Strongyloides stercoralis infection in the UK: A systematic review and meta-analysis of published cases

Strongyloidiasis is a helminth infection where symptoms vary, and asymptomatic presentation is common. Chronic strongyloidiasis can cause a high mortality ‘hyper-infection’ in immunocompromised states. Understanding at risk populations and symptomology can guide screening and early treatment to reduce hyper-infection risk. A systematic review of studies describing patients in the UK with strongyloidiasis pooled a total of 1,308 patients. Weighted pooled prevalence (WPP) of asymptomatic cases was 27.7% (95% CI 17.1–39.5%, I2 = 92%, p < 0.01). At-risk populations included migrants, returning travellers and armed forces personnel. The most common symptoms reported were abdominal pain (WPP 32.1% (95% CI 20.5–44.8%), I2 = 93%, p < 0.01), rashes (WPP 38.4% (95% CI 13.1–67.7%), I2 = 99%, p < 0.01) and diarrhoea (WPP 12.6% (95% CI 6.7–19.9%), I2=70%, p = 0.03). Symptomatology varied with cohort characteristics. Although asymptomatic presentation is common, patients may present with abdominal pain, diarrhoea or rashes. A low threshold for screening symptomatic individuals in at-risk groups is required.


Introduction
Strongyloides stercoralis is a helminth endemic to tropical and subtropical regions which infects humans after contact with larvaecontaminated soil. 1 , 2 After initial infection, S. stercoralis larvae generated in the intestinal tract can reinfect the host through a process of autoinfection, causing a lifelong chronic infection.Up to 70% of patients with chronic infection may be asymptomatic, with raised blood eosinophil count often being the only indicator of strongyloidiasis. 3n patients with clinical manifestations, symptoms include mild gastrointestinal and respiratory complaints, as well as a characteristic skin rash (larva currens). 4Though chronic infection can be asymptomatic or mild, it has the potential to lead to a hyper-infection state -a lifethreatening syndrome which has a reported case fatality rate approaching 95%. 5 Hyper-infection commonly results from impaired host immunity (which may be iatrogenic), wherein exaggerated parasite dissemination occurs and causes complications including meningitis, renal and respiratory failure and disseminated intravascular coagulation and shock. 3 , 5Chronic strongyloidiasis can be readily treated with antihelminth agents. 6Therefore, screening of populations at risk of har-bouring chronic strongyloidiasis is vital to enable prompt treatment and eradication, averting the risk of hyper-infection.
The worldwide prevalence of strongyloidiasis is estimated to be 8.1%, with regions of sub-Saharan Africa, South America and South-East Asia being recognised endemic areas with prevalence exceeding 20%. 7In the UK, the frequency of strongyloidiasis is unknown.Cases are mainly reported in migrants, and it is estimated that up to one in five arriving to the UK from endemic regions may be affected by helminth infections. 8 , 9Hyper-infection cases in the UK have been reported in the literature in a variety of clinical settings, [10][11][12][13][14][15] and cases may rise given greater global travel and migration, as well as an increasing number of immunocompromised individuals (eg due to increasing immunosuppressive therapies, organ transplants and long-term corticosteroid use).
It is therefore important to screen at-risk populations in the UK for chronic strongyloidiasis.There are currently no studies that have systematically reviewed the reports of strongyloidiasis infection within the UK, and therefore there is a paucity of data to identify the groups most at risk.We aimed to evaluate the demographics, clinical features and outcomes of reported strongyloidiasis in the UK through a meta-analysis of published studies.Our objective was to elucidate symptoms and charachttps://doi.org/10.1016/j.clinme.2024.100227teristics that should increase suspicion and prompt screening of patients for strongyloidiasis in general medical settings (such as the acute medical unit and general practice).

Search strategy and selection criteria
Systematic review and meta-analysis was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. 16The review was registered in PROS-PERO (CRD42023416600) prospectively.
Database searches of PubMed (MEDLINE), Scopus, Web of Science and Cochrane for English language peer-reviewed primary research articles published until 25 January 2023 were conducted.Search terms used a combination of the words ' Strongyloides ' and 'United Kingdom'.Full search terms can be found in Supplementary Table 1.
The Population, Intervention, Comparator, Outcome, Study Design (PICOS) criteria were used prior to selection ( Table 1 ).All full-text studies reporting cases of S. stercoralis diagnosed in the UK were included, regardless of selective populations.Studies with case numbers below 10 and those not reporting demographic features of their cohorts were excluded to reduce sampling bias.

Data extraction
Titles and abstracts from database searches were imported into End-Note X8 (Clarivate Analytics, Philadelphia, USA) and de-duplicated.Remaining studies were uploaded to Rayyan (Qatar Computing Research Institute, Doha, Qatar) for screening.Screening of titles and abstracts and subsequently full-text manuscripts was undertaken in parallel by two reviewers (AMA and CO) and conflicts were resolved through consensus.Data were then extracted by the two authors (AMA and CO).
For included studies, the following fields were gathered: publication year, study type, duration (months), location, cohort characteristics, number of patients with confirmed strongyloidiasis, and if available, sex ratios, average age, reported symptoms at diagnosis, diagnosis modality, average eosinophil count, treatment type, and outcomes (including presence of hyper-infection, death or successful treatment).
Included studies were split into subgroups according to the cohort characteristics: • Mixed cohorts were studies which recruited from unspecified populations.• Migrant cohorts were studies solely recruiting patients who have migrated to the UK.• Returning traveller cohorts were studies recruiting patients who had returned to the UK from travel abroad.• Armed forces personnel cohorts were studies recruiting patients who were returning armed forces personnel or returning prisoners of war.

Quality assessment
Retrospective studies were classified according to the Newcastle Ottawa Scale 17 and randomised trials were assessed according to the Cochrane Risk of Bias 2.0 tool. 18

Statistical analysis
Data analysis of descriptive statistics was performed using SPSS (Version 27, IBM, Armonk, USA).R statistics (Rstudio Version 4.0.1) was used to perform meta-analysis, and create figures, forest and funnel plots (ggplot2 and meta packages).
DerSimonian-Laird random-effects modelling was used to report weighted pooled prevalence (WPP) of clinical variables.A Freeman-Tukey double arcsine transformation was applied to our data.Subgroup analyses were performed.Forest plots were generated with natural scales, and heterogeneity was assessed using Cochran's Q test and I 2 , while Egger's test was used to test for publication bias.
Sensitivity analysis was performed to assess the effect of variables in our analysis.We separated studies which recruited from specific populations to adjust for the effect of mixing heterogeneous studies.
Quality assessment revealed two studies had a high risk of bias, four had a moderate risk, and two had low risk.Full risk of bias assessment is detailed in Supplementary Table 2.

Sensitivity analysis
During sensitivity analysis ( Table 3 ), the WPP of asymptomatic cases was higher when excluding armed forces (27.7% vs 30.7%), while the WPP was lower when excluding returning travellers (27.7% vs 23.7%).
The WPP of abdominal pain reduced more significantly when excluding migrant cohorts (32.1% vs 27.0%).The WPP of abdominal increased when excluding armed forces personnel (32.1% vs 39.1%).
The WPP of diarrhoea increased when discounting armed forces cohorts, although this only left one study reporting diarrhoea (12.6% vs 16.4%).
When excluding armed forces personnel, the WPP of rashes decreased significantly (38.4% vs 19.4%), whereas excluding studies recruiting returning travellers led to an increase in WPP of rash (38.4% vs 52.1%).

Discussion
This systematic review and meta-analysis explored the symptoms and characteristics of published strongyloidiasis diagnoses in the UK to help recognise which populations and presentations should prompt screening.Consistent with previous studies conducted in non-endemic areas, our findings highlight the diagnostic challenges associated with strongyloidiasis due to its tendency to either remain asymptomatic or present with non-specific symptoms. 10 , 27 , 28A quarter of patients with S. stercoralis included in our study were asymptomatic at diagnosis, while commonly reported symptoms included non-specific abdominal pain, diarrhoea and rash.Variation in presentation was observed between different cohorts.

Case selection/sampling
The systematic review yielded at total of eight articles, conducted in either in London or Liverpool, likely due to the long-standing existence of tropical medicine institutions in both these locations.These centres provide services to a large geographical area of patients and have specialised parasitology laboratories.Therefore, although positive cases may be detected at these centres, they may have been referred from other UK regions.Nonetheless, there remains a need for further studies conducted in other populous urban areas with substantial foreign-born populations to facilitate better determination of appropriate indications for testing and treatment at a population level. 29

Asymptomatic infection
Over one-quarter (27.7%) of patients were asymptomatic at diagnosis.Asymptomatic infection occurs when mature helminths persist in the small intestine, maintaining a state of equilibrium with the host immune system. 30Within this balanced state, chronic strongyloidiasis results in granulocyte and eosinophil production. 30 , 31pproximately 70% of patients exhibit peripheral eosinophilia, which may serve as the sole basis for investigating strongyloidiasis. 32he proportion of asymptomatic cases we observed is comparatively lower than the reported figures in other non-endemic areas.A previous systematic review (which included five studies from Japan, Spain, Italy and the UK), found that 41.4% of diagnoses were asymptomatic. 32mong the studies included in these meta-analyses was a cohort study of 1,245 patients in Spain which reported that 82.1% of patients were asymptomatic at diagnosis. 33This difference may suggest that screening is less common in the UK.However, of the five included studies in the systematic review, diagnosis was primarily through combined serological testing and eosinophilia.In comparison, several of our studies included stool microscopy or culture in diagnosis.Numerous diagnostic tests are available for the diagnosis of strongyloidiasis, including stool examination, serological testing and PCR.Serological testing may overestimate the prevalence of the infection since it can remain positive even after the infection has resolved or cross-react with other helminth infections. 34This can lead to false-positive results and dilute the association of symptoms with strongyloidiasis. 32The sensitivity of microscopic diagnosis is improved by increased larval quantity, therefore prevalent use of microscopic diagnosis in our meta-analysis may skew our findings towards those with higher larvae burdens, which may in turn bias towards increased infection severity and symptomatic presentation.Additionally, our study included cases with co-infections, which introduces the possibility that positive symptoms were suggestive of other helminth infections.
There was heterogeneity when comparing asymptomatic infection between cohorts.Excluding returning travellers led to a decrease in the WPP of asymptomatic diagnoses, suggesting that this groups is being screened in some capacity. 35The presence of infective symptoms may depend on a high burden of larvae resulting from a significant period of autoinfection. 22 , 36This could explain the higher asymptomatic rates observed in returning travellers, who may have a limited opportunity for a high burden to occur due to relatively recent exposure to S. stercoralis .
In comparison, excluding armed forces personnel raised the WPP of asymptomatic diagnosis.This suggests this cohort is not screened as much as the others included in our study.Previous studies have shown that prisoners of war, for example, experience more severe helminth infection due to factors associated with their length in captivity, living condition and nutrition at the time of infection. 37Similarly, armed forces personnel have unique occupational exposures in endemic regions and are highlighted as a high-risk group. 38Given this, any patient who has served abroad in the armed forces should be considered for screening of strongyloidiasis in the UK, especially in the presence of eosinophilia.

Symptomology
Exploring symptomology of chronic strongyloidiasis is important given that 30% may not exhibit eosinophilia. 32The most common symptoms reported in our analysis were abdominal pain, diarrhoea and the presence of a rash, which aligns with findings in other metaanalyses 32 , 39 Presence of these symptoms should prompt screening in atrisk populations.Symptoms in strongyloidiasis arise when the balance between the host immune system and larvae production is disrupted.Host immune responses can vary for a multitude of reasons (eg genetic, environmental, comorbidity-related, microbiome-related). 30 Unsurprisingly, differences in symptomology were seen between distinct cohorts.
Overall, 32% of patients reported abdominal pain at diagnosis, while 13% had diarrhoea.In sensitivity analysis, these rates may have been higher among the migrant cohorts.Migrants have an increased probability of experiencing concurrent infections that can lead to abdominal symptoms, such as giardiasis or amoebiasis.S. stercoralis detection in these cohorts may be incidental in association with symptomatic coinfections.Given the correlation between the burden of larvae migration through the gastrointestinal tract and amplification of symptoms, migrants may be more symptomatic due to longer time periods in endemic areas, increasing the risk of repeated Strongyloides larval exposure. 39Interestingly, this finding was not observed in the army personnel, which on sensitivity analysis showed lower rates of abdominal symptoms, despite extended stays in endemic settings allowing for significant larvae burden.Such differences could be due to variations in the human gut microbiome between cohorts.The gut microbiome can influence immune responses to helminths and parasites, and microbiomes differences may contribute to variations in gastrointestinal presentations among populations. 40inally, 38.4% of patients reported a rash at diagnosis.Presentation with rashes was highest in armed forces personnel compared to other cohorts.This rash may have been larva currens or 'creeping eruption', which is often considered pathognomonic of strongyloidiasis, occurring due to larval migration through the skin resulting in a local hypersensitivity response.Genetic, immunological and environmental factors can influence hypersensitivity responses to parasites, and these may differ among ethnic groups. 41These variations may account for our findings; however, alternative infectious aetiologies for rashes in returning military personnel, may also result in the incidental diagnosis of strongyloidiasis. 42

Importance of screening immunocompromised cohorts
Hyper-infection is a significant risk in chronic strongyloidiasis.In our analysis, only two patients had reported hyper-infection.However, there was a paucity of data on the clinical presentation of strongyloidiasis in immunocompromised individuals, and further research in this subgroup of patients at the highest risk of hyper-infection is required.
Notably, one of the excluded studies in our analyses screened for strongyloidiasis specifically in people living with HIV, despite disseminated strongyloidiasis not being considered prevalent in advanced HIV disease. 43In comparison those living with human T-cell lymphotropic virus type 1 (HTLV1) are at significant risk, and though the rates of HTLV infection in England and Wales are low, most diagnosed cases are among individuals born outside the UK, increasing their risk of S. stercoralis infection. 44In Ming et al 's study, seven patients with HTLV-1 were included and it was noted that this sub-population may not present with peripheral blood eosinophilia, highlighting the need to understand symptomology suggestive of strongyloidiasis in the absence of increased eosinophils. 24More studies should focus on sub-population, and screening for HTLV-1 and strongyloidiasis should be routine when either is diagnosed. 456][47][48][49] In organ transplantation, most severe disease occurs within the first few months, and recent calls have been made to include screening in transplantation guidelines. 50 , 51[54]

Limitations and future directions
This review is subject to certain limitations.Firstly, a limited number of studies provided detailed information on symptomology and demographics, which restricts the comprehensiveness of our findings.The possibility of selective reporting could have influenced the accuracy and completeness of our findings -especially in the army personnel cohorts which were both conducted in Liverpool.
Additionally, due to the scarcity of reported cases in the literature, we incorporated studies that examined co-infections, and this inclusion contributed to the heterogeneity observed within our analysis as these studies utilised different diagnostic approaches.Moreover, it is difficult to conclusively attribute reported symptoms to strongyloidiasis due to possibility the crossover in presentation with other similarly acquired parasitic and other infections.
Most included studies were conducted in a small number of distinct centres and in very limited UK regions.Although London possesses a higher population of individuals with origins from regions with Strongyloides endemicity, other known at-risk groups for strongyloidiasis may not be as well represented in this region.Given our findings of varied typical presentation in different at risks groups, sampling from other UK regions is essential to better elucidate possible indications for Strongyloides testing in an unselected cohort.Additionally, epidemiology in other urban areas (eg Leicester, Luton, Manchester) with significant migrant populations from Stronygloides -endemic areas have not been studied. 29n confirmed asymptomatic infection, studies of the prevalence of other clinical signs (particularly respiratory symptoms which were scarcely reported in the included studies) and positive biomarkers of disease (including eosinophilia) from a general cohort of patients will be useful given the high frequency of asymptomatic individuals who may eventually be at risk of hyper-infection.Additionally, existing datasets of population comorbidity and immunosuppression (eg for COVID-19 shielding and vaccination) could be utilised to identify those at risk of Strongyloides hyper-infection, the prevalence of undetected strongyloidiasis within this cohort, typical symptomatology, and other positive indicators for infection.
To better stratify the risk and presentation of undetected strongyloidiasis and hyper-infection in the general, further UK-wide studies are imperative with a view to improved guidance on screening and treatment.

Conclusion
Our findings demonstrated that in a UK cohort, over one in four patients diagnosed with S. stercoralis may be asymptomatic, and presence eosinophilia in at-risk populations should prompt screening.The most common symptoms at diagnosis included abdominal pain, diarrhoea and rash.However, specific cohorts may have present differently at diagnosis, with those having extended periods in endemic region being most likely to present with symptoms.In these at-risk groups (eg migrants and armed forces personnel), non-specific abdominal pain, diarrhoea or rash with no other cause identified should prompt screening for strongyloidiasis.Given the risk of hyper-infection, clinicians should have a low threshold to screen any patient with exposure to endemic regions, especially if they are or will be considered for immunosuppression.Further high-quality evidence on strongyloidiasis in immunosuppressed cohorts, and specifically in those with HTLV-1, is needed ( Box 1 ).

Key practice implications:
• Asymptomatic strongyloidiasis is common and screening should be considered in all individuals with a history of travel to endemic areas prior to immunosuppressive treatment to reduce the risk of hyper-infection.• Symptomatology can differ based on patient characteristics.
• Strongyloides testing should be considered in migrants from endemic areas presenting with chronic gastrointestinal symptoms.
• Strongyloides testing should be considered in armed forces personnel with prior travel to endemic areas.

Fig. 1 .
Fig. 1.PRISMA flow diagram of study selection for inclusion in meta-analysis.

Table 1
PICOS inclusion criteriaReview questionWhat are the demographics, clinical features and outcomes of patients who are diagnosed with Strongyloides stercoralis in the UK? Population All patients diagnosed with S. stercoralis in the UK.

Table 3
Sensitivity analysis of weighted pooled prevalence of symptoms in different cohorts with strongyloidiasis.