Introduction

Leptospirosis is a globally distributed zoonotic disease caused by spirochete bacteria of the genus Leptospira, with a wide clinical spectrum ranging from subclinical infection to severe disease with multiorgan failure. Transmission to humans occurs through direct or indirect exposure to urine from infected wild or domestic animals, particularly rodents, dogs, pigs, and cattle. In reservoir hosts, Leptospira colonizes the proximal renal tubules and is excreted in urine, contaminating soil and water. In settings with inadequate sanitation, frequent flooding, and precarious housing, these environmental conditions favor bacterial persistence and increase the risk of human exposure [1,2].

Leptospira has worldwide distribution, with the highest incidence in tropical and subtropical regions, where it contributes substantially to febrile illness morbidity and mortality. Regions with the greatest estimated burden include Sub-Saharan Africa, Latin America and the Caribbean, South and Southeast Asia, and Oceania. Global incidence exceeds one million cases annually, with approximately 60,000 deaths [3]. In the Peruvian Amazon, leptospirosis is endemic, with epidemic surges occurring after periods of heavy rainfall and flooding. In Loreto, outbreaks and increased transmission have been documented in association with Amazon River overflows, and national and regional surveillance in 2023 reported a marked increase in cases in Loreto during the post-rainy season, including outbreak investigations in affected localities. El Niño-related rainfall and flooding events in Peru have likewise been associated with increased leptospirosis transmission [4,5]. As of epidemiological week 52 of 2024, Peru had reported 9,520 leptospirosis cases, an incidence of 27.91 per 100,000 inhabitants, and 16 confirmed deaths, with a substantial proportion concentrated in high-burden Amazonian departments, particularly Loreto, Ayacucho, Madre de Dios, and Ucayali [6]. In this high-burden setting, where many suspected cases are initially assessed outpatient primary-care settings, understanding age-stratified clinical profiles is relevant to improve early clinical suspicion, guide targeted diagnostic testing, and support timely treatment [7].

Age has been associated with clinically meaningful differences in leptospirosis presentation and outcomes. Early manifestations are often nonspecific and overlap with other causes of acute febrile illness in tropical settings, including dengue, malaria, rickettsial infections, and bacterial sepsis, which may delay recognition and appropriate management. Defining age-stratified clinical profiles may therefore help strengthen early clinical suspicion in endemic areas [8–11]. Classic severe manifestations may be less frequently observed in pediatric patients than in adults, but they may also present with more nonspecific or atypical patterns that are harder to recognize in primary health care centers [12,13]. In early childhood, clinical assessment may be complicated by limited self-reporting of subjective symptoms, and children with leptospirosis may less frequently exhibit the classic severe features seen in adults, such as jaundice and overt renal dysfunction, thereby reducing early diagnostic suspicion at the first point of care [13–16].

Belén, a district of Maynas Province in Loreto, experiences seasonal river flooding, inadequate sanitation, precarious housing, and limited access to safe water, all of which create favorable conditions for leptospirosis transmission. These environmental and social conditions make Belén a priority area for leptospirosis surveillance and public health intervention.

Although leptospirosis has been extensively described in general populations, relatively few studies have systematically compared clinical and epidemiological differences across age groups, particularly in outpatient or primary health care centers and among children under 5 years of age. This gap is especially relevant in the Peruvian Amazon, where early presentations frequently overlap with other causes of acute febrile illness and where most suspected cases are managed initially at the first level of care. Therefore, this study aimed to describe and compare the clinical manifestations and epidemiological characteristics of confirmed leptospirosis cases in patients under 5 years and those aged 5 years or older evaluated at four primary health care centers in Belén, Loreto. By defining age-stratified outpatient profiles in this setting, the findings may inform prevention strategies and support diagnostic suspicion and initial clinical decision-making at the primary health care centers.

Methods

Study setting

The study was conducted in four first-level Ministry of Health primary health care centers serving Belén and nearby urban and urban-marginal sectors of Iquitos: 6 de Octubre health center, 9 de Octubre health center, Cardozo health center, and Belén health center (Fig 1). These facilities provide primary care in areas characterized by high population density, frequent flooding, and environmental conditions favorable to tropical infectious diseases. 6 de Octubre health center is a category I-3 health center that also provides 24-hour emergency services and participates in epidemiological surveillance, whereas the other participating centers provide routine first-level care, generally with more limited emergency coverage. These differences in service organization and catchment characteristics should be considered when interpreting the distribution of confirmed cases across centers.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Geographic location of the study area and included primary health centers in Loreto, Peru.

https://doi.org/10.1371/journal.pntd.0013473.g001

Belén is a district of the province of Maynas, Loreto region, in the northeastern Peruvian Amazon. It is among the most socially and environmentally vulnerable urban areas of Iquitos, with an estimated population of approximately 76,000 inhabitants, around 75% of whom live in flood-prone areas, particularly in the lower part of the district along the Itaya River. The area has a warm, humid, rainy tropical climate, with temperatures ranging from 17°C to 38°C and annual rainfall exceeding 2,600 mm [17]. During seasonal river flooding, many households become partially submerged, creating unsanitary conditions due to water stagnation, inadequate waste disposal and limited sanitation infrastructure. These conditions increase exposure to contaminated water and other environmental hazards, facilitating the transmission of infectious diseases. Children under 5 years of age in Loreto also bear a high burden of other common conditions, including anemia, acute respiratory infections, diarrheal disease, and parasitic infections [18–20]. In this context, the environmental and social conditions of Belén, including seasonal flooding, limited access to potable water, inadequate sanitation, and precarious housing, may facilitate leptospirosis transmission and support the need for surveillance and public health interventions.

Population and sample

The study population consisted of patients evaluated for suspected leptospirosis at four primary health care centers in Belén during the study period. A total of 2,325 patients with clinical suspicion of leptospirosis were identified. Of these, 1,920 were excluded because leptospirosis was not confirmed by MAT, either due to a negative result or because the test was not performed, and 5 additional records were excluded because they contained insufficient information for analysis. The final analytical sample included 400 patients with MAT-confirmed leptospirosis (Fig 2). No formal sample size calculation was performed because the study included all eligible MAT-confirmed cases with sufficient information during the study period. Therefore, subgroup comparisons, particularly those involving children under 5 years of age, should be interpreted as exploratory.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Flow diagram of the inclusion and exclusion of patients with suspected leptospirosis evaluated at four primary health centers.

https://doi.org/10.1371/journal.pntd.0013473.g002

Patients were eligible for inclusion if they had a MAT-confirmed diagnosis of leptospirosis and sufficient information in the medical records and epidemiological forms for the variables included in the analysis. Confirmatory diagnosis was based on MAT performed at the National Institute of Health laboratory in Lima. A case was considered confirmed when a single serum sample showed a MAT titer of at least 1:800, with serogroup reactivity recorded. This threshold was selected in accordance with the Peruvian National Technical Health Standard for leptospirosis, which indicates that, when paired sera are not available, a single MAT titer ≥1:800 may confirm the diagnosis in the Peruvian Amazon [21]. Given the retrospective nature of this primary care study, paired acute and convalescent samples were not routinely available for all patients.

Results

There were 81 cases in 2022 (20.3%), 105 cases in 2023 (26.3%), and 214 cases in 2024 (53.5%) (Table 1). Thus, a higher number of confirmed cases were identified in records in 2024 than in prior years. However, as this was a retrospective review and the number of suspected cases evaluated was not available by year, these differences should not be interpreted as annual changes in incidence. Among children under 5 years, most confirmed cases were recorded in 2024 (20/28, 71.4%), whereas among patients aged 5 years or older, 194/372 (52.2%) were recorded in the same year. Cases were unevenly distributed across primary health centers: 6 de Octubre accounted for 288 of 400 confirmed cases (72.0%), followed by Cardozo with 71 (17.8%), 9 de Octubre with 37 (9.2%), and Belén with 4 (1.0%) (Table 1; Fig 3). Center-stratified demographic and clinical characteristics are presented in S1 Table. When monthly counts were aggregated across 2022–2024, patients aged 5 years or older showed the highest numbers of confirmed cases in March (n = 62), September (n = 49), and October (n = 40). Among children under 5 years, the total number of confirmed cases was small (n = 28), with low monthly counts throughout the year and a maximum of 5 cases in August (Fig 4).

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Epidemiological characteristics of patients with confirmed leptospirosis evaluated at primary health centers in the Peruvian Amazon, 2022–2024.

https://doi.org/10.1371/journal.pntd.0013473.t001

Download:

• PNG
  larger image
• TIFF
  original image

Fig 3. Kernel density map of MAT-confirmed leptospirosis cases by reporting primary health center in Belén, Loreto, Peruvian Amazon, 2022–2024.

https://doi.org/10.1371/journal.pntd.0013473.g003

Download:

• PNG
  larger image
• TIFF
  original image

Fig 4. Monthly distribution of confirmed leptospirosis cases aggregated across 2022–2024 by age group (<5 vs ≥5 years) in primary health centers in Belén, Loreto, Peruvian Amazon.

https://doi.org/10.1371/journal.pntd.0013473.g004

Among children under 5 years, 12 (42.9%) were from rural areas, 11 (39.3%) from urban areas, and 5 (17.9%) from urban-marginal areas. Among patients aged 5 years or older, urban residence predominated (163 cases, 43.8%), followed by urban-marginal (114, 30.6%) and rural areas (95, 25.5%). Residence in flood-prone areas was reported in 14 children under 5 years (50.0%) and 160 patients aged 5 years or older (43.0%). Females accounted for 18 of 28 cases (64.3%) among children under 5 years and 199 of 372 cases (53.5%) among patients aged 5 years or older (Table 1). Overall, the most frequent interval from symptom onset to consultation was 3–6 days (183 cases, 45.8%). Among children under 5 years, the largest proportion presented within 0–3 days (13/28, 46.4%), whereas among patients aged 5 years or older, the largest proportion presented within 3–6 days (171/372, 46.0%) (Table 2).

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Time from symptom onset to consultation among confirmed leptospirosis cases by age group (<5 vs ≥ 5 years), 2022–2024.

https://doi.org/10.1371/journal.pntd.0013473.t002

Regarding clinical presentation, the most frequently recorded symptoms in children under 5 years were malaise (22/28, 78.6%) and fever (20/28, 71.4%). Among patients aged 5 years or older, the most frequently recorded symptoms were fever (311/372, 83.6%), headache (247/372, 66.4%), malaise (190/372, 51.1%), and chills (133/372, 35.8%). Other symptoms more commonly recorded in patients aged 5 years or older included nausea/vomiting (23.1%), dizziness (22.6%), myalgia (16.4%), and musculoskeletal pain (10.2%) (Table 3).

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Distribution of clinical symptoms among confirmed leptospirosis cases by age group (<5 vs ≥ 5 years) in primary health centers in the Peruvian Amazon, 2022–2024.

https://doi.org/10.1371/journal.pntd.0013473.t003

The predominant MAT serogroup reactivity in both age groups was Varillal, detected in 24 of 28 children under 5 years (85.7%) and 316 of 372 patients aged 5 years or older (84.9%). Panama reactivity was identified in 2 children under 5 years (7.1%) and 5 patients aged 5 years or older (1.3%), whereas Hurstbridge reactivity was recorded only among patients aged 5 years or older (47 cases, 12.6%). Overall, 317 patients (79.3%) showed reactivity to a single serogroup, while 83 (20.8%) showed reactivity to two or more serogroups (Table 4).

Download:

• PNG
  larger image
• TIFF
  original image

Table 4. Distribution of MAT serogroup reactivity among confirmed leptospirosis cases by age group (<5 vs ≥ 5 years) in primary health centers in the Peruvian Amazon, 2022–2024.

https://doi.org/10.1371/journal.pntd.0013473.t004

Paracetamol was the most frequently recorded antipyretic overall, administered in 303 cases (75.8%), including 17 of 28 children under 5 years (60.7%) and 286 of 372 patients aged 5 years or older (76.9%). Among antibiotics, amoxicillin was the most frequently recorded in children under 5 years (20/28, 71.4%), whereas doxycycline predominated in patients aged 5 years or older (222/372, 59.7%). Ciprofloxacin, ceftriaxone, and ampicillin were recorded only among patients aged 5 years or older, whereas erythromycin was recorded in 3 children under 5 years (10.7%) and 14 patients aged 5 years or older (3.8%). Intravenous 0.9% sodium chloride was administered in 3 children under 5 years (10.7%) and 49 patients aged 5 years or older (13.2%) (Table 5).

Download:

• PNG
  larger image
• TIFF
  original image

Table 5. Medications recorded among confirmed leptospirosis cases by age group (<5 vs ≥ 5 years) evaluated at primary health care centers in the Peruvian Amazon, 2022–2024.

https://doi.org/10.1371/journal.pntd.0013473.t005

In exploratory bivariate comparisons by age group, descriptive statistical differences were observed for headache, general malaise and chills (Table 6). Due to the small number of children under 5 years and the exploratory nature of these comparisons, p-values should be interpreted descriptively rather than as confirmatory evidence of age-related differences. Undernutrition was more frequent among children under 5 years (50.0% vs 14.8%; p < 0.001). In addition, presentations more than 3 days after symptom onset were less frequent in children under 5 years than in patients aged 5 years or older (53.6% vs 73.7%; p < 0.001). The complete set of exploratory bivariate comparisons is presented in S2 Table.

Download:

• PNG
  larger image
• TIFF
  original image

Table 6. Exploratory bivariate comparisons of selected clinical characteristics between age group (<5 vs ≥ 5 years), primary health centers in the Peruvian Amazon, 2022-2024.

https://doi.org/10.1371/journal.pntd.0013473.t006

Discussion

This study provides descriptive data on the epidemiological and clinical profile of MAT-confirmed leptospirosis in primary health care centers of the Peruvian Amazon, with emphasis on differences between children under 5 years and patients aged 5 years or older. Despite the limited pediatric sample, this study provides rare primary-care data on MAT-confirmed pediatric leptospirosis in a hyperendemic Amazonian setting, where outpatient presentations are underrepresented in the literature. We observed a seasonal clustering of cases that coincided with the climatic cycle of Belén, including periods of river rise and intense rainfall, which likely increase environmental exposure and may contribute to sustained transmission. This context is especially relevant in Belén, where environmental vulnerability and limited sanitation may intensify exposure risks. A clearly defined seasonal pattern was not observed among children under 5 years; however, this likely reflects the small pediatric sample rather than a true absence of temporal variation.

The main age-related differences observed in this primary care cohort involved symptom profile, nutritional classification, and timing of consultation. These comparisons should be viewed as exploratory given the small pediatric sample size, multiple comparisons, and retrospective facility-based data. The small number of children under 5 years limited statistical power for subgroup comparisons and increased the risk of unstable estimates; therefore, age-group comparisons should be interpreted as exploratory and hypothesis-generating rather than confirmatory. In addition, some observed differences may reflect variation in symptom ascertainment, caregiver reporting, and healthcare-seeking behavior rather than true biological differences between age groups. Accordingly, these findings are best interpreted as descriptive and exploratory, particularly for children under 5 years, and require confirmation in larger prospective pediatric cohorts.

Children under 5 years most often presented with nonspecific symptoms, particularly malaise and fever, whereas headache and chills were more frequently recorded in older patients. This pattern is clinically important because it may reduce early diagnostic specificity in first-level care, especially in endemic Amazonian settings where leptospirosis overlaps with other common febrile illnesses such as dengue, malaria, and arboviral infections. In very young children, especially those who are pre-verbal, subjective symptoms are also more difficult to identify and document, which may further contribute to underrecognition. Taken together, these findings support the possibility that mild pediatric leptospirosis may be underdiagnosed in routine outpatient practice unless clinicians maintain a high index of suspicion [10,24,25].

A higher proportion of children under 5 years were evaluated within the first 3 days after symptom onset. This pattern may reflect earlier caregiver concern in young children, although the categorical structure of the available data and the retrospective design limit more detailed interpretation. Caregiver recall, documentation practices, and age-related differences in how symptoms are expressed may also have influenced these findings. Rather than indicating a definitive age-related difference in help-seeking behavior, this observation should be interpreted as a descriptive pattern that warrants further study [8,24,26].

Another notable finding was the higher frequency of undernutrition among children under 5 years. This result should be interpreted carefully because nutritional status was classified using age-specific anthropometric standards, and the retrospective design does not allow causal inference. Moreover, nutritional status in this setting may be shaped by broader social and environmental determinants, including poverty, food insecurity, inadequate sanitation, coinfections, and barriers to care, none of which were captured in the available records [24,27]. Therefore, the observed difference should not be interpreted as evidence that undernutrition independently influenced presentation or recovery, but rather as a contextual finding that may reflect underlying vulnerability in this population.

From a microbiological perspective, Varillal was the predominant MAT serogroup reactivity in both age groups. Panama reactivity appeared proportionally more frequent among children under 5 years, although this observation was based on very few cases and should not be overinterpreted. Since MAT reflects serologic reactivity rather than organism-level characterization, these findings do not permit conclusions about virulence, source of infection, or age-specific susceptibility. Larger studies with broader microbiological characterization are needed to determine whether these patterns are reproducible [28].

Treatment patterns also differed by age group, with amoxicillin more frequently recorded in children under 5 years and doxycycline predominating among older patients [27]. These patterns likely reflect routine primary-care practice and age-related prescribing considerations and are broadly consistent with the Peruvian National Technical Health Standard, which recommends amoxicillin or erythromycin as oral options for mild leptospirosis in children, while doxycycline, amoxicillin, ciprofloxacin, or erythromycin are listed as oral options for adults [21]. However, because this was a retrospective review of primary care records, detailed information on severity, timing of treatment initiation, dosing, duration, adherence, and clinical outcomes was not consistently available. Accordingly, these data should be interpreted as describing local prescribing patterns rather than treatment effectiveness or adherence to guidelines [8,29].

This study reflects a predominantly ambulatory and mild clinical spectrum of leptospirosis managed at primary health care centers. Patients with warning signs or suspected severe disease may have been referred to higher levels of care and are therefore likely underrepresented. In addition, the analysis depended on routinely collected variables and MAT results reported in the medical records, which limited assessment of other potentially important determinants such as socioeconomic conditions, sanitation, coinfections, and complementary diagnostic methods. These considerations are important when interpreting the scope and generalizability of the findings.

Overall, children under 5 years of age with confirmed leptospirosis in this primary care series commonly presented with nonspecific symptoms such as malaise and fever and were frequently identified in flood-prone settings. In the Amazonian context, where leptospirosis coexists with other endemic febrile illnesses, this pattern may complicate early recognition at the primary health care centers. These findings support the need to strengthen clinical suspicion, improve diagnostic capacity in primary health care centers, and reinforce epidemiological surveillance for pediatric leptospirosis in endemic areas. Further studies are needed to better characterize age-related differences in presentation and to clarify how social and environmental factors shape risk in young children.

Supporting information