Primary causes of hospital admission in HTLV-1 infected patients: infectious complications, predictors of mortality and long-term survival outcomes

Abstract

Background

Human T-cell lymphotropic virus type 1 (HTLV-1) is a globally prevalent retrovirus associated with adult T-cell leukemia-lymphoma (ATL) and HTLV-1-associated myelopathy/tropical spastic paraparesis (HAM/TSP). Despite substantial research, the high mortality risk among HTLV-1 carriers remains poorly understood.

Objective

This study aimed to identify primary causes of hospital admission, predictors of mortality, and long-term survival outcomes in HTLV-1-infected patients.

Methods

A retrospective cohort study was conducted in two referral hospitals in Mashhad, Iran, from 2011 to 2022, including 308 adult HTLV-1-infected patients. Hospital admissions, clinical outcomes, and mortality predictors were analyzed. Survival analysis was performed on 95 patients who completed follow-up.

Results

Among the 308 patients, a total of 582 hospital admissions were evaluated. The median age of the patients was 55 years, and 61.4% were female. The primary causes of hospitalization were infectious complications (23.7%), HTLV-1-associated myelopathy/tropical spastic paraparesis (HAM/TSP) (14.8%), and adult T-cell leukemia (ATL) (12.5%). The in-hospital mortality rate was 16.8%, with 98 patients succumbing during their hospital stay. Among patients discharged from the hospital, the median post-discharge survival was approximately 33 months. Factors significantly associated with in-hospital mortality included extended hospital stays, elevated leukocyte counts, high percentages of polymorphonuclear cells, thrombocytopenia, and co-infection with hepatitis B.

Conclusion

HTLV-1 infection is associated with high mortality during hospitalization and limited long-term survival after discharge, largely driven by infectious complications and co-infections such as hepatitis B. The substantial post-discharge mortality emphasizes the need for improved patient management and robust follow-up care to mitigate long-term mortality risk in this population.

Clinical trial number

Not applicable.

Introduction

Human T-cell lymphotropic virus type 1 (HTLV-1), the first human retrovirus identified in the early 1980s, predominantly infects T-lymphocytes and is known for its global prevalence. Despite many carriers remaining asymptomatic, HTLV-1 can lead to severe, life-threatening complications. The virus is strongly associated with specific diseases, most notably adult T-cell leukemia-lymphoma (ATL), a rare and aggressive cancer, and HTLV-1-associated myelopathy/tropical spastic paraparesis (HAM/TSP), a debilitating neurological condition. The lifetime risk of developing ATL among HTLV-1 carriers ranges from 2% to 7%, while the risk of developing HAM/TSP is between 0.25% and 3% [1].

HTLV-1 is transmitted through several routes including maternal-to-child transmission via breastfeeding, sexual contact, and contaminated blood products [2]. This virus exhibits a highly heterogeneous global distribution with focal endemic areas. As of 2012, it was estimated that 5 to 10 million people worldwide were living with HTLV-1. Notable regions with high prevalence include Japan, Iran, the Americas, the Caribbean, Melanesia, Central and West Africa, and Australia [3]. In Iran, HTLV-1 prevalence is notably high, with Mashhad in Northeastern Iran identified as one of the most endemic regions, showing a prevalence of 2.1%. Transmission in Iran primarily occurs through blood transfusions, breastfeeding, and sexual contact [4].

Despite extensive research into the diseases associated with HTLV-1, the increased mortality risk for HTLV-1 carriers is not fully explained by existing morbidity data. While some HTLV-1-related conditions, such as ATL, are known to be fatal, their low incidence cannot solely account for the heightened mortality rates observed. Significant gaps remain in understanding the virus’s impact on cardiovascular, cerebrovascular, and metabolic diseases. Therefore, the overall burden of HTLV-1 may be more extensive and complex than currently appreciated [1].

This study aims to address these gaps by investigating the primary causes of hospital admission among HTLV-1-infected patients, identifying predictors of mortality, and evaluating long-term survival outcomes.

Methods

This retrospective cohort study included all adult patients with HTLV-1 who were hospitalized between 2011 and 2022 at two main referral hospitals in Mashhad, Iran. As the second-largest city in the country and located in the northeast, Mashhad is a key region for HTLV-1 research. The objective was to evaluate the causes of hospital admissions, the clinical outcomes of these patients during their hospital stay and post-discharge, and the predictors of mortality.

Diagnosis of HTLV-1 infection and inclusion criteria

In most patients, HTLV-1 infection had been diagnosed prior to hospital admission. In Iran, the standard diagnostic protocol for HTLV-1 infection involves initial screening by enzyme-linked immunosorbent assay (ELISA), followed by confirmation with Western blot testing. All adult patients with confirmed HTLV-1 infection who were hospitalized between 2011 and 2022 were included in the study, regardless of the reason for admission. Both symptomatic individuals (e.g., with HTLV-1-associated conditions such as HAM/TSP or ATL) and asymptomatic carriers admitted for other medical reasons were eligible for inclusion.

Geographical representation

The study included two maps to provide a geographical context for the cohort of HTLV-1-infected patients. The first map highlights Razavi Khorasan Province, where Mashhad is located. This map features numbers representing the count of HTLV-1-infected individuals born in each city who were admitted to the two referral centers in Mashhad (Fig. 1). The second map presents an overview of Iran, indicating the nationwide distribution of HTLV-1-infected patients included in the study (Fig. 2).

Fig. 1.

Geographic distribution of HTLV-1-infected patients born in cities across razavi khorasan province admitted to referral centers in Mashhad

Fig. 2.

Nationwide distribution of HTLV-1-infected patients born in provinces across iran admitted to referral centers in Mashhad

Statistics

Descriptive statistics for continuous variables were presented as medians with interquartile ranges (IQR), while categorical variables were expressed as frequencies and percentages. To identify factors associated with mortality, binary logistic regression was performed. Initially, univariate analysis was conducted to assess the association between individual variables and mortality. Variables with a p-value of less than 0.05 in the univariate analysis were subsequently entered into the multivariable logistic regression model to adjust for potential confounding factors. Collinearity among variables was evaluated, and no significant collinearity was detected. Survival analysis was carried out using the Kaplan–Meier method. A p-value of less than 0.05 was considered statistically significant for all analyses.

Ethics

All patients or their legally authorized representatives provided informed consent. This study received approval from the ethics committee of Mashhad University of Medical Sciences under the code IR.MUMS.MEDICAL.REC.1401.318 and was performed by the ethical standards set forth by the Helsinki Declaration.

Results

Demographic and clinical characteristics

The cohort comprised 308 HTLV-1 infected patients with a median age of 55 years (IQR: 43, 65.3). Among them, 84 patients (27.3%) were elderly (≥ 65 years). The majority of the patients were female, accounting for 61.4% (189/308) of the cohort, and a significant proportion were married, with 84.7% (261/308) reporting this marital status.

A notable portion of the cohort had underlying medical conditions. Diabetes mellitus was present in 13% (40/308) of the patients, while malignancies were identified in 15.3% (47/308) of the patients including 14.6% hematological malignancies and 0.64% solid tumors. Myelopathy was prevalent in this cohort, affecting 36% (111/308) of the patients, either as a known underlying condition among HTLV-1 infected individuals or as the final diagnosis of hospital admission. Additionally, 12.7% (39/308) of the patients were undergoing chemotherapy. Co-infections with hepatitis B virus (HBV) and hepatitis C virus (HCV) were relatively rare, affecting 4.2% (13/308) and 1% (3/308) of the patients, respectively. No patients in the cohort were co-infected with HIV (Table 1).

Table 1.

Patients’ characteristics

Variables
Age, years	55 (43, 65.3)
Elderly (≥ 65 years)	84/308 (27.3)
Gender, female	189/308 (61.4)
Marital status, married	261/308 (84.7)
Underlying conditions
Diabetes mellitus	40/308 (13)
Malignancy	47/308 (15.3)
Hematological malignancy*	45/308 (14.6)
Solid tumor**	2/308 (0.64)
Myelopathy	111/308 (36)
Chemotherapy	39/308 (12.7)
HBV co-infection	13/308 (4.2)
HCV co-infection	3/308 (1)
HIV co-infection	0
Episodes of hospital admission
One episode	163/308 (52.9)
Two episodes	64/308 (21)
Three episodes	34/308 (11.3)
Four episodes	21/308 (7)
Five episodes or more	26/308 (8.4)

* These include ATL, NHL (Non-Hodgkin Lymphoma), DLBL (Diffuse Large B-cell Lymphoma), CLL (Chronic Lymphocytic Leukemia)

**These include Lung cancer, Bladder cancer

The study found that out of 308 HTLV-1 infected patients, 163 (52.9%) experienced one episode of hospital admission, while 145 (47.1%) had multiple admissions: 64 patients (21.0%) had two episodes, 34 patients (11.3%) had three episodes, 21 patients (7.0%) had four episodes, and 26 patients (8.4%) had five or more episodes.

Cause of hospital admission

The study revealed diverse final diagnoses for the 582 hospital admissions. Infectious complications were the most common, accounting for 23.7% (138/582) of the admissions. HTLV-1 associated myelopathy/tropical spastic paraparesis (HAM/TSP) was diagnosed in 14.8% (86/582) of the cases, while adult T-cell leukemia/lymphoma (ATLL) was found in 12.5% (73/582). Complications related to ATLL and cardiovascular and HAM/TSP were present in 11.9% (69/582) and 3.4% (20/582) and 3.1% (18/582) of admissions, respectively. Other neoplastic disorders accounted for 18 admissions (3%), and 2.4% (14/582) of the cases had an unknown final diagnosis. Cutaneous T-cell lymphoma/Sezary syndrome and neurologic complications each represented 2.1% (12/582) of admissions, while rheumatologic disorders were present in 1.9% (11/582) of cases, respectively. The remaining 19% (111/582) of admissions were categorized as “other” types.

The median length of stay (LOS) was 7 days (IQR: 3, 12). In-hospital mortality was observed in 16.8% (98/582) of the cases (Table 2).

Table 2.

Causes of hospital admission and clinical outcome in HTLV-1-incted patients

Final diagnoses
Infectious complications	138/582 (23.7)
HAM/TSP	86/582 (14.8)
ATLL	73/582 (12.5)
ATLL complications*	69/582 (11.9)
HAM/TSP complications**	18/582 (3)
Other types of neoplastic disorders***	18/582 (3)
Unknown	14/582 (2.4)
CTCL/Sezary syndrome	12/582 (2.1)
Neurologic complications^	12/582 (2.1)
Rheumatologic disorders#	11/582 (1.9)
Cardiovascular complications@	20/582 (3.4)
Others	111/582 (19)
LOS	7 (3, 12)
In-hospital mortality	98/582 (16.8)

*Such as Chemotherapy, Hypercalcemia, Reaction to chemotherapy medications, Packed cell injection, ATLL relapse

** Such as Corticosteroid pulse, HAM/TSP exacerbation

*** Such as NHL, DLBL, Lung cancer, Thyroid cancer, Pancreatic cancer, Multiple myeloma

^ Such as HTLV-1 induced- polyneuropathy, Neurogenic bladder

# Such as Systemic Lupus Erythematous, Wegner Granulomatosis, Still Disease, Dermatomyositis, Inclusion body myositis

@ Such as Acute coronary syndrome, Heart failure, Ischemic Heart Disease, Deep Venous Thrombosis, Pulmonary Thrombo-Embolism

Causes of infectious complications

We detailed the causes of infectious complications among the 582 hospital admissions. Infectious complications were identified in 23.7% (138/582) of the admissions. Among these, pleuropulmonary infections were the most common, affecting 39.1% (54/138) of patients. Urinary tract infections were the second most frequent, present in 23.2% (32/138) of cases. Gastrointestinal infections accounted for 8.7% (12/138) of the infections. Skin and soft tissue infections (SSTIs) were observed in 7.2% (10/138) of patients. Primary bacteremia and neutropenic fever each represented 4.3% (6/138) of infectious complications. Sepsis of unknown source was found in 2.9% (4/138) of the cases. Brain infections and Invasive fungal infection occurred in 1.4% (2/138) of the patients. The remaining 7.2% (10/138) of cases were classified as “other” types of infections (Tables 3 and 4).

Table 3.

Infectious complications in HTLV-1-incted patients

Infectious complications
Pleuropulmonary infections	54/138 (39.1)
Urinary tract infections	32/138 (23.9)
GI infections	12/138 (8.7)
SSTIs	10/138 (7.2)
Primary bacteremia	6/138 (4.3)
Neutropenic fever	6/138 (4.3)
Sepsis of unknown source	4/138 (2.9)
Brain infection	2/138 (1.4)
Invasive fungal infection	2/138 (1.4)
Others	10/138 (7.2)

GI: Gastrointestinal; SSTI: Skin and soft tissue infection

Table 4.

Isolated Pathogens, Clinical Syndromes, and Associated Underlying Conditions in HTLV-1-Infected Patients

Isolated pathogens	Number of episodes	Clinical syndromes	Underlying conditions
E. coli	17	Bacteremia, UTI, infected surgical wound, infected pressure ulcer,	ATL, chemotherapy, myelopathy
Cystoisospora belli	7	GI infection	ATL, chemotherapy
M. tuberculosis	5	Pneumonia
SARS-CoV-2	5	Pneumonia	DM
Candida species	4	Esophagitis, UTI, pneumonia	ATL, myelopathy, chemotherapy
Acinetobacter species	4	Bacteremia, pneumonia, UTI, infected pressure ulcer,	ATL, chemotherapy, myelopathy
Klebsiella pneumoniae	3	Bacteremia, UTI	Myelopathy, DM, HB
HSV	3	Gingivostomatitis, encephalitis, esophagitis	ATL, chemotherapy, DM
Mucor	2	Sinusitis	ATL, chemotherapy, DM
CMV	2	Pneumonia	ATL, chemotherapy
PJ	2	Pneumonia	ATL, chemotherapy
Enterococcal species	2	Bacteremia, UTI	ATL, chemotherapy, myelopathy
Enterobacter species	1	Bacteremia,	HB
CoNS	1	Bacteremia,	ATL, chemotherapy, myelopathy
Salmonella typhi	1	Bacteremia,	ATL
NTM	1	pneumonia
S. aureus	1	Infected pressure ulcer	ATL, chemotherapy, myelopathy
Pseudomonas species	1	UTI	DM, myelopathy
Proteus mirabilis	1	Infected pressure ulcer	Myelopathy
S. lugdunensis	1	Pneumonia	ATL, chemotherapy

UTI: Urinary tract infection; ATL: Adult T-cell leukemia; GI: Gastro intestinal; DM: Diabetes mellitus; HB: Hepatitis B

Factors associated with in-hospital mortality

Table 5 presents the findings from the univariable and multivariable logistic regression analyses, detailing the associations of various factors with mortality.

Table 5.

Univariable and multivariable Logistic Regression Analyses: Factors associated with in-hospital mortality in HTLV-1-infected patients

Variable	Univariate analysis OR (95% CI)	P-value	Multivariable analysis OR (95% CI)	P-value	Reference / Unit of comparison
Age	1.015 (1.001–1.028)	0.030	1.000 (0.983–1.018)	0.961	Per 1-year increase
Sex (male)	0.560 (0.364–0.860)	0.010	0.725 (0.402–1.308)	0.286	Ref = female
Peripheral leukocytes	1.016 (1.007–1.025)	< 0.001	1.035 (1.016–1.055)	< 0.001	Per 1,000 / µL increase
Polymorphonuclears	1.018 (1.004–1.033)	0.013	1.034 (1.017–1.051)	< 0.001	Per 1% increase
Hemoglobin level	0.887 (0.811–0.969)	0.008	0.900 (0.801–1.011)	0.076	Per 1 g/dL increase
Platelet count	0.994 (0.992–0.997)	< 0.001	0.996 (0.993–0.999)	0.003	Per 1,000 / µL increase
Diabetes mellitus (yes)	1.565 (0.852–2.873)	0.170	—	—	Ref = no diabetes
HBV co-infection (yes)	3.000 (1.275–7.060)	0.023	3.823 (1.130–12.927)	0.031	Ref = no HBV co-infection
Myelopathy (present)	0.403 (0.236–0.685)	< 0.001	0.560 (0.276–1.135)	0.108	Ref = no myelopathy
Malignancy (present)	2.426 (1.572–3.745)	< 0.001	1.123 (0.239–5.278)	0.883	Ref = no malignancy
Chemotherapy (yes)	2.301 (1.484–3.566)	< 0.001	0.996 (0.204–4.862)	0.997	Ref = no chemotherapy
Diagnosis of infection as primary cause	1.918 (1.058–3.479)	0.032	—	—	Ref = non-infectious cause
Length of hospital stay	1.047 (1.024–1.071)	< 0.001	1.053 (1.022–1.085)	0.001	Per 1 day increase

Continuous variables are expressed per incremental unit of measurement (leukocytes and platelets = 1,000 cells/µL; hemoglobin = 1 g/dL; polymorphonuclears = 1% point; hospital stay = 1 day). For categorical variables, the reference groups are: female (sex), no diabetes, no HBV co-infection, no myelopathy, no malignancy, no chemotherapy, and non-infectious cause of admission. OR: odds ratio; CI: confidence interval; HBV: hepatitis B virus

In the univariable analysis, several variables demonstrated significant associations with mortality. For each additional year of age, the risk of mortality increased by 1.5% (OR = 1.015, 95% CI: 1.001–1.028, p = 0.030). Elevated levels of peripheral leukocytes and higher percentages of polymorphonuclear leukocytes were significantly associated with increased mortality. Specifically, each 1000/µl increase in peripheral leukocytes increased the risk of mortality by 1.6% (OR = 1.016, 95% CI: 1.007–1.025, p < 0.001), while each additional percentage unit of polymorphonuclear leukocytes increased the risk by 1.8% (OR = 1.018, 95% CI: 1.004–1.033, p = 0.013). Conversely, a lower platelet count was associated with higher mortality, with each additional 1000/µl decrease in platelet count increasing the risk of mortality by 0.6% (OR = 0.994, 95% CI: 0.992–0.997, p < 0.001). Hemoglobin levels also showed a significant association with mortality; each mg/dl decrease in hemoglobin level was associated with a 1.3% increase in mortality risk (OR = 0.887, 95% CI: 0.811–0.969, p = 0.008). Additionally, the presence of myelopathy was associated with a reduced mortality risk, with an odds ratio of 0.403 (95% CI: 0.236–0.685, p < 0.001). Malignancy and chemotherapy were both linked to higher mortality risks, with odds ratios of 2.426 (95% CI: 1.572–3.745, p < 0.001) and 2.301 (95% CI: 1.484–3.566, p < 0.001), respectively. A longer length of hospital stay was associated with a higher risk of mortality, with each additional day increasing the mortality risk by 4.7% (OR = 1.047, 95% CI: 1.024–1.071, p < 0.001).

The multivariable logistic regression analysis identified several significant factors associated with in-hospital mortality among HTLV-1-infected patients. Length of hospital stay was significantly associated with mortality. Each additional day of hospitalization increased the odds of mortality by 5.3% (OR = 1.053, 95% CI: 1.022–1.085, p = 0.001). Peripheral leukocyte count was another significant predictor of mortality. Each increase of 1,000 leukocytes per microliter was associated with a 3.5% increase in the odds of mortality (OR = 1.035, 95% CI: 1.016–1.055, p < 0.001). Polymorphonuclear cell percentage also showed a significant relationship with mortality. A 1% increase in Polymorphonuclears was associated with a 3.4% increase in the odds of mortality (OR = 1.034, 95% CI: 1.017–1.051, p < 0.001). Thrombocytopenia was significantly associated with mortality. Each decrease of 1,000 platelets per microliter was associated with a 0.4% increase in the odds of mortality (OR = 0.996, 95% CI: 0.993–0.999, p = 0.003). HBV co-infection was associated with a significantly higher risk of mortality. Patients with HBV co-infection had nearly 3.8 times the odds of mortality compared to those without HBV (OR = 3.823, 95% CI: 1.130–12.927, p = 0.031). Diagnosis of infection as the primary cause for hospital admission was significantly linked to an increased risk of mortality, with an odds ratio of 1.918 (95% CI: 1.058–3.479, p = 0.032).

Survival probability of HTLV-1-infected patients following discharge from the hospital

Of the 308 individuals infected with HTLV-1, 98 died during hospitalization. Post-discharge follow-up was completed for 95 patients by the end of the study, while the remaining individuals were lost to follow-up. Among those discharged, the median survival time was 32.97 months (95% CI: 19.42–46.51 months), indicating that half of these patients survived at least this long after leaving the hospital.

Discussion

This study provides a comprehensive analysis of the demographic and clinical characteristics, causes of hospital admission, and survival outcomes for 308 patients with HTLV-1 infection. In-hospital mortality was approximately 17%, emphasizing the substantial risk of mortality during their hospital stay. Among patients discharged from the hospital, the median survival time was 32.97 months (95% CI: 19.42–46.51 months), indicating that half of these patients survived at least this long after discharge. These findings highlight that, despite initial recovery from acute illness, long-term outcomes remain poor, emphasizing the need for enhanced post-discharge care and extended monitoring as the heightened mortality risk persists well beyond hospitalization. In a recent systematic review and meta-analysis of epidemiological associations between HTLV-1 and adverse health outcomes, strong evidence was identified linking HTLV-1 infection to an increased risk of premature death. The significant overall increase in mortality observed was not fully explained by the risk factors typically associated with HTLV-1. Most of the conditions studied were not fatal, and the low incidence of adult T-cell leukemia-lymphoma (ATL), despite its high case-fatality rate, did not account for the observed mortality increase. The authors concluded that the clinical conditions underlying the elevated risk of non-neoplastic diseases in HTLV-1 carriers may be diverse [1].

Multivariable logistic regression analysis identified several factors significantly associated with in-hospital mortality. Prolonged hospitalization was associated with increased mortality risk, with each additional day raising the odds of death by 5.3% (OR = 1.053, 95% CI: 1.022–1.085, p = 0.001). Elevated levels of peripheral leukocytes and higher percentages of polymorphonuclear leukocytes were linked to increased mortality. Thrombocytopenia was a significant predictor as well. HBV co-infection significantly increased the risk of mortality, with nearly four times the odds compared to non-HBV patients (OR = 3.823, 95% CI: 1.130–12.927, p = 0.031), underscoring the severe impact of HBV co-infection on survival outcomes. The diagnosis of infection as the primary cause of hospital admission was also significantly associated with increased mortality (OR = 1.918, 95% CI: 1.058–3.479, p = 0.032).

Despite the relatively low prevalence of HBV and HCV co-infection in this cohort, HBV co-infection was associated with a markedly higher risk of mortality, highlighting that even infrequent co-infections can substantially impact outcomes and should be considered important in the management of HTLV-1–infected patients. This association should be interpreted with caution, as detailed liver function data were not uniformly available, preventing assessment of whether hepatic dysfunction contributed to the increased mortality risk. The finding likely reflects an association rather than a causal relationship and may relate to overlapping risk factors or immunological interactions between HTLV-1 and HBV infections, warranting further investigation. HTLV-1, as both a blood-borne and sexually transmitted infection, shares transmission routes with several other pathogens, including HBV, HCV, and HIV. Although HTLV-1 is globally distributed, it predominantly affects individuals in low- and middle-income tropical regions where the prevalence of other infectious agents is elevated, thereby predisposing to co-infection [5]. Previous studies have reported a high prevalence of HBV and HCV among individuals with HTLV-1 infection, and while the impact of HTLV-1 on HCV outcomes remains debated, most evidence suggests it may accelerate the clinical progression of hepatitis C [5]. Additionally, individuals with anti-HCV antibodies have been reported to be up to 20 times more likely to be co-infected with HTLV-1 [5].

In our cohort, however, no cases of HIV co-infection were identified. Iran is a country with relatively low HIV prevalence in general population [6, 7], and the predominant route of HTLV-1 transmission is vertical, through prolonged breastfeeding [8], which contrasts with many other endemic regions where sexual or parenteral routes are more common [9]. Consequently, the prevalence of HIV among HTLV-1 carriers is expectedly low. Nonetheless, the possibility of missed HIV/HTLV-1 co-infections cannot be excluded, as repeat HIV testing was not performed in all patients during hospitalization. This limitation may have led to underestimation of the true burden of viral co-infection in this population.

In our study, univariable—but not multivariable—analysis showed that HTLV-1–infected patients with HAM/TSP had better survival outcomes compared to those hospitalized without HAM/TSP. However, this finding should not be interpreted as indicating a protective or prognostic effect of myelopathy itself. Rather, it likely reflects differences in the clinical context of hospitalization. Many patients with HAM/TSP were admitted for exacerbations of chronic neurological disability or for supportive care, whereas those without HAM/TSP were more often hospitalized for severe systemic illnesses such as adult T-cell leukemia/lymphoma or sepsis, which inherently carry higher mortality. Due to the retrospective design and limited sample size, subgroup analysis could not be performed to further explore this observation; thus, the association should be regarded as contextual and non-causal.

This result contrasts with findings from the São Paulo study, where HTLV-1–infected individuals with HAM/TSP had a significantly higher mortality risk compared to asymptomatic HTLV-1 carriers, showing an approximately fivefold increase in the risk of death, particularly in the presence of co-infections such as HIV or HCV [10]. The divergent outcomes between the two studies likely reflect differences in the comparison groups—our study comparing hospitalized patients with and without HAM/TSP, versus the Brazilian study comparing symptomatic myelopathy cases to asymptomatic individuals. Taken together, these observations suggest that while HAM/TSP represents a serious neurological manifestation of HTLV-1 infection, its apparent prognostic impact on mortality depends on the reference population and clinical setting, underscoring the importance of context in interpreting survival outcomes.

Some other factors such as malignancy and chemotherapy showed significant associations with mortality in univariate analysis but did not remain significant in the multivariable model after adjusting for confounders. These findings should therefore be interpreted as associations observed in unadjusted analysis.

The study highlights a diverse range of causes for hospital admissions among HTLV-1 infected patients, with a significant portion attributed to infectious complications, highlighting the susceptibility of HTLV-1 infected patients to severe infections. Myelopathy and ATL also emerged as prominent causes, reflecting the severe manifestations of HTLV-1 infection itself. In-hospital mortality rate was 16.8%, with about half of patients experiencing multiple hospital admissions. Several other studies have explored the causes of hospital admission in patients with HTLV-1 infection across different regions. In a study conducted by Ramos et al. in Spain between 1997 and 2015, involving 135 patients (115 with HTLV-1 and 20 with HTLV-2), the most common causes of hospitalization were myelopathy (45.2%), leukemia/lymphoma (22.2%), solid organ transplants (10.4%), hepatitis C (7.4%), childbirth (5.2%), and parasitic infections (3.7%), including Strongyloides stercoralis (1.5%). The median length of hospital stay was 9 days, and approximately 8% of patients died during the study [11]. Another study by Adedayo et al. in Dominica (1995–1999) on 66 HTLV-1-infected patients reported that the leading causes of hospitalization were ectoparasitic/endoparasitic infections (40.9%), HAM/TSP (18.2%), lymphoma (15.2%), and ATL (7.6%). Notably, there were no fatalities reported in this cohort [12]. Additionally, a study by Einsiedel et al. in Australia (2000–2010) included 507 patients with 39,967 hospitalization episodes. The causes of admission were categorized into four groups: non-communicable diseases (e.g., smoking, alcohol use, chronic kidney and liver disease, malignancies), infections (e.g., sepsis, pneumonia, skin infections), respiratory conditions (e.g., bronchiectasis, COPD, asthma), and HTLV-1-related conditions (e.g., Strongyloides stercoralis infection, HAM/TSP, ATL, and infectious dermatitis) [13]. These studies underscore the diverse clinical presentations and regional variations in the causes of hospital admission among patients with HTLV-1 infection. Notably, infectious complications consistently emerge as a significant cause of hospitalization in HTLV-1-infected individuals.

There is substantial evidence showing that patients with ATL experience severe immunosuppression, which is linked to opportunistic infections and other malignancies [5]. It is a highly immunosuppressive malignancy, with a significantly elevated incidence of infections per patient-year [14, 15]. HTLV-1 can also cause immune dysfunction even in the absence of clinical disease [5]. There is substantial in vitro and clinical evidence indicating immune dysfunction in asymptomatic HTLV-1 carriers. This evidence includes an elevated risk of acute and chronic infections by various pathogens, increased rates of tuberculosis reactivation, case reports of opportunistic infections, and diminished cell-mediated immunity as demonstrated by delayed-type hypersensitivity (DTH) skin testing [16]. In fact, modest to severe immune suppression is observed in otherwise healthy HTLV-1 carriers [17]. When opportunistic infections occur in HTLV-1-positive patients, it is important to consider the potential progression from asymptomatic HTLV-1 infection to smoldering ATL [17]. The mechanisms behind immunosuppression in HTLV-1-infected individuals are unclear, but it is suggested that reduced T lymphocyte production in the thymus may contribute to immunodeficiency [17]. HTLV-1 infection predominantly affects CD4 + T cells, especially CD4 + CD25+ regulatory T cells, which are major reservoirs for the HTLV-1 provirus. This preferential infection may contribute to inflammation and play a role in developing HTLV-1-associated conditions like ATL. Significant immunosuppression in HTLV-1-infected individuals is thought to be linked to increased CD4 + CD25+ suppressor T cell phenotypes. Elevated counts of these regulatory T cells correlate with higher susceptibility to Strongyloides hyperinfection. Moreover, Foxp3 expression by CD4 + CD25+Foxp3 + regulatory T cells is lower in HTLV-1-infected individuals compared to healthy controls [17]. In addition to these immune alterations, the proviral load (PVL) plays a central role in determining the clinical course of HTLV-1 infection and the progression from asymptomatic carrier status to HAM/TSP or ATL. Higher PVL is consistently associated with increased disease activity and worse clinical outcomes. Although CD8⁺ T cells carry only approximately 5% of the total PVL, they play a crucial role in viral persistence and neuropathology in HAM/TSP, and their expansion is correlated with higher PVL and failure to control viral replication. This ineffective antiviral response contributes to sustained immune activation and tissue injury [18].

Moreover, in HAM/TSP, HTLV-1 infection leads to a reduction of the Th17 effector subset, further impairing host defense mechanisms. This shift in T-helper cell balance results in a dominant Th1-skewed immunophenotype, characterized by excessive production of pro-inflammatory cytokines. Rather than providing effective viral control, this dysregulated immune response contributes to chronic inflammation, tissue damage, and progressive neurological dysfunction, thereby worsening disease severity and long-term outcomes.

Increasing evidence suggests that HTLV-1 infection may influence the risk and severity of several other infectious diseases, including opportunistic infections [16]. HTLV-1 is predominantly found in low- and middle-income countries (LMICs), regions that also have high rates of other infections such as Strongyloides stercoralis and Mycobacterium tuberculosis. This overlap means that HTLV-1-infected individuals are often co-infected with multiple pathogens, leading to complex interactions that can influence the progression of both HTLV-1 and the coexisting infections. The immune response to HTLV-1 is typically Th1-dominant, whereas an effective response to some parasites, like helminths, generally requires a Th2-dominant response. Consequently, the immune system’s reaction to co-infections can be conflicting. Additionally, the clinical manifestations of these co-infections may either exacerbate or occasionally benefit the host, depending on how the immune responses interact [5]. Reports have identified cases of CMV gastritis, gastroenterocolitis, Pneumocystis pneumonia (PCP), and disseminated herpes zoster in HTLV-1 carriers. Additionally, infections such as pulmonary histoplasmosis and cryptococcal meningitis have been observed [16]. In the case of strongyloidiasis, coinfection with HTLV-1 appears to result in a higher rate of chronic carriage, increased parasite load, and a risk of more severe infection [16]. Frequent reports indicate hyperinfection with Strongyloides stercoralis in HTLV-1 carriers, linked to a selective defect in the humoral response to the organism [16].

Among the 582 hospital admissions in our cohort, 23.7% were due to infectious complications, with pleuropulmonary infections being the most prevalent, affecting 39.1% of these cases. This high incidence underscores the vulnerability of HTLV-1 infected patients to respiratory illnesses, likely exacerbated by their immunocompromised status. In a study from Australia, HTLV-1 infection was found to be significantly associated with higher incidences of respiratory conditions including asthma (coefficient 0.99; 95% CI, 0.27–1.7), lower respiratory tract infections (coefficient 0.19; 95% CI, 0.04–0.34), and bronchiectasis (coefficient 0.60; 95% CI, 0.02–1.18) [13]. Several case-control studies have investigated pulmonary infections, such as pneumonia [19] and pulmonary tuberculosis [20–22], in HTLV-1-infected individuals. In the study by Atsumi et al., S. pneumoniae and Haemophilus influenzae were the most frequently isolated microorganisms in both HTLV-1-positive and HTLV-1-negative patients with community-acquired pneumonia. However, P. jiroveci was the third most frequent organism in HTLV-1-positive patients, while Moraxella catarrhalis was more common in HTLV-1-negative patients. Additionally, Klebsiella pneumoniae tended to be more frequent in HTLV-1-positive patients [19]. In our study, tuberculosis accounted for 3.6% of the infectious complications among hospitalized HTLV-1-infected patients. Previous research from Japan and Brazil has also reported an increased risk of tuberculosis among individuals with HTLV-1 [21, 22]. Grassi et al. found that HTLV-1 infected patients had a relative risk of 2.6 (95% CI: 1.6–4.2) for developing tuberculosis compared to those who were not infected with HTLV-1 [23]. Patients with a history of tuberculosis were twice as likely to be infected with HTLV-1, whereas those with active tuberculosis had a threefold higher likelihood of being co-infected with HTLV-1 [5]. However, not all studies have shown consistent results [24]. In Bissau, HTLV-1 infection was found to be more prevalent among tuberculosis patients compared to a general population control group. However, this difference was not significant when specifically looking at HIV-negative patients with tuberculosis. It was suggested that while HTLV-1-related immunosuppression alone may not significantly increase the risk of tuberculosis, it could contribute to higher risk of tuberculosis in individuals already infected with HIV. Additionally, higher mortality was observed in patients with pulmonary tuberculosis who were co-infected with HIV-2 and HTLV-1 compared to those who were HIV-2-positive but HTLV-1-negative [25]. Taken together, these observations indicate that the clinical course of HTLV-1 infection is shaped by a complex interaction of viral, host, and environmental factors. Proviral load, host genetic background—particularly specific HLA alleles such as HLA-DRB113, HLA-A28, B54, C07, DQB103:01, and DRB107:01—and the presence of co-infections jointly influence susceptibility to disease progression, hospitalization, and mortality. Certain HLA types are associated with more effective immune control of HTLV-1, whereas others predispose to higher PVL and increased risk of HAM/TSP or ATLL. These factors help explain why clinical outcomes vary widely among HTLV-1-infected individuals and why prolonged hospitalization, co-infections, and laboratory abnormalities alone cannot fully account for differences in prognosis [26].

In our study, urinary tract infections (UTIs) were identified as the second most common infectious complication, accounting for 23.9% of admissions. Evidence shows that HTLV-1-infected individuals have notably higher rates of bladder and kidney infections. However, Murphy et al. noted that this association should be interpreted with caution due to the established connection between HTLV-1 and HAM/TSP. Symptoms such as urinary frequency and urgency, which often signal bladder hyperreactivity from myelopathy, may prompt HTLV-1-infected individuals to seek medical attention. Consequently, diagnoses of UTIs or kidney disease might result from unrecognized neurogenic bladder dysfunction, or patients may be treated presumptively for UTIs based on their symptoms [27]. A cross-sectional study of urinary symptoms in HTLV-1-infected individuals revealed that while urinary symptoms indicative of UTIs are common, actual UTIs are relatively infrequent. The study suggested that neurogenic bladder is the predominant cause of urinary symptoms in this group [28]. Nonetheless, HTLV-1-infected individuals, especially those with myelopathy, are likely to experience a higher rate of urinary infections due to bladder dysfunction [29]. In a 21-year study from São Paulo, Brazil, researchers followed 727 HTLV-1-infected individuals, including 248 with HAM/TSP. Over 3,800 person-years, 27 patients died, with sepsis from UTIs and decubitus ulcers being the leading cause of death in HAM/TSP patients [10].

Gastrointestinal (GI) infections accounted for 8.7% of the infectious complications in our cohort. The GI infections were linked to pathogens such as Cystoisospora belli, emphasizing the diverse etiological factors contributing to gastrointestinal symptoms in this population. Several reports have documented cases of cystoisosporiasis in HTLV-1-infected patients, including those with ATL and asymptomatic HTLV-1 carriers from various regions around the world [30–35]. Most of these patients exhibited debilitating, profuse watery diarrhea, which was prone to frequent relapses and frequently led to malabsorption issues. One of our patient, a 90-year-old man with ATL in remission, began experiencing symptoms of cystoisosporiasis while undergoing chemotherapy. He was diagnosed after a year of persistent chronic diarrhea, which resulted in a weight loss of approximately 30 kg. Although he initially responded well to treatment, he experienced multiple relapses of cystoisosporiasis and ultimately succumbed to a relapse of ATL. HTLV-1 infection may affect the persistence of other parasites within the co-infected host [13]. This is evident from the increased rates of treatment failure for Schistosoma mansoni and Strongyloides stercoralis observed in individuals infected with HTLV-1 [5]. HTLV-1 infection was found to be significantly associated with more frequent hospital admissions for strongyloidiasis (coefficient 0.563; 95% CI, 0.17–0.95) [13]. Co-infection with HTLV-1 and Strongyloides stercoralis leads to altered immune responses to the parasite. In these patients, elevated IFN-γ levels suppress the production of IL-4, IL-5, IL-13, and IgE, which are crucial for defense against helminths. This altered immune response reduces the effectiveness of treatment for Strongyloides stercoralis and contributes to a higher prevalence of strongyloidiasis among HTLV-1-infected individuals. Consequently, HTLV-1 infection emerges as a major risk factor for severe, disseminated strongyloidiasis [36].

Skin and soft tissue infections (SSTIs) and primary bacteremia were less common but still significant, affecting 7.2% and 4.3% of our patients, respectively. An association between HTLV-I infection and Staphylococcus- and Streptococcus-related infective dermatitis among Jamaican children has been identified [37]. Neutropenic fever, sepsis of unknown source, brain infections, and invasive fungal infections each represented a smaller proportion of the infectious complications but are critical due to their severity and the complexity of their management. In a study from Australia, HTLV-1 infection was found to be significantly associated with increased rates of bloodstream infections (BSIs) (adjusted negative binomial regression, coefficient 0.21; 95% CI, 0.02–0.41) [13].

Our study, conducted across two major referral centers in Mashhad, Iran—a region known for its high HTLV-1 endemicity—provides valuable insights into the causes of hospitalization and long-term outcomes for HTLV-1–infected patients. One notable strength is the comprehensive nature of the cohort, which encompasses a diverse patient population and reflects the endemic characteristics of the region.

Nonetheless, this study has several important limitations. First, its retrospective design may have led to incomplete data capture, information bias, and residual confounding. Because the cohort was drawn from tertiary referral hospitals that predominantly manage patients with severe HTLV-1–related complications such as HAM/TSP and ATL, a selection bias toward more advanced or complicated cases is likely. Consequently, the findings primarily reflect the clinical features and outcomes of hospitalized individuals and cannot be generalized to community-based or asymptomatic HTLV-1 carriers.

Second, the study cohort included a clinically diverse population with distinct prognoses, comprising hematologic and neurologic presentations (e.g., ATL and HAM/TSP) as well as other complications. Due to small subgroup sizes, separate analyses were not feasible. Therefore, mortality factors were analyzed collectively to identify broad associations among hospitalized HTLV-1-infected patients. The results should be interpreted as descriptive trends within this mixed population rather than disease-specific predictors.

Third, pre-infection health status and the temporal relationship between HTLV-1 acquisition and comorbidities could not be determined due to the retrospective nature of the data. Although major underlying conditions (diabetes mellitus, malignancy, myelopathy, and hepatitis B co-infection) were included in the multivariable model to mitigate confounding, unmeasured or residual confounders may still exist.

Fourth, one of the most important limitations of this study is the incomplete post-discharge follow-up. Of 210 discharged patients, survival outcomes were available for only 95 (45.2%). This loss to follow-up may have affected the accuracy of survival estimates and introduced selection bias, as patients who remained in contact with healthcare services may differ systematically from those lost to follow-up. Nonetheless, the available data provide meaningful insight into post-hospitalization outcomes for nearly half of the discharged cohort, highlighting the poor long-term prognosis in this population. As this was a hypothesis-generating exploratory study, survival findings should be interpreted as descriptive and not as evidence of causality or population-level survival rates.

Fifth, the absence of a non-HTLV-1 control group of hospitalized patients limits the ability to directly compare causes of admission and outcomes and restricts causal inference. Additionally, incomplete laboratory data—such as missing liver function tests—further constrained mechanistic interpretation.

Finally, the combination of retrospective design, selection bias, incomplete follow-up, and lack of a control group collectively limits the generalizability of the findings. Nevertheless, this investigation represents one of the largest systematic description to date of hospitalized HTLV-1–infected patients in an endemic region and provides valuable exploratory insights into their clinical spectrum, causes of admission, and mortality-associated factors. The results should be regarded as hypothesis-generating observations that form a foundation for future prospective, population-based studies with standardized data collection and appropriate comparison groups.

Conclusion

In conclusion, this study highlights the significant impact of hospital admission on the survival outcomes of HTLV-1-infected patients. The in-hospital mortality rate of 17% and the median post-discharge survival of approximately 33 months underscore the severe clinical trajectory of HTLV-1 infection. The identification of factors significantly associated with mortality, including prolonged hospitalization, elevated leukocyte levels, thrombocytopenia, and HBV co-infection, can help anticipate the risk of early mortality during the hospital stay and guide targeted interventions. These findings provide valuable insights for improving patient management, emphasizing the need for early intervention and robust post-discharge care to mitigate the persistent long-term mortality risk in this population.

Abbreviations

HTLV-1

Human T-cell lymphotropic virus type 1

ATL

Adult T-cell leukemia-lymphoma

HAM/TSP

HTLV-1-associated myelopathy/tropical spastic paraparesis

IQR

Interquartile ranges

HBV

Hepatitis B virus

HCV

Hepatitis C virus

HIV

Human immunodeficiency virus

ATLL

Adult T-cell leukemia/lymphoma

CI

Confidence interval

OR

Odds ratio

LOS

Length of stay

LMICs

Low- and middle-income countries

PCP

Pneumocystis pneumonia

UTIs

Urinary tract infections

BSIs

Bloodstream infections

Author contributions

F.M., M.H., M.S., and F.S. conceived of the presented idea. F.M., M.H., M.S., and F.S. wrote the manuscript. F.M., M.H., M.S., and F.S. contributed to the design and implementation of the research. F.S. and M.H. prepared Figs. 1 and 2 with support from F.M., and M.S. All authors reviewed the manuscript.

Funding

None.

Data availability

The data that support the findings of this study are available on request from the corresponding author.

Declarations

Ethics approval and consent to participate

All patients or their legally authorized representatives provided informed consent. This study received approval from the ethics committee of Mashhad University of Medical Sciences under the code IR.MUMS.MEDICAL.REC.1401.318 and was performed by the ethical standards set forth by the Helsinki Declaration.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.