Strongyloides hyper-infection in a lung transplant recipient: Case report and review of the literature

Abstract

CASE PRESENTATION

A 63-year-old man with a left single lung transplant for end-stage combined restrictive and obstructive lung disease developed persistent pulmonary infiltrates and recurrent gram-negative bacteremia post-transplant. Bronchoalveolar lavage fluid revealed a nematode on Papanicolau staining compatible with Strongyloides stercoralis larvae on day 50 post-transplant. Although Strongyloides serology performed post-transplant was negative, a retrospective review of the medical record revealed marked peripheral blood eosinophilia on several occasions before transplantation. Despite reduction in immunosuppression and treatment with albendazole and ivermectin, the patient developed another episode of Escherichia coli bacteremia. He died 3 months post-transplant from pulmonary and neurological complications.

DIAGNOSIS

Strongyloides hyper-infection.

DISCUSSION

Strongyloides hyper-infection syndrome is known to occur in immunocompromised patients, but it has only been reported once in a lung transplant recipient. This case illustrates the importance of screening for parasitic infections before transplantation in patients with marked eosinophilia, especially among immigrants from countries in which Strongyloides is endemic. Hyper-infection syndrome may appear years after infection in the context of immunosuppression or immunodeficiency. This case also highlights the association between Strongyloides hyper-infection and bacteremia with enteric organisms.

Introduction

Lung transplant recipients frequently have infectious complications; however, parasitic infection is rare and not often considered. Strongyloides stercoralis is an intestinal nematode that can cause disease predominantly among immunosuppressed hosts. It has a worldwide distribution but is more common in tropical and subtropical rural areas and in some parts of the southeastern United States. The parasite lives in the mucosa of the small bowel. Three variations of the life cycle of the parasite have been described: direct, indirect, and auto-infection. In the direct cycle, the rhabditiform larvae transform to filariform larvae in the soil, penetrate the skin, and travel through the bloodstream to the right heart, lungs, and airways. The larvae are then coughed up in respiratory secretions and swallowed into the esophagus. Once in the small bowel, they infect the intestinal mucosa, where they develop into the adult female parasite and produce eggs that hatch into rhabditiform larvae in the mucosa. In the indirect cycle, the rhabditiform larvae grow into adult organisms in the soil and produce eggs. The eggs then produce filariform larvae, which start the direct cycle just described. The auto-infection cycle occurs when the rhabditiform larvae transform into filariform larvae in the intestinal lumen. The filariform larvae then penetrate the intestinal mucosa, reach the circulation, and continue the direct cycle without the need of external reinfection (1).

With immunosuppression, this auto-infection cycle is accelerated, and non-infectious rhabditiform larvae mature into infectious filariform larvae in the host’s gastrointestinal tract. With increased larval migration, patients may present with or have exacerbated gastrointestinal and pulmonary symptoms, with the detection of increased numbers of larvae in stool and pulmonary samples, such as sputum or bronchoalveolar lavage (BAL), leading to the hyper-infection syndrome. Strongyloides hyper-infection syndrome is often accompanied by bacteremia as a result of enteric organisms arising from the gut translocating into the bloodstream and occasionally causing meningitis, with mortality rates of 70%–85% (2).

In disseminated Strongyloides infection, there is migration of larvae to organs beyond the range of the pulmonary auto-infective cycle. This results in widely disseminated larval infection not only to the lungs but also to other organs not commonly involved in the normal helminthic life cycle, such as liver, heart, kidneys, endocrine glands, and central nervous system.

Hyper-infection syndrome has been described in many immunocompromised people, including renal, heart, intestinal, and stem cell transplant recipients, but it has been described only once after lung transplantation (2–6). We describe the second case report, to our knowledge, of hyper-infection after lung transplantation, in which despite successful identification of the nematode in the recipient’s BAL and adequate treatment, the patient died as a result of ongoing gram-negative bacteremia and multiple pulmonary and systemic complications.

Case report

A 63-year-old man who had immigrated from Laos to Canada 40 years earlier was listed for lung transplantation as a result of combined end-stage restrictive and obstructive lung disease. His medical history included chronic obstructive pulmonary disease since 2005 (ex-smoker with 40 pack-year history); non–small cell carcinoma (staging T4N0) treated with chemotherapy and radiotherapy to the right upper lobe in 1996; pneumonia complicated by acute respiratory distress syndrome, requiring intubation and mechanical ventilation in 2010 with resultant fibrosis; and a history of myocardial infarction with concomitant atrial fibrillation. He was oxygen dependent and had modified Medical Research Council grade 4 dyspnea at the time of referral.

His pre-transplant assessment revealed occlusion of his left anterior descending coronary artery, requiring stenting. Further routine assessment included tuberculin skin testing, which was negative, and screening for hepatitis B and C, which revealed a positive hepatitis B core antibody. He was positive for cytomegalovirus (CMV), Epstein-Barr virus, and varicella zoster virus antibody and negative for HIV. Sputum cultures were negative for fungi and mycobacteria. He was not tested for Strongyloides or human T-cell leukemia/lymphoma virus (HTLV) antibodies before transplant.

The patient underwent a left single-lung transplant from a brain-dead, hepatitis C-positive donor. The transplant was performed on veno-arterial extracorporeal membrane oxygenation (ECMO) bypass. The surgery was technically difficult because of chest wall adhesions and bleeding. Postoperatively, the patient required inhaled nitric oxide, which was weaned on postoperative day (POD) 2. His initial immunosuppression consisted of methylprednisone, mycophenolate mofetil, and basiliximab as an initial calcineurin inhibitor sparing agent because of acute kidney injury. Because the donor respiratory secretions were positive for Haemophilus influenzae, he was treated with piperacillin–tazobactam and, as per our centre’s post-transplant protocol, prophylactic valganciclovir and trimethoprim–sulfamethoxazole. Piperacillin–tazobactam was switched to cefazolin on POD 3 and continued for another 24 hours.

As a result of marginal allograft function and anticipated prolonged mechanical ventilation, a tracheostomy was placed on POD 9. Tacrolimus was initiated on POD 10 once renal function improved.

With persistent pulmonary infiltrates on chest X-ray and increased oxygen requirements, a bronchoscopy and BAL were performed on POD 4 with negative results. Blood cultures were negative after 72 hours of incubation. The patient had never presented with any gastrointestinal symptoms. He received a steroid bolus for presumed rejection (10 mg/kg of methylprednisone for 3 d). On POD 20, he developed hemodynamic instability, with deterioration in respiratory status and new bilateral parenchymal infiltrates, worse on the left hemithorax, along with a new chest wall hematoma, for which he underwent exploratory thoracotomy. He was placed on venous–venous ECMO to allow protective allograft ventilation. Blood and intra-operative pleural cultures grew extended spectrum beta-lactamase (ESBL)–producing Escherichia coli, and the patient was commenced on intravenous meropenem.

Despite clinical improvement, after 1 week of meropenem treatment, blood cultures were persistently positive for ESBL-producing E. coli. A transesophageal echocardiogram revealed degenerative aortic valve disease, and infective endocarditis was excluded. No other source of infection for the recurrent E. coli bacteremia could be identified. The patient received 4 weeks of antibiotics for E. coli empyema and bacteremia.

He was gradually weaned from the ventilator and was transferred to the ward on POD 37. After being off antibiotics for 1 week, septic shock recurred, and he was transferred back to the intensive care unit. Both blood and tracheal aspirate cultures were again positive for ESBL-producing E. coli. Bronchoscopy was performed because of his pulmonary infiltrates on POD 50, and the BAL sample demonstrated a nematode consistent with third-stage larvae of S. stercoralis (Figure 1). Therapy with 400 mg of albendazole every 12 hours and 12 mg of ivermectin every 24 hours (200 μg/kg/d) via gastric tube was immediately instituted for 2 weeks’ duration. His immunosuppressive doses of tacrolimus and mycophenolate mofetil were lowered. Therapy with intravenous meropenem was continued for 4 weeks in an effort to prevent secondary bacteremia and central nervous system involvement. Stool testing for ova and parasites (five samples), enzyme-linked immunosorbent assay Strongyloides serology, and HTLV 1 and 2 serologies were all negative. Immunoglobulin (Ig) quantitation (IgG, IgA, and IgM) showed markedly low IgM levels but normal IgG and IgA levels. The patient had never developed any episode of CMV viremia. He had never developed signs of meningitis; therefore, lumbar puncture was not performed. Retrospectively, his absolute eosinophil count was elevated on several occasions before transplantation, ranging between 600/mm³ and 730/mm³ (normal values: 40–400 mm³). Of note, after POD 1, the eosinophilia had resolved.

Figure 1:

Papanicolau staining showing a nematode compatible with Strongyloides stercoralis larva in patient’s bronchoalveolar lavage fluid on post-transplantation day 50

His clinical course was further complicated by a right leg deep vein thrombosis and atrial fibrillation for which he was on a therapeutic dose of intravenous heparin. At almost 3 months post-transplant, he had an abrupt decrease in level of consciousness. Computed tomography of the brain revealed an acute-on-chronic subdural intracranial and subarachnoid hemorrhage with uncal herniation. Neurosurgery was consulted, and the patient was deemed an unsuitable candidate for surgical evacuation. He was declared brain dead the following day. After a prolonged discussion with his family regarding his prognosis, he was provided comfort measures and died on POD 79. No autopsy was performed.

Discussion

Strongyloides hyper-infection syndrome produces significant patient morbidity that can be life threatening among patients who are immunosuppressed. Cell-mediated, humoral, and mucosal immunity are involved in protection against Strongyloides. Factors that impair any of these forms of immunity, such as immunosuppressive therapy, chemotherapy for hematologic malignancies, or HIV infection, increase the likelihood of developing hyper-infection (7).

Strongyloides infection after solid organ transplantation (SOT) may occur by de novo acquisition via travel or residence in endemic areas after transplantation, reactivation of disease in the recipient, or donor-derived transmission. Reactivation is the most common form, although confirmed transmission from donor-derived strongyloidiasis in SOT recipients has previously been reported (8,9). Our patient last visited Laos, which is endemic for Strongyloides (10,11), in 2010 and likely had latent infection before transplant, as suggested by his mild eosinophilia.

Risk factors for hyper-infection include high-dose corticosteroids, which induce eosinophil apoptosis, and immunosuppressive agents, such as azathioprine, tacrolimus, and antithymocyte globulin, which impair T-cell–mediated immunity (12). A low eosinophil count is a poor prognostic indicator (13).

Surprisingly, there is some evidence that cyclosporine has activity against Strongyloides in vitro, and its use in immunosuppression prevented Strongyloides hyper-infection among kidney transplant patients, although this association has not been confirmed clinically (14–16). The immunosuppression protocol at our transplant centre uses cyclosporine as the first-line calcineurin inhibitor; however, it is difficult to affirm whether this is the sole explanation, because despite performing many lung transplants for immigrants from endemic countries, Strongyloides hyper-infection syndrome has not previously been reported in our centre. In a previous study screening for intestinal parasite infections among newly arrived Southeast Asian refugees in Canada between July 1982 and February 1983, the seroprevalence of Strongyloides infection ranged from 11.8% to 76.6% (17). Although our recipient’s initial drug regimen used tacrolimus rather than cyclosporine as per the protocol for donors positive for hepatitis C, the steroid pulse given for presumed rejection is most likely the main reason contributing to the hyper-infection in this case.

Symptoms related to disseminated disease are often a result of direct organ invasion by the larvae or bloodstream seeding of bacteria from the gut or lungs. Gram-negative bacteremia has been reported concomitantly with S. stercoralis infection, as in our case in which multiple E. coli bacteremic infections occurred, without another source for this infection. There have also been reports of significant gastrointestinal and central nervous system bleeding among patients with hyper-infection syndrome as a result of the invasion of larvae into organs (18,19). Although this patient was on heparin infusion for deep vein thrombosis and atrial fibrillation, which could have accounted for his neurological complications, it is also plausible that these complications could have been the result of larval migration. Unfortunately, the patient’s family did not consent to an autopsy to confirm our hypothesis.

Because of the limitations of single stool testing, which can fail to detect larvae in up to 70% of cases, serologic tests are used (20). Most serologic tests measure IgG response to a crude soluble extract of larvae obtained from experimentally infected animals or related Strongyloides species. Limitations include cross-reactivity among patients with filarial infection or infection from other soil-transmitted helminths, diminished sensitivity among immunosuppressed patients such as SOT recipients, hematologic malignancy or HTLV I infection, inability to distinguish between current and prior infection, and lack of standardization across centres. In addition, there is evidence that even in cases with microscopically proven strongyloidiasis, serologic testing was associated with only 81% sensitivity (21). This study suggests that there might be a correlation between immunological control of strongyloidiasis and the amplitude of the humoral response. Serology shows greater sensitivity among immunocompetent hosts (95%) than among those individuals on immunosuppressive agents (20,22). Of note, our patient was tested twice, and even in the absence of IgG deficiency and HTLV infection, his serology was negative. In this scenario, we reinforce that serology should not preclude empiric treatment for patients with suspected Strongyloides infection, given the high chance of false-negative results.

Screening for asymptomatic Strongyloides infection before initiation of transplant immunosuppression can prevent the development of hyper-infection syndrome (23). Eosinophilia, defined as the presence of more than 500 eosinophils/mm³ of blood (or >7% of the leukocyte differential) should raise the suspicion of latent parasitic disease (7). However, in a recent study with more than 86 patients with microscopically proven infection, most of them were asymptomatic, and 23% had a normal eosinophil count, implying that neither the presence of symptoms nor eosinophilia is sufficiently sensitive for diagnosis of infection (21). This reinforces the need to adopt universal screening protocols for SOT candidates and donors from endemic areas (7,24,25). In previous screening protocols, the prevalence of positive serology tests varied from 4.9% to 10.5% among recipients and from 3.9% to 9.3% among donors from endemic areas (23,26,27). Of note, our patient had eosinophilia at least 2 years before transplant, which disappeared on POD 1 after transplant, probably as a result of corticosteroid use, which induces eosinophil apoptosis. Because of difficulties in management of deceased donors and recipients, the pitfalls of serology with false-negative results, and the relative simplicity of prophylactic treatment, some authors advocate for a more pragmatic approach with prophylactic ivermectin on the basis of origin and history of both recipients and donors (28).

Although there is no standardized treatment for Strongyloides hyper-infection syndrome, we suggest adopting the American Society of Transplantation guidelines that recommend using 200 μg/kg of ivermectin orally daily, which may be combined with albendazole, until 2 weeks after the last positive stool sample (29–32). Subcutaneous and rectal ivermectin have been reported as an option in the setting of malabsorption (33,34). Antibacterial pre-emptive treatment is often advised because of the risk of translocation of gram-negative bacteria from the gastrointestinal tract during invasion by the larvae. Of note, sepsis and bacteremia are predictors of mortality and poor prognosis among this population (35).

In conclusion, we report the second case, to our knowledge, of direct identification of S. stercoralis larvae in the BAL fluid of a lung transplant recipient. Evaluation and pre-emptive treatment for S. stercoralis infection should be considered for at-risk potential recipients before transplantation as well as for donors from endemic countries. Because serology is less useful among SOT recipients, patients with clinical findings consistent with hyper-infection or disseminated strongyloidiasis should be extensively tested by checking for larvae in stools, BAL, pleural fluid, and cerebrospinal fluid, and empiric treatment should be done pre-emptively. Testing for strongyloidiasis should also be performed when assessing recurrent bloodstream infections with enteric organisms of an unknown source among patients who are potentially at risk for this parasite.

Ethics Approval:

N/A

Informed Consent:

Informed patient consent has been secured from the patient’s next of kin.

Registry and the Registration No. of the Study/Trial:

N/A

Funding:

No funding was received for this work.

Disclosures:

N Belousova received financial assistance for attending the ISHLT 2022 meeting; C Rotstein has received grants or contracts from Cidara, Merck Canada Inc., and Pfizer Canada Inc. He has received consulting fees from Avir Pharma and Merck Canada Inc. and payment or honoraria from Roche Pharma Canada, Sunovion Pharmaceuticals Canada, and Merck Canada Inc. The other authors have nothing to disclose.

Peer Review:

This manuscript has been peer reviewed.

Animal Studies:

N/A