Extracorporeal membrane oxygenation support for tropical infections: a scoping review

Abstract

Background

Extracorporeal membrane oxygenation (ECMO) for infectious causes of refractory cardiopulmonary failure is established as appropriate therapy in high-income countries. Its use in low- and middle-income nations for tropical infections is not well-studied, however, perhaps because most of these countries have not been historically able to offer ECMO support. Tropical infections remain an important cause of global morbidity and mortality, but the role of ECMO is poorly described.

Methods

We identified a list of viral, bacterial, fungal, and parasitic infections that qualified as tropical infectious diseases. These included infections that were either a World Health Organization (WHO) designated Neglected Tropical Disease (NTD) or an infectious disease with a higher prevalence in the Tropics than elsewhere. We conducted a comprehensive review of existing literature regarding ECMO use to support these infections.

Results

Multiple viral, bacterial, fungal, and parasitic tropical infections have been supported by ECMO with varying success. leptospirosis, melioidosis, and tuberculosis are conditions suitable for venovenous ECMO support with frequent use in the literature and reported survival as high as 84% in Leptospirosis. Venoarterial ECMO has been successfully used in American trypanosomiasis-related cardiogenic shock as a bridge to transplant, with one center reporting a 71% survival rate. Dengue and malaria have been successfully supported with both venovenous and venoarterial ECMO. Mortality is relatively high (> 65%) in patients who receive ECMO for Middle Eastern Respiratory Syndrome. ECMO has also supported Echinococcal infections perioperatively. There are multiple tropical infections where ECMO use has not been published.

Conclusion

As the use of ECMO expands globally, more patients with tropical infections may require ECMO in both endemic and non-endemic settings. We present a scoping review on the evidence base of ECMO use to support tropical infectious diseases, with data regarding feasibility in specific disease processes as well as clinical considerations for tropical diseases on the ECMO circuit.

Background

The use of extracorporeal membrane oxygenation (ECMO) for refractory cardiac and pulmonary failure continues to increase throughout the world, with clinical studies demonstrating benefit, as well as observational data supporting its use in various emerging infectious disease outbreaks [1–5]. ECMO has successfully supported a variety of infections, including venovenous (VV) configurations for acute respiratory distress syndrome (ARDS) [1, 6] as well as in the venoarterial (VA) configuration for cardiomyopathy secondary to sepsis [7, 8]. As indications for ECMO expand and its use becomes more widespread, clinicians are more frequently faced with decisions regarding ECMO candidacy for specific diseases with which they have minimal experience and limited published data to inform them. One example of this is tropical infections, which provide a unique challenge because these diseases are generally found in low- and middle-income settings, where the use of ECMO is limited to larger centers, if available at all.

Clinicians managing ECMO patients should understand tropical infections, regardless of their practice location. While there have been international efforts to combat tropical infections, changes in both urbanization and human interaction with the environment have led to reemergence of various tropical infections as well as expansion of tropical infections beyond historical borders [9, 10]. Tropical infections are a frequent cause of admission to intensive care units in low- and middle-income countries [11]. Historically, ECMO was unavailable in most countries located in the tropics and, while this remains true throughout much of equatorial Africa, ECMO use in middle-income countries in Latin America and Asia has increased considerably over the last two decades [12]. With the increasing use of ECMO in the tropics, more clinicians in tropical countries will be consulted for cannulation decisions. Even in regions which are not considered tropical, increases in global travel may lead to more returning travelers with life-threatening tropical infections outside of areas where these diseases are endemic.

With the continued threat of tropical infections, clinicians face challenges determining the potential utility of ECMO support for various tropical infections. In this scoping review, we describe published outcomes of adult and pediatric patients with tropical infections who have been supported on ECMO and discuss what considerations may be relevant to clinicians in cases where there is no published or minimal evidence.

Methods

A literature review was performed on diseases that the authors determined to be tropical infections. To meet these criteria, a tropical infection was regarded as either a World Health Organization (WHO) designated Neglected Tropical Disease (NTD) or an infectious disease which has demonstrably higher prevalence between the Tropics of Cancer and Capricorn. A literature search was performed using PubMed and Google Scholar to find peer-reviewed manuscripts published in English from 1988 to 2024 which described patient outcomes for the tropical diseases previously identified. Search terms included “(ECMO) AND [each identified infection]”. The type of studies evaluated included case reports, case series, retrospective cohort studies, and meta-analyses. Variables were extracted from each article by the investigators in regard to countries involved, disease characteristics, patient demographics (age, comorbidities), and ECMO support (mode, duration, complications). Mortality was defined as any patient who did not survive to hospital discharge. For diseases in which there were no published ECMO data available, we considered the pathophysiology of each organism’s disease process to identify tropical diseases that might conceivably respond to ECMO support.

Tropical infections with no reported ECMO use

Many tropical infections have no published data describing outcomes for patients receiving ECMO because they mostly cause disfigurement and other infectious morbidity, rather than severe cardiopulmonary failure. Notable NTDs in which ECMO likely would not be considered for these reasons include Buruli ulcer, Chromoblastomycosis, Dracunculiasis, Hookworm, Leishmaniasis (both cutaneous and visceral), Leprosy, Lymphatic filariasis, Mycetoma, Noma, Onchocerciasis, Scabies, Taeniasis, Trachoma, and Trichuriasis. ECMO is also unlikely to be helpful in tropical infections that cause multiorgan failure, hypovolemic shock, or neurologic dysfunction such as severe encephalitis. Viral hemorrhagic fevers (e.g., Ebola, Congo-Crimean Hemorrhagic Fever, and Marburg) often cause hypovolemic shock where ECMO is unlikely to be effective [13]. Yellow fever has the potential for secondary pulmonary infections and cardiac dysfunction, but ECMO is unlikely to alter outcomes in cases of severe multiorgan failure due to prominent encephalitis symptoms. Both rabies and African trypanosomiasis can have severe extra-neural manifestations such as acute respiratory distress syndrome (ARDS) and myocardial dysfunction [14–16], but they also more commonly cause severe encephalitis and cerebral edema, in which case ECMO is unlikely to change the natural history of the disease.

There are multiple NTDs that may cause severe cardiopulmonary disease for which ECMO support has not been documented but theoretically could benefit from ECMO in specific situations. The utility of ECMO support is likely highest in these diseases when there is primary cardiac or respiratory failure. Diseases which may have potential benefit from ECMO support but with no reported cases of ECMO use include Chikungunya, Paracoccidioidomycosis, Paragonimiasis, Schistosomiasis, and Trematodiases. Chikungunya is an arbovirus that can occasionally lead to either respiratory failure or myocarditis. Parasitic infections such as Schistosomiasis and more rarely Clonorchis are causes of pulmonary arterial hypertension in tropical settings in which venoarterial (VA) or venovenous (VV) ECMO may be considered for support of severe right heart failure as a bridge to initiation of directed therapy, recovery, or transplant. Paragonimiasis is a food-borne zoonosis that commonly causes lung and pleural disease in which VV ECMO may be considered as well. Paracoccidioidomycosis is a fungal infection endemic to the tropics that rarely causes acute lung injury, and VV ECMO might be a reasonable option for these patients. These tropical infections where ECMO has not been described remain an area of uncertainty. Future efforts to characterize the outcomes of patients with these infections with severe cardiac or respiratory failure supported by ECMO should be further explored.

Bacterial infections

Leptospirosis

Leptospirosis is a zoonotic bacterial disease caused by the bacteria genus Leptospira. Transmission occurs when humans have contact with or ingestion of water or soil contaminated by rodent urine (Table 1) [17]. Severe manifestations of leptospirosis include refractory ARDS (with or without pulmonary hemorrhage) and Weil’s disease with the triad of respiratory failure, jaundice, and acute kidney injury. Both conditions have been supported with ECMO. The mainstays of therapy include penicillin or doxycycline, where standard doses are generally used on ECMO [18].

Table 1.

Tropical bacterial diseases with published ECMO data

Disease	NTD?	Countries	Published ECMO cases	ECMO mode	Age (median [IQR])	Survival?	Evidence	Comments
Leptospirosis	Yes	Brazil, Canada, Chile, China, Croatia, France, Germany, India, Italy, Japan, Netherlands, Philippines, Singapore, Sri Lanka, Taiwan, Thailand, United Kingdom, United States	89	VV VA	37.5 [32–48.5]*	84%	Abstract Case reports Case series	90% male, Majority no comorbidities, 74% Renal Replacement Therapy, Hemorrhagic complications, Elevated Liver enzymes Pancreatitis
Melioidosis	No	Australia, China, Saudi Arabia, Singapore	4	VV VA	16 [6–29]	50%	Case reports	Neutropenia, Multiorgan dysfunction
Q fever	No	Australia,	1	VA	32	100%	Case report	Infective endocarditis, Perioperative
Scrub typhus	No	China, South Korea	5	VV VA	67 [40.5–74.5]	100%	Case reports	Multiorgan dysfunction
Tuberculosis	No	Belgium, Canada, Chile, China, France, Germany, Iceland, India, Italy, Japan, South Korea, Switzerland, United Kingdom, United States, Vietnam	169	VV VA	39 [28–48]*	51%	Case reports Retrospective Study Systematic Review ELSO Registry	PK/PD of TB drugs is uncertain
Vibrio vulnificus	No	South Korea	1	VA	52	0%	Case reports	Bridge to lower extremity amputation for necrotizing fasciitis

^* = age not fully available for all cited cases of Leptospirosis requiring ECMO, age data listed are from cases with age reported

^** = age not fully available for all cited cases of Tuberculosis requiring ECMO, age data listed are from largest adult cohort of patients

IQR interquartile range, VA venoarterial, VV venovenous

The 89 published cases in which ECMO was used to support leptospirosis comprised 26 different case reports and case series [19–43]. A recently described cohort of 47 patients was supported with VV ECMO at a single center in the Philippines [44]. Duration of ECMO in these patients ranged from 17 h to 37 days. Pulmonary hemorrhage and severe thrombocytopenia were significant disease complications pertinent to ECMO management, and multiple reports have described challenges with recurrent airway bleeding while treating leptospirosis on ECMO. Despite these bleeding tendencies, multiple anticoagulation strategies have been employed. Some authors used heparin with a partial thromboplastin time goal of 40–60 s [25, 28, 31], one reported success with an activated clotting time between 160 and 180 s [27] and one other successfully targeted anti-Xa range 0.2–0.3 IU/L [37]. Weil’s disease may lead to liver impairment and potentially acute liver failure, which could further complicate ECMO management. Elevated liver-associated enzymes were common in this cohort of patients but were not associated with worse outcomes. Other complications, such as acute pancreatitis, have also been reported in patients receiving VV ECMO [41].

Melioidosis

Melioidosis is an infection caused by the Gram-negative bacterium Burkholderia pseudomallei that is a leading cause of severe pneumonia and sepsis in the tropics, predominantly Southeast Asia and northern Australia. More recently, it has been found in Central and South America [45] and its distribution is likely more global than was previously described. The pulmonary disease can resemble tuberculosis with an upper lobe predominance and focal bronchopneumonia, a cavitary process, or a diffuse miliary pattern [46]. The diagnosis of melioidosis is challenging because it requires culture of the organism, which may be unavailable in many tropical locations or centers unfamiliar with this soil saprophyte [47]. B. pseudomallei is intrinsically resistant to many first-line empiric antibiotics, including penicillin, ampicillin, first and second generation cephalosporins. Treatment of melioidosis is ceftazidime or a carbapenem for those with critical illness [43, 48–50], for which standard doses can be used for patients receiving ECMO. However, melioidosis can also cause central nervous system (CNS) abscesses, in which case higher doses should be used. Both VA and VV configurations of ECMO have been employed for support in melioidosis. There are 4 published reports of ECMO support for melioidosis [48, 51–53]. Two patients died of multiorgan failure, and two were successfully supported with ECMO. All patients eventually were treated with either ceftazidime, meropenem, or both. These four cases of ECMO support in melioidosis highlight the fulminant nature of this condition and its potential for multiorgan involvement.

Q fever

Q fever is a disease caused by the intracellular Gram-negative bacteria Coxiella burnetii [54]. Q fever can cause acute pneumonia and a chronic infection that leads to infective endocarditis. While this disease has a worldwide distribution, it is thought to be emerging in certain tropical regions [55–57]. For example, the annual incidence in French Guiana (37/100,000) was higher than that seen in Europe, the United States, or even Australia (2.1/100,000) [58, 59].

ECMO has been implemented once in the literature to support a critically ill patient with Q fever in the setting of infective endocarditis. A 32-year-old Australian male received VA ECMO support post-operatively for septic shock and high-output cardiac failure after aortic valve repair for infective endocarditis due to chronic Q fever [60]. This case highlights the use of ECMO to support cardiac complications in chronic Q fever, but there is no data on the use of ECMO in acute Q fever.

Scrub typhus

Scrub typhus is a rickettsial illness caused by the arthropod-borne Gram-negative bacillus Orientia tsutsugamushi [61] and predominantly affects the Asia-Pacific region [62]. Global travel has resulted in its occurrence in areas where it is not endemic. Scrub typhus is transmitted by the Leptotrombidium mite, and shortly after, an eschar develops. It commonly presents as undifferentiated fever but can cause ARDS, myocarditis, septic shock, disseminated intravascular coagulopathy, and renal failure. Treatment is often delayed due to difficulty in diagnosing the condition. The diagnosis is often missed because rapid point-of-care testing is not widely available, and the diagnosis relies on the Weil–Felix test, which may take > 48 h to process and has poor sensitivity and specificity [63]. Treatment includes doxycycline or azithromycin, which are given in standard doses on ECMO [64].

The available evidence regarding ECMO use in support of scrub typhus comes from 5 case reports in returning travelers to South Korea and China [65–69]. Four of these cases were treated with VV ECMO for ARDS and one was treated with VA for myocarditis. The four cases of VV ECMO used to support scrub typhus were in adults (age 20–81 years old). All patients survived to discharge with ECMO courses that lasted less than 2 weeks [65–68]. Extracorporeal cardiopulmonary resuscitation (ECPR) was used in a 61-year-old Korean male for biopsy-proven myocarditis. The patient survived to discharge after 3 days of VA ECMO [69].

Tuberculosis

Tuberculosis (TB) is a disease caused by the bacterium Mycobacterium tuberculosis with a significant disease burden in the tropics [70]. Furthermore, there is a syndemic between TB and human immunodeficiency virus (HIV), making severe infections more common in countries with a high burden of HIV, which are also more common in the tropics [71]. TB can progress to respiratory failure and ARDS. TB comprises 2–4.8% of ARDS cases, even in highly TB-prevalent countries such as South Africa and India [72]. Given the rarity of ARDS due to TB, there is a paucity of data in regard to best management. There have been several studies evaluating the utility of ECMO for patients with respiratory failure due to TB. While there are a variety of medications used for TB, there is limited data on the pharmacodynamic and pharmacokinetic effects of the ECMO circuit on many anti-tuberculosis drugs. Antimicrobials used in various regimens, including amikacin, imipenem, and levofloxacin, suggest no adjustment is required, but linezolid may require higher doses to reach therapeutic levels [18].

A retrospective analysis utilizing the extracorporeal life support organization (ELSO) registry database [73], one international retrospective cohort study [74] and one systematic review and meta-analysis [75] provide information regarding ECMO support for critically ill TB patients. An analysis of the ELSO registry included 169 patients with TB requiring ECMO support, with successful weaning in 63% and survival to discharge in 55%. Idris et al. reviewed 43 patients from 1975 to 2022 who received ECMO from 15 countries with 37 VV ECMO, 5 VA ECMO, and 4 patients who had their configuration changed during the treatment course. They described a survival of 81% in a very heterogeneous group which included newborns, infants, children, and adults with both cardiac and pulmonary failure. A second cohort described by the TB ECMO study group focused on adult patients from 2002 to 2022 in low prevalence countries and found older age, higher pre-ECMO SOFA scores, drug-resistant TB, and cavitary TB (56% vs 28% in miliary TB) as significant predictors of 90-day mortality. Seven out of eight patients with drug-resistant TB did not survive to ECMO decannulation. The total mortality in this cohort of patients was 51%.

Vibrio vulnificus

Vibrio vulnificus is a Gram-negative aquatic bacteria that causes infections in humans from either consumption of contaminated seafood or salt-water exposure to an open wound [76]. Vibrio thrives in warm waters globally and is commonly found in tropical seas. Clinical manifestations include primary bacteremia, wound infection, and diarrhea. Treatment includes a third-generation cephalosporin plus a tetracycline, which are dosed similarly to other critically ill patients in patients receiving ECMO. Early surgical consultation is recommended for serious wound infections. Only one case of VA ECMO used to support Vibrio infection has been published to date. VA ECMO was used for septic cardiomyopathy due to necrotizing fasciitis from a severe infection [77]. ECMO was used as a bridge to lower extremity amputation. ECMO was weaned with improved cardiac function after 15 days. A second ECMO run was employed after this same patient developed refractory septic shock, but the patient died from bacteremia. While not a classic form of cardiopulmonary failure, this demonstrates possible utility for ECMO in severe Vibrio infections that lead to cardiac failure.

Viral infections

Dengue

Dengue is a viral NTD transmitted by the Aedes mosquito (Ae. aegypti, Ae. albopictus) endemic to many tropical countries (Table 2). While there are several vaccines around the world to prevent severe dengue, especially in children, there were over 3 million cases and 1,400 deaths in the first six months of 2025 around the world [78]. Vaccines against the four major serotypes of dengue are available for both those with and without previous infection. Reinfection is possible with dengue due to a different serotype and increases the risk for severe disease manifestations such as dengue shock syndrome due to antibody-dependent enhancement. There are no established anti-viral treatments available for dengue, and treatment is supportive. In severe dengue, patients may develop respiratory failure progressing to ARDS or myocarditis. For both of these complications, there is a potential role for ECMO support.

Table 2.

Tropical viral diseases with published ECMO evidence

Disease	NTD?	Countries	Published ECMO cases	ECMO mode (n)	Age (median [IQR])	Survival?	Evidence	Comments
Dengue	Yes	Colombia, France, India, Taiwan	11	VA VV	31 [11–52]*	54%	Case reports Case series	Thrombocytopenia, Hemorrhagic complications Vasoplegia
MERS	No	Saudi Arabia, South Korea	54	VV	45.5 [ 28.5–58.5]	22%**	Case reports Retrospective cohorts Case control	Poorer outcomes in older patients with more comorbidities

NTD neglected tropical disease, MERS Middle Eastern Respiratory Syndrome Coronavirus

IQR interquartile range, VA venoarterial, VV venovenous

^* = reported as a mean and range

^** = data only available on 27 individuals, 6 survived

All published cases of the use of VV ECMO for patients with respiratory failure come from Colombia, where dengue is endemic [79, 80]. In these reports involving three adults and one child, ECMO duration lasted between 3 and 34 days. Two patients survived. The published cause of death was only available in one case, which was due to intracranial hemorrhage in the setting of pre-eclampsia, thrombocytopenia, and systemic anticoagulation. The available evidence regarding the use of VA ECMO support in dengue-associated myocarditis is limited to seven individuals in endemic countries in both adults and children [81–84]. ECMO durations were generally short in this patient population and were challenging to treat due to the frequent thrombocytopenia and capillary leak. Hemorrhagic complications remain a significant concern in patients with dengue supported on ECMO given the frequency of thrombocytopenia in this patient group. Another consideration in regard to VA ECMO use for dengue is the vasodilatory component of dengue shock syndrome. In general, dengue myocarditis with cardiogenic shock is a more suitable indication for ECMO than the distributive shock seen in dengue shock syndrome [13].

Middle East Respiratory Syndrome (MERS)

Middle East Respiratory Syndrome (MERS) was first described in 2012 and is caused by a novel coronavirus (MERS-CoV) [85]. This zoonosis has a natural reservoir in dromedary camels. The vast majority of cases were noted in 2017 throughout Saudi Arabia and the Middle East. It remains in circulation after the initial outbreak, with challenges related to its lack of vaccine and anti-viral medications [85]. Severe respiratory failure with progression to ARDS is the primary cause of death, which can be complicated by septic shock and multiorgan failure.

There are 54 published cases of ECMO use in adults with cardiopulmonary failure due to MERS from 2014 to 2018. A case control trial showed decreased mortality (65% vs 100%, P = 0.02) with ECMO compared to matched critically ill patients [86]. A different study in older patients with more comorbidities receiving ECMO for MERS reported 100% mortality in their center [87]. Additional case series evaluating patients suffering from MERS frequently had small sub-cohorts that received ECMO in both adults and children, but did not report individual patient outcomes [88–94].

Fungal infections

Histoplasmosis

Histoplasmosis is a dimorphic fungus with a worldwide distribution of cases (Table 3) [95] in North America, Australia, Southeast Asia, and Africa. Histoplasmosis is a major cause of life-threatening illness among persons with HIV in Latin America [96, 97]. Severe disease is mostly seen in immunocompromised individuals with a higher fungal burden, with respiratory failure being the most common severe manifestation. Treatment of histoplasmosis is itraconazole for mild to moderate cases and liposomal amphotericin B with methylprednisolone for severe disease with critical illness. Previous reviews have suggested using doses of 5–8 mg/kg of liposomal amphotericin B in adults receiving ECMO [18]. There are two case reports where ECMO was used to support severe respiratory failure due to disseminated histoplasmosis. One patient was an immunocompetent 24-year-old female who was supported on VV ECMO for 19 days for disseminated histoplasmosis treated with amphotericin B followed by itraconazole [98]. The second patient was a 57-year-old female on methotrexate and a TNF-α inhibitor who developed ARDS, multiorgan failure, and hemophagocytic lymphohistiocytosis successfully treated with amphotericin B, methylprednisolone and ECMO support [99].

Table 3.

Tropical fungal diseases with ECMO evidence

Disease	NTD?	Countries	Published ECMO cases	ECMO mode (n)	Age (median [IQR])	Survival?	Evidence	Comments
Histoplasmosis	No	United States	2	VV	24, 57	100%	Case Reports	Amphotericin B Glucocorticoids
Talaromycosis	No	China	1	VV	50	100%	Case Report	Hematogenous spread

IQR Interquartile Range, VA venoarterial, VV venovenous

Talaromyces

Talaromycosis is a fungal infection caused by the dimorphic fungus Talaromyces marneffei (formerly Penicillium marneffei). Talaromycosis is endemic throughout southeast Asia where it is an important cause of morbidity and mortality in immunocompromised patients, although travel-related exposure leading to infection has been reported [95]. Clinical manifestations are usually related to hematogenous spread and involvement of the reticuloendothelial system with lymphadenopathy and hepatosplenomegaly. Severe disease is manifested by bloodstream infection with septic shock and ARDS. Diagnosis requires a high index of suspicion because culture or histopathology showing characteristic features of Talaromyces is required. Unlike other dimorphic fungi, serology, PCR, and antigen-based testing are not routinely available. Treatment of talaromycosis includes amphotericin B and itraconazole. There has been one case of successful ECMO use for Talaromyces-associated ARDS in a 50-year-old male with 10 days of VV ECMO support [100]. The patient was treated with liposomal amphotericin B, glucocorticoids, and intravenous immunoglobulin.

Parasitic infections

Ascaris

Ascaris lumbricoides is a soil-transmitted nematode that causes the NTD Ascariasis (Table 4). Transmission occurs via ingestion of water or food contaminated with infectious eggs. The prevalence of A. lumbricoides infection is highest in tropical countries, with the majority of individuals living in Asia, Africa, and South America. While intestinal, hepatobiliary, or pancreatic infections are the most common, pulmonary infections occur as well. There are cases of severe pulmonary ascariasis reactions occurring in pediatric patients who are critically ill. ECMO support has only been reported in a single pediatric patient with Ascaris pneumonitis complicating inhalational lung injury, who was treated with Mebendazole and supported on VV ECMO for weeks before passing away on ECMO [101].

Table 4.

Tropical parasitic diseases with ECMO evidence

Disease	NTD?	Countries	Published ECMO cases	ECMO Mode	Age (median [IQR])	Survival?	Evidence	Comments
Ascaris	No	United States	1	VV	NR	100%	Case reports	Inhalational injury Pediatrics
Chagas	Yes	Brazil, Colombia, Mexico	4	VA	35 [20.5–54]	75%	Case report Case series Expert experience	Pericardial tamponade Bridge to Heart Transplant
Echinococcus	Yes	Cyprus, Germany, Romania, Saudia Arabia, Slovenia, Spain	6	VV	24 [ 21–39]	83%	Case reports	Bridge to Thoracic Surgery
Malaria	No	Austria, Colombia, Germany, India Madagascar, Portugal, Turkey, United States	17	VV VA VVA	39 [24–53]	89%*	Case reports Case series	Hemolysis DIC Thrombocytopenia
Strongyloides	Yes	Canada China	2	VV	58.5 [55–62]	0%	Case report	Delayed diagnosis Chronic Steroids CKD

NR not reported

IQR Interquartile Range, NR Not Reported, VA venoarterial, VV venovenous

*Data only available on 9 patients, 8 survived

American trypanosomiasis

Chagas disease is an NTD caused by the protozoan parasite Trypanosoma cruzi. Vector-borne transmission is the major route of transmission via an infected triatomine (‘assassin bug’). Chagas disease is endemic in North, Central, and South America [102]. The epidemiology of Chagas disease has changed due to the migration of infected individuals, with thousands of patients with Chagas infections now living in the United States, Spain, Australia, Japan, and other European countries [103]. Less than 1% of acute infections cause life-threatening complications such as acute myocarditis and meningoencephalitis [104], but chronic infection can lead to chronic Chagas cardiomyopathy in 20–30% of patients [105]. Treatment of acute T. cruzi includes the anti-parasitic agents benznidazole and nifurtimox, which have not been studied during ECMO.

There are 4 published cases of VA ECMO to support myocarditis related to Chagas disease from Brazil, Colombia, and Mexico. Three patients with either acute or chronic Chagas myocarditis survived to hospital discharge with VA ECMO. Their ages ranged from 17 to 62 years, and ECMO was generally used as a bridge to cardiac transplant [79, 106, 107]. There is one reported case of a 46-year-old female who spent 8 days on ECMO for cardiogenic shock with recovery of cardiac function, but she ultimately died of septic shock from ventilator-associated pneumonia, cannula-associated infection, and inguinal infection with exposure of femoral vessels after 55 ICU days [79]. Antiparasitic therapy with nifurtimox was administered prior to ECMO in two of these cases based on results of endomyocardial and pericardial biopsies.

Echinococcus

Echinococcosis is an NTD caused by tapeworms of the genus Echinococcus, with the majority of cases concentrated in tropical and subtropical countries, especially due to Echinococcus granulosus. Alveolar echinococcosis and cystic echinococcosis are the most important forms of disease, and both cause slowly enlarging lesions in the liver, lungs, and other organs [108]. Antiparasitic therapy with albendazole, mebendazole, or praziquantel may be used in certain cases as an adjunct to surgical or percutaneous treatment of cysts. Surgical removal of intact hydatid cysts remains the treatment of choice when feasible. The rupture of a cyst can precipitate allergic reactions ranging from mild to fatal anaphylaxis or severe pneumonitis resulting in respiratory failure.

There are six cases of VV ECMO use to support echinococcus either for severe respiratory failure due to cyst rupture or as a bridge through complex thoracic surgery. Four of these patients received perioperative ECMO support for echinococcosis, including post-operatively for a patient with a hydatid cyst overgrowing her left pulmonary artery requiring left pneumonectomy [109] or perioperatively to surgically remove life-threatening lesions [110–112]. A 51-year-old male tolerated single lung ventilation to suture massive air leaks with support of VV ECMO. VV ECMO was used in two patients as rescue for severe hypoxemia due to cyst rupture, who ultimately had surgical resection after improvement and removal of ECMO [113, 114]. All patients received empiric antibiotics along with either albendazole, praziquantel, or a combination of both. The favorable outcomes of these cases suggest VV ECMO may have a role in support of severe respiratory failure due to acute cyst rupture and to facilitate lifesaving thoracic surgery due to thoracic echinococcus infection.

Malaria

Malaria is a global health problem caused by the obligate intraerythrocytic protozoa of the genus Plasmodium [115]. Malaria primarily causes infections in low- and middle-income countries around the tropics, but imported cases of malaria in returning travelers from these regions also cause significant morbidity [116]. In 2023, there were 263 million cases and 597,000 deaths reported by the World Health Organization [117]. Malaria primarily causes fever and hemolytic anemia, but it can affect multiple organ systems and rarely causes ARDS. Malaria-associated ARDS has been reported with P. falciparum and P. vivax infections [118]. Treatment of severe malaria is preferentially parenteral artesunate. Enteral therapy can be delivered in critical illness, but is limited by its slow onset of action.

There have been 17 cases of ECMO and one case of extracorporeal carbon dioxide removal [119] published to support severe malarial infections. Eight cases of ECMO were VV for respiratory failure [79, 120–124], one case was VA support for suspected myocarditis [125], one case of VVA for ARDS with stress cardiomyopathy [124] and seven others were unspecified forms of ECMO support [126, 127]. Five of the VV ECMO patients described presented with multiorgan dysfunction, cleared their parasitemia with anti-malarial drugs, yet still progressed with ARDS requiring VV ECMO and eventual survival to discharge [120, 121, 123]. VA ECMO was used in a 61-year-old male from Indonesia who developed refractory cardiogenic shock due to a P. falciparum infection [125]. The median ECMO duration was 7 (IQR: 4–90) days, with two patients having prolonged ECMO runs of 40 and 90 days and both surviving to discharge. Six patients suffered malaria from P. falciparum, and four had P. vivax (seven cases did not specify the malaria species). There was rare mention of dose adjustment or drug level monitoring for their anti-malarial treatments while on ECMO. One patient supported with VV ECMO for malaria-associated ARDS successfully cleared their parasitemia with doxycycline 100 mg every 12 h [124] with pharmacokinetic data suggesting that standard doses of doxycycline may produce therapeutic levels and be effective while on ECMO. Severe malaria can cause hemolysis, DIC, and thrombocytopenia, which can be exacerbated by ECMO. These hematologic abnormalities were common in malaria patients supported with ECMO, but there was no mention regarding specific anticoagulation strategies or coagulation monitoring to mitigate these problems.

Strongyloides

Strongyloidiasis is an NTD caused by Strongyloides stercoralis, a helminth transmitted in the soil. It is most prevalent in the tropics, but human infections occur in North America, Europe, Japan, and Australia. Immunosuppressed individuals are at high risk for hyperinfection and disseminated disease. Disseminated Strongyloides can be fatal from septic shock, acute kidney injury, or respiratory failure. Severe disease is treated with ivermectin, with or without albendazole. Ivermectin is ideally given enterally but may be given intramuscularly, subcutaneously, or rectally in cases where enteral administration is not feasible. There are two published cases of VV ECMO reported to support disseminated Strongyloides [128, 129]. Both patients were on prednisone for chronic renal insufficiency and reported walking barefoot in rural environments. Strongyloides was diagnosed approximately one week after critical illness. ECMO duration was seven and eight days in each case. One patient was treated with nasogastric ivermectin, intramuscular ivermectin, and nasogastric albendazole, and the other treated with albendazole. Both patients died of progressive multiorgan failure despite antihelminth therapy and ECMO support.

Discussion

ECMO is established in international guidelines for the support of ARDS [130] and cardiogenic shock [131, 132]. VV ECMO provided gas exchange in cases of ARDS caused by dengue, malaria, MERS, melioidosis, tuberculosis, histoplasmosis, and talaromycoses. VV ECMO was similarly employed in cases of pulmonary hemorrhage due to leptospirosis. VV ECMO was instrumental in the management of Echinococcus granulosus infections in the setting of cyst rupture or perioperatively with high-risk cardiothoracic surgery. VA ECMO provided mechanical circulatory support in multiple cases of cardiogenic shock caused by Chagas disease, malaria, dengue, leptospirosis, and melioidosis. While physicians use ECMO for various causes of severe cardiac or respiratory failure, they may feel more reluctant to consider ECMO to support specific tropical infections. In this scoping review, we described the published literature and demonstrated that ECMO has utility for a variety of bacterial, viral, fungal, and parasitic tropical diseases that cause respiratory and cardiac failure.

With any infection, there is a risk of nosocomial transmission to healthcare workers and other patients. Several tropical diseases, such as Ebola and MERS, have been associated with widespread transmission in hospital settings during outbreaks [133, 134]. Best practices for infection control in the tropics include the use of both systemic interventions to ensure proper environmental controls, training protocols, and equipment to minimize risk as well as individual actions, such as hand hygiene and use of personal protective equipment [135]. Compounded with these challenges in tropical infections, ECMO requires frequent blood draws, the potential need for invasive procedures, and large, multidisciplinary treatment teams, which should be taken into account when coordinating an ECMO response to a tropical infection [136, 137].

This scoping review has several limitations. Many diseases are described only in case reports, which are likely subject to publication bias. Nonetheless, knowing that ECMO has successfully supported a specific tropical infection may help clinical decision-making. Conversely, regions experienced with specific disease processes may not publish ECMO cases because its use may not be seen as novel in these diseases. In resource-constrained environments, with limited access to diagnostics, there are additional challenges to managing a patient with respiratory or cardiac failure, and a microbiological or serological diagnosis may not be able to be made. Alternatively, outside the tropics, there may also be difficulties accessing diagnostics for diseases that are not endemic to that region. Several tropical diseases only have a few cases published, which limited the generalizability of these manuscripts’ conclusions. Finally, many studies included did not provide important clinical details, such as the timing of ECMO initiation, complications, antimicrobial doses, and anticoagulation strategies.

Conclusions

Advances in care will allow for continued expansion of ECMO to new settings and for consideration of new indications that may benefit from this support. The utility of ECMO in tropical infections is likely best in isolated respiratory or cardiac failure and unlikely to be helpful with severe neurologic dysfunction, multiorgan failure, severe vasodilatory or hypovolemic shock. While this scoping review provides proof of concept for ECMO support in multiple tropical infections, it also highlights gaps that can guide future research efforts, including in ECMO optimization and drug delivery for tropical infections.

Abbreviations

ARDS

Acute respiratory distress syndrome

CNS

Central nervous system

DIC

Disseminated intravascular coagulopathy

ECPR

Extracorporeal cardiopulmonary resuscitation

ELSO

Extracorporeal life support organization

HIV

Human immunodeficiency virus

MERS

Middle East Respiratory Syndrome

NTD

Neglected tropical disease

TB

Tuberculosis

SOFA

Sequential Organ Failure Assessment

VA

Venoarterial

VV

Venovenous

WHO

World Health Organization

Author contributions

The study was conceived by GM and JM. The literature search and data synthesis were performed by ZJ and JM. The manuscript was written by ZJ, JM, and GM. The tables were written by ZJ. Additional unpublished data were provided by LS. LS, KR, and DT critically revised the manuscript for important intellectual content. All authors read and approved the final draft.

Funding

None.

Data availability

No datasets were generated or analyzed during the current study.

Declarations

Ethics approval and consent to participate

The views expressed in this manuscript reflect the results of research conducted by the author(s) and do not necessarily reflect the official policy or position of the Defense Health Agency, Department of Defense, nor the U.S. Government.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.