Hantavirus global briefing — September 19, 2026: Europe defines its contact terms, and a steppe survey finds hantavirus in 47 rodent test units

Executive read

This briefing covers global hantavirus activity reported between 16 and 19 September 2026 (display timezone UTC+08:00), the interval since the September 16 edition. One official document and one survey carry most of this window, and the human case layer moved only in Germany.

Europe's agency supplied the definitions its own advice had left open. An addendum published on 18 September keeps the recommendations of the 11 September Andes virus contact advice and sets out what self-monitoring, active follow-up and self-isolation each mean in practice, together with the rule for handling contacts who have no symptoms while the 42-day monitoring period runs. 1

A rodent survey posted on 18 September reports hantavirus RNA in 47 of 137 testing units drawn from 343 small mammals trapped across four districts of the West Kazakhstan Region, with a model-based individual infection probability of 18.0% (95% profile-likelihood confidence interval 13.5–23.1%). The paper is a preprint and has not been peer reviewed. 2

Germany's statutory notification table for week 37, published 17 September, records 5 hantavirus notifications in the week and 215 for the year to date, against 217 over the same stretch of 2025. 3

Argentina's national bulletin has not moved past week 35, so its hantavirosis row — 58 events for 2026 against a 2022–2025 median of 41 — is carried forward here without renewal. 4

Europe supplies the definitions its advice was missing

The advice ECDC published on 11 September told EU/EEA countries to classify contacts of imported Andes virus cases by exposure level and to monitor them for 42 days. The words doing the work went undefined, and national authorities, which had diverged most sharply on what to do with asymptomatic contacts, filled them in for themselves. The addendum of 18 September closes that gap. 1

Three terms now carry fixed meanings. Self-monitoring means the contact watches for compatible symptoms for 42 days from the last exposure while continuing everyday activities. Active follow-up means a public health authority stays in regular contact, preferably by telephone, across the same period to ask whether symptoms have appeared. Self-isolation means physical separation from other people, and it applies to a contact who develops symptoms. 1

The two-level classification scheme carries over from the 11 September version. What the addendum settles is the asymptomatic case: a contact with no symptoms is monitored and followed up according to exposure level, and laboratory testing follows from symptoms appearing, whichever exposure class the contact was placed in. A contact who develops symptoms during the 42 days is tested as soon as possible, with whole blood in EDTA the preferred specimen, self-isolates until symptoms resolve, and then returns to the monitoring approach originally assigned. 1

The document also fixes the case definitions it works from — suspected, probable, confirmed and non-case — and the exposure window that decides who counts as a contact. Monitoring begins from the onset of the earliest symptom compatible with infection; where the prodromal illness is poorly defined, authorities may extend the window 48 hours earlier as a precaution. Day 0 for the 42-day count is the day of last exposure. 1

ECDC attaches its own caveat, and it matters for anyone adopting the thresholds locally. The exposure levels and the operational examples built on them, including duration thresholds and proximity criteria, are intended to support decisions and have not been formally validated for Andes virus infection. The evidence behind the advice comes mostly from retrospective outbreak investigations in Argentina and Chile, and the period during which a case is infectious remains incompletely characterised. National authorities may apply more or less stringent measures according to their own assessment of an event. 1

The addendum also restates the cruise-ship event that prompted the whole exercise, and its numbers are worth separating from the ones WHO published. ECDC records 12 confirmed cases and three deaths plus one probable case from the MV Hondius cluster, with 237 contacts from 15 countries monitored for up to 42 days and no secondary case detected among them. 1 WHO's closing notice counts contact identification across 33 countries and overseas territories, with 317 higher-risk contacts completing quarantine and monitoring and about 336 lower-risk contacts completing self-monitoring. 5 Both agencies counted the same event; the two contact tallies cover different populations, and a reader who adds them together will overstate the network.

Official posture: what moved and what held

ECDC's weekly communicable disease threats report for 12–18 September, published on 18 September, covers Ebola, malaria, botulism, avian influenza A(H9N2), influenza A(H5), respiratory virus epidemiology in the EU/EEA, West Nile virus, chikungunya, dengue, Crimean-Congo haemorrhagic fever, Vibrio, measles and expert deployments. Hantavirus and Andes virus are absent from that list. 6 The report tracks the events ECDC counts as an active threat at European level, and hantavirus has stayed off that list through the whole late-summer stretch.

The European Commission's outbreak page carries no entry newer than its 4 September flash report of the Health Security Committee working group, and its assessment holds: risk to the general population in the EU/EEA from imported Andes virus is very low, and the virus has been identified as Andes hantavirus, the only hantavirus known to pass between people, typically through close and prolonged contact. 7

The WHO outbreak news index carries no hantavirus item later than the 2 July notice that closed the cruise-ship event at 13 cases, including three deaths, and a case fatality ratio of 23%. 5 The United States CDC situation page remains archived at 7 July 2026 with national case reporting through 2023. 8

The frame behind each human count

Germany's Robert Koch Institute publishes hantavirus notifications under the statutory reporting system, and its week-37 table, published on 17 September 2026, reads as follows. 3

Measure	Week 37, 2026	Weeks 1–37, 2026	Weeks 1–37, 2025
Hantavirus notifications	5	215	217

The cumulative column does not advance by the weekly column, and anyone tracking Germany week to week should read it accordingly. In the bulletin published on 10 September, the same row read 205 notifications for weeks 1–36. 9 One week later the year-to-date figure stands at 215, so ten notifications joined the 2026 total while five were recorded in the reporting week. The remaining five arrived as late reports against earlier weeks. The same correction applies to the 2025 column, which makes the two-notification gap between 215 and 217 provisional rather than a settled comparison.

Argentina's position is unchanged because the source has not been renewed. The Boletín Epidemiológico Nacional index page, last updated on 14 September, still lists the week-35 issue as its newest, and that issue's hantavirosis quick table reads 58 accumulated events for 2026 against a 2022–2025 median of 41 at the same point, above the expected range for the year to date and within range across the most recent four weeks. 4 10 The row still carries no provincial breakdown, so the national figure supports no statement about where in the country the events occurred.

Chile's national position stands where earlier editions left it, with nothing dated inside this window.

No fresh outbreak event with geographic detail was published during 16–19 September, which leaves Germany's weekly figure as the only new human count in this window sourced to a national statutory system.

Reservoir: a steppe survey with the settlements attached

The West Kazakhstan Region has recorded HFRS since 2000, when official registration began; the paper counts 329 cases across the region from 2000 to 2025, with hospitalisations peaking between November and January in most years and patients drawn mainly from rural areas. 2

The new survey was run by Talgat Nurmakhanov and colleagues in four districts — Burlinsky, Chingirlauisky, Terektiinsky and Baytereksky — during August and September 2025, at 17 sampling sites. Field teams set Hero traps at two-metre intervals with bread bait and captured 343 small mammals, identified to species by morphology. Lung tissue was pooled by host species and capture site into 123 testing units of one to six animals, and 14 specimens from the same campaign were tested individually, giving 137 testing units over the 343 animals. Samples were screened by real-time RT-qPCR against a 230-base-pair fragment of the L segment, with conventional RT-PCR alongside. 2

Hantavirus RNA turned up in 47 of the 137 testing units, 45 from pooled units and two from individually tested animals. The authors report that as a descriptive minimum infection rate of 13.7%, and then model the individual level separately: under a pooled-testing likelihood model that accounts for differences in unit size, the estimated probability that any given captured animal carried the virus is 18.0%, with a 95% profile-likelihood confidence interval of 13.5–23.1%. Three host species carried positives. 2

District	Rodents collected, 2025	Descriptive minimum infection rate
Chingirlauisky	63	19.0%
Baytereksky	151	12.7%
Terektiinsky	86	12.6%
Burlinsky	43	11.6%

Capture totals are summed from the paper's site-level table; the paper reports the district-level rate only, and warns that the rates are descriptive minimums whose unit sizes differed between districts. 2

The part of the paper a surveillance reader will want is the comparison with human disease. The authors set their sampling sites against the human HFRS cases reported in the region in 2025, 42 in total, and the overlap is partial.

Settlement	Human HFRS cases, 2025	Rodents collected, 2025
Daryinsk	8	48
Rubezhinskoye	2	15
Utvinka	2	1
Uralsk	16	No sampling reported

The remaining 30 human cases fell in settlements where the 2025 campaign collected no animals. 2

Three limits bound what the survey establishes. No sequencing was performed, so the detected RNA could not be assigned to a hantavirus species or genetic lineage, which leaves open which virus the region's rodents carry. The survey covers a single campaign in one year, so it gives a snapshot rather than a trend. And spatial overlap between sampling sites and settlements that reported human cases is a coincidence of geography; the authors state plainly that it establishes no transmission link. 2

A second reservoir measurement belongs to the same layer. Petra Strakova and colleagues tested common voles in South Moravia in the Czech Republic and found TULV, a hantavirus, in 35 of 140 animals, or 25%, alongside Leptospira in 70 (50%), Bartonella in 72 (51.4%) and Brucella in 50 (35.7%), with frequent co-infections in the same animals. The paper was accepted on 7 September and appears in the September 2026 issue of APMIS, so it sits just outside this window and is flagged here with its own date. 11

Research triage

Ultrasound for severity in viral haemorrhagic fevers — PLOS Neglected Tropical Diseases, September 17

Authors and group: Daniel Z. Hodson is first author, with authors from the University of California Los Angeles, the University of Marília in Brazil, Hospital Kuala Lumpur, the Lighthouse Clinic Trust in Malawi, the University of Cologne, the University Medical Centre Hamburg-Eppendorf, the Bernhard Nocht Institute for Tropical Medicine and the University of Tübingen.
Design: Systematic review, registered with PROSPERO as CRD42024620862, searching PubMed through May 2025 for original patient-level ultrasound findings in viral infections associated with haemorrhagic fevers with serological or molecular confirmation. Case series and duplicative datasets were excluded, and risk of bias was assessed with QUADAS-2. Of 457 publications screened, 56 were included, covering 6,979 participants.
Dates: Published September 17, 2026. DOI 10.1371/journal.pntd.0014609.
What it reports: The evidence base is dominated by dengue: 44 publications and 6,189 participants. Hantavirus disease contributes three publications and 134 participants, with Crimean-Congo haemorrhagic fever at six and 494, yellow fever at two and 116, and Lassa fever at one and 46. Findings recorded across these infections include gallbladder wall thickening, ascites, pleural effusions and abnormal lung ultrasound, pericardial effusions, abdominal lymphadenopathy, splenomegaly, abnormal pancreatic ultrasound, and hyperechoic kidneys or liver. Pericardial effusions and several other findings were associated with severe disease, and the authors propose that ultrasound findings be folded into severity definitions, floating an ultrasound-based assessment of plasma leakage as a hypothesis-generating framework. Crude proportions and random-effects meta-analysis were used where the data allowed; the review does not report a pooled effect size specific to hantavirus disease.
Limits: A single database was searched, which the authors say may have missed work in local journals. Recruitment was overwhelmingly in-hospital at tertiary referral centres during high-transmission periods, heterogeneity was high, and the dengue studies carried bias in the timing domain.
Original: A systematic review of ultrasound to predict severe disease in patients with dengue and other viral infections associated with hemorrhagic fevers, PLOS Neglected Tropical Diseases

One clinical item published just before this window opened is worth a line for anyone seeing undifferentiated fever and joint pain. A letter in The Journal of Rheumatology, online on 15 September, sets out hantavirus infection as a mimic of rheumatic disease and proposes a selective diagnostic approach for telling the two apart. 12

Watchpoints

Argentina's week-36 bulletin. The index page's newest issue is week 35. A new issue would show whether the 58-against-41 row holds or moves. 4
National uptake of the three definitions. The addendum binds no member state, and the point of past divergence was quarantine of asymptomatic contacts. National guidance documents that adopt the definitions, or keep their own, are the observable. 1
The Kazakh autumn. The region's HFRS hospitalisations peak from November to January, and this survey ran in August and September. A second campaign, or sequencing of the stored samples, would turn an 18.0% infection probability into a typed virus. 2
Germany's revised totals. The cumulative column gained ten notifications in one week, so the 215-against-217 comparison with 2025 will shift again as late reports land. The next bulletin covers week 38. 3