Initially reported as battery acid exposure, it was later determined that the patient was a “cooker” from a metamfetamine laboratory that had caught fire. Four hospital staff members experienced nausea and mucous membrane irritation. The symptoms resolved within 24 h in two members and lasted more than a day in the other two. The hospital was evacuated for 7.5 h.

The symptoms in these cases corresponded with inhalational exposure to solvents, especially the central nervous system effects. Mucous membrane irritation can also be caused by acids or alkaline agents (depending on their volatility). At least, it can be concluded that serious secondary toxicity did not occur.

Sarin

In 1995 in Tokyo, a terrorist attack on the subway potentially exposed some 5,500 people to sarin. Okumura et al. [37] and Nozaki et al. [38] described secondary exposure to hospital personnel. A nearby hospital received 640 patients, which could not all be decontaminated due to limited capacity [37]. Hospital personnel wore standard working clothes with gloves and surgical masks. In retrospect, 110 out of 1063 hospital staff reported symptoms: eye symptoms (66), headache (52), throat pain (39), dyspnoea (25), nausea (14), dizziness (12), nose pain (11). Symptoms were mild and did not require treatment. The authors [37] attributed the secondary exposure to poor ventilation and lack of a decontamination area. Many patients were treated in the poorly ventilated hospital chapel. Almost half of medical staff working there reported symptoms (38 of 83). One nurse who had treated patients all day at this location was admitted with nausea, headache and shortness of breath (no antidote treatment). Incidence of secondary exposure on the ED on the other hand was relatively low (8 of 48), probably due to better ventilation (entrance wide open).

At another hospital, thirteen out of fifteen doctors experienced symptoms: dim vision (11), rhinorrhoea (8), dyspnoea or chest tightness (4), cough (2), salivation (1) and sore throat (1) [38]. Most doctors with severe miosis (6 of 8), were directly involved in resuscitation, intubation or decontamination of two severely exposed patients. Six doctors received atropine (one in combination with pralidoxime). In two doctors, the cholinesterase activity was determined. The values were within normal range. All doctors could continue to treat patients. Ventilation was improved by opening doors and windows and patients’ belongings were bagged/sealed and stored outside of the ED.

Saunders et al. [39] cited this terrorist attack as an example of “the potentially devastating effect of chemically contaminated patients” to advocate the need for hospital preparedness. Hick et al. [4] on the other hand focussed on the fact that “all [exposed physicians] continued their patient care duties essentially uninterrupted”. From our point of view, the Tokyo sarin attack illustrates that even highly toxic substances are unlikely to cause serious secondary toxicity in hospital personnel, even if a large number of patients are involved. Of course, elaborate PPE is appropriate in such a rare event but the lesson to be learned here is that critical care must not be delayed, even if full PPE is not available immediately.

Features of secondary toxicity

The limited number of case reports [5–9,15,16,18,22,25, 29–31,35] indicate that if secondary exposure does occur, the symptoms are generally mild and reversible and do not require treatment. Most commonly reported are irritation of eyes and respiratory tract, nausea, headache and dizziness/light-headedness after inhalation exposure. The only serious case involved direct unprotected skin contact [10] and is much debated [32–34]. Overall, the reported symptoms correspond with limited inhalational exposure to the agents involved but are rather unspecific nonetheless, so anxiety and odour-mediated responses could have played an additional role in some cases.

In many cases of secondary exposure, special circumstances increased inhalational exposure. In two of three secondary exposure cases to irritating solid particles, multiple patients were involved [8,15]. For “gases formed in the stomach”, in two [16,25] out of five cases the parent compound returned from the stomach (laryngoscopy, gastric lavage/gastrectomy). In the three cases [5,18,22] where this did not occur the symptoms were especially minor. From all included case reports, two cases [17,30] reported turning off the ventilation system and transferring the patient to a separate room. In one of these cases the authors specifically mention this created a “poor working environment” [30]. Poor ventilation was also an important factor in secondary exposure of hospital personnel to sarin after the 1995 Tokyo attack [37]. For the other cases, it is not clear from the reports if ventilation was inadequate but it could have been an important factor.

When to use more than standard PPE

We believe that the available literature is sufficient to assess the risk of secondary toxicity and to determine which circumstances justify the use of more extensive PPE. Based on the literature we conclude that normal hygienic precautions, including nitrile gloves and water-resistant gown covering arms/shoulders/torso, will protect hospital staff adequately when treating a single patient contaminated with common chemical agents. These include patients contaminated due to spilling or vomiting after ingestion of a consumer/professional product with volatile ingredients (including solvents in pesticides), and patients that release toxic gases in expired air. It is assumed that direct mouth-to-mouth resuscitation will not be performed in the hospital.

Since minor symptoms are occasionally reported with these agents, probably due to special circumstances increasing exposure, we recommend the following additional precautions. In case of solid particle contamination, clothing should be removed and bagged carefully to prevent resuspension. In case of gases formed in the stomach or contamination with volatile liquids, the patient should not be isolated in a poorly ventilated room. Vomit or gastric content after lavage should immediately be disposed of adequately to prevent further off-gassing. After direct unprotected skin contact with a chemical agent, the contact area should be washed immediately with plenty of water and mild soap. Whenever health effects are experienced, hospital staff can be rotated. In some of the case reports, this was reported every 5–10 min [15], every 10 min [22] or every 20 min [30].

For single contaminated patients, we recommend a mask (surgical, but preferably P1/P2) and eye protection (safety glasses/goggles) when solid particle contamination is extensive. Furthermore, if (partial) wet decontamination is performed on single patients, splash proof outer clothing (splash suit) and eye protection to protect from splashes are preferred.

Additional protection against inhalational exposure is only necessary in rare circumstances, such as extensive contamination with a highly toxic volatile fluid (such as sarin) when contaminated clothing has not been removed. In addition, in the case of less hazardous substances but a large number of patients, Level C PPE during decontamination is advisable, because of the increased exposure time. These scenarios are most likely to happen as a result of a large-scale industrial incident, transportation incident, or terrorist attack. Level C protection consists of an air-purifying respirator, splash suit, eye protection, nitrile and/or butyl gloves and boots [4,14,40]. The respirator makes use of filters, cartridges or canisters that adsorb airborne agents. Hick et al. [4] advised a combination of at least organic vapour (e.g., solvent exposure) and high-efficiency particulate air (HEPA) filters. Additional “acid gas” filters could be incorporated for increased protection [4]. In a hospital setting Level C protection will suffice and an encapsulating suit with supplied air respirator or self-contained breathing apparatus (Level B) is not necessary [4,14,40].

Conclusions

Based on the information in this review, we conclude that secondary exposure risk for hospital personnel treating a chemically contaminated patient is low when skin contact is prevented by applying normal hygienic precautions, including gloves and water-resistant gown. These precautions will suffice for single patients contaminated with common chemicals. When solid particle contamination is extensive, a mask and eye protection can be applied. Splash proof outer clothing (splash suit) and eye protection is preferred if (partial) wet decontamination is performed on single patients.

The use of more elaborate personal protective equipment (Level C) is only necessary in rare circumstances, such as an extensive contamination with a highly toxic volatile fluid (such as sarin). It can also be considered in case of a large number of patients. Since only mild secondary toxicity is expected in a worst case scenario, critical care must never be delayed, even if the protection of the healthcare workers is not (yet) optimal.

Acknowledgements

The authors thank Professor Allister Vale for his invaluable contribution to improve this review.

Disclosure statement

The authors report no conflict of interest.