http://www.diva-portal.org

This is the published version of a paper published in Infectious Diseases.

Citation for the original published paper (version of record):

Furberg, M., Anticona Huaynate, C., Schumann, B. (2019)Post-infectious fatigue following Puumala virus infectionInfectious Diseases, 51(7): 519-526https://doi.org/10.1080/23744235.2019.1605191

Access to the published version may require subscription.

N.B. When citing this work, cite the original published paper.

Permanent link to this version:http://urn.kb.se/resolve?urn=urn:nbn:se:umu:diva-159095

Full Terms & Conditions of access and use can be found athttps://www.tandfonline.com/action/journalInformation?journalCode=infd20

Infectious Diseases

ISSN: 2374-4235 (Print) 2374-4243 (Online) Journal homepage: https://www.tandfonline.com/loi/infd20

Post-infectious fatigue following Puumala virusinfection

Maria Furberg, Cynthia Anticona & Barbara Schumann

To cite this article: Maria Furberg, Cynthia Anticona & Barbara Schumann (2019) Post-infectious fatigue following Puumala virus infection, Infectious Diseases, 51:7, 519-526, DOI:10.1080/23744235.2019.1605191

To link to this article: https://doi.org/10.1080/23744235.2019.1605191

© 2019 The Author(s). Published by InformaUK Limited, trading as Taylor & FrancisGroup

Published online: 11 May 2019.

Submit your article to this journal

Article views: 305

View related articles

View Crossmark data

INFECTIOUS DISEASES,2019; VOL. 51,NO. 7, 519–526

https://doi.org/10.1080/23744235.2019.1605191

ORIGINAL ARTICLE

Post-infectious fatigue following Puumalavirus infection

Maria Furberga , Cynthia Anticonab and Barbara Schumannb

aDepartment of Clinical Microbiology, Umea University, Umea, Sweden; bDepartment of Epidemiology and Global Health, UmeåUniversity, Umeå, Sweden

ABSTRACTBackground: Puumala virus infection or nephropathia epidemica (NE) is common in northern Sweden. NE causes haemor-rhagic fever with renal syndrome. Most patients make a full recovery, but a convalescent phase with fatigue has beenreported. Although post-infectious fatigue has been demonstrated for other viral infections, it is not well studied in relationto NE. This study assessed recovery time and levels of fatigue in former NE patients, as compared to the gen-eral population.

Methods: NE patients diagnosed in northern Sweden between 2007 and 2011, together with a comparison sample fromthe general population, answered a questionnaire on demographic and health-related factors, including the FatigueSeverity Scale (FSS), and characteristics of NE infection. Self-reported recovery time was assessed, and fatigue levels werecompared across the two groups by multiple linear regression, stratified by gender.

Results: In total, 1132 NE patients and 915 comparison group subjects participated. Time to complete recovery wasreported to exceed 3 months for 47% and 6 months for 32% of the NE patients. Recovery time differed by gender andage. NE patients had significantly higher FSS scores than the comparison group. Differences were greater among womenthan men, and adjustments for current illness, body mass index, smoking and current residence only slightly modifiedthe estimates.

Conclusions: Individuals with previous NE infection show higher fatigue scores than non-infected individuals, even 5 yearsfollowing the infection. Full recovery takes half a year or longer for a substantial proportion of former NE patients.

KEYWORDSPuumala virusnephropathia epidemicapost-infectious fatiguerecovery timeNorthern Sweden

ARTICLE HISTORYReceived 30 November 2018Revised 26 February 2019Accepted 4 April 2019

CONTACTMaria Furberg

maria.furberg@umu.seDepartment of Clinical Microbiology, Umea

University, Umea, Sweden

� 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis GroupThis is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://creativecommons.org/licenses/by-nc-nd/4.0/),which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is not altered, transformed, or built upon inany way.

Background

Northern Sweden has a high incidence of Puumala virus(PUUV) infection, also known as nephropathia epidemica(NE) or vole fever. The causative agent, PUUV, is a hanta-virus that causes haemorrhagic fever with renal syn-drome. The virus is spread by bank vole (Myodesglareolus) excreta, although the voles themselves remainunaffected by the infection. Transmission to humans isbelieved to occur via the inhalation of infectious aero-sols; no human-to-human transmission of PUUV hasbeen reported [1].

Following an incubation period of 1–8 weeks (median4) PUUV causes a sudden onset and mild form of haem-orrhagic fever with renal syndrome, varying from sub-clinical cases to severe disease requiring haemodialysis[2]. The symptoms include high fever, malaise, headache,abdominal and back pain, dizziness, vomiting, blurredvision, thrombocytopenia, proteinuria, haematuria andtransient renal failure. Case fatality is low (<0.5%) [3,4].No specific treatments are available, and the infectioncreates lifelong immunity [5].

The Swedish Communicable Diseases Act mandatesthat all NE cases must be reported [6]. The diagnosticcriteria constitute either a positive PUUV–RNA PCR bloodtest or a positive test of specific PUUV IgM and IgG anti-bodies, consistent with current infection. The annualincidence of NE in Sweden varies considerably, withpeaks every 3–4 years related to the bank vole popula-tion cycle. Recorded incidences are on average 20/100,000 population per year, although the true inci-dence is considered to be at least eight times higher[7,8]. During a large NE outbreak in 2007, the incidence

reached 313 cases/100,000 population in the county ofV€asterbotten [3] (Figure 1).

NE infection is considered a relatively benign dis-ease, and its long-term consequences have been onlybriefly studied, with particular emphasis placed onhypertension, persisting haematuria and proteinuriaand prolonged renal impairment. Initial studies indi-cated long-lasting health effect [9–11], although theseare not corroborated by more recent studies, with theexception of the case of persisting haematuria [12].Most patients make a full recovery following NE infec-tion, although a convalescent phase of several weekswith fatigue has been reported [2,13]. Long recoverytimes associated with post-infectious fatigue have alsobeen reported in other viral infections such asEpstein–Barr Virus (EBV, also called mononucleosis)[14], Dengue fever [15] and Ross River virus [16].Fatigue following EBV infection has been most widelyinvestigated, in both retrospective and prospectivestudies [16,17]. Post-infectious fatigue state does notexhibit specific diagnostic criteria or a commonlyaccepted definition, and is neither equal to chronicfatigue syndrome as defined by the Centers forDisease Control and Prevention [18], nor the less well-defined chronic fatigue (CF).

Thus, previous research has indicated a high preva-lence of post-infectious fatigue for some viral infectiousdiseases, but only two small studies mention fatigue inrelation to NE [2,13]. The aim of this study was to assessrecovery times following NE infection and to examinelevels of fatigue in former NE patients compared to thegeneral population.

Figure 1. Cumulative incidence of NE in Sweden (dotted), Norrbotten (dashed), and V€asterbotten (solid) counties. Source: Public HealthAgency of Sweden, 2018.

520 M. FURBERG ET AL.

Methods

Recruitment

We contacted all adult patients diagnosed with NE infec-tion between 2007 and 2011 at the two laboratoriesserving northernmost Sweden: the Sunderby ClinicalMicrobiology Laboratory in Luleå and the ClinicalMicrobiology Laboratory at Norrlands University Hospitalin Umeå. Quick-test ReaScan Puumala IgM [19] andimmunofluorescence (IFA) IgM and IgG analyses [20]were used at both laboratories to diagnose NE. All IgMtests had to be confirmed by IgG seroconversion or afour-fold titer increase in paired samples, to meet thereportable criteria.

A comparison sample from the general populationwas obtained from Statistics Sweden, matched by age,gender and municipality of current residence of thepatients. Via their Swedish personal identification num-bers, any previous diagnosis of NE in the two labora-tories during the last 10 years could be ruled out forthe individuals in the comparison group.

The sample of NE patients and individuals for com-parison received a letter of invitation, a questionnaireand a written informed consent form by postal service.All potential participants who did not respond receivedone reminder letter.

Questionnaire

The questionnaire consisted of three sections. Section 1included demographic information and health-relatedfactors. Section 2 (for patients only) included the charac-teristics of the NE infection, such as place of diagnosis,hospitalisation, time in bed, sick leave and time untilcomplete recovery. In addition, patients were asked toidentify which of the following symptoms dominatedprior to their recovery: fatigue, pain or sleep disturb-ance, or any alternative symptom, which they wereasked to name. Section 3 included a series of validatedinstruments to measure fatigue (FSS—Fatigue SeverityScale), general health (SF 36 – Short Form 36 HealthSurvey), sleep quality (PSQI – Pittsburgh Sleep QualityIndex) and pain (BPI-SF – Brief Pain Inventory-Short Form).

In this paper, we report the findings of the fatigueassessment. The FSS is a nine-item, self-report question-naire that assesses the effects of fatigue on daily living.The final fatigue score is measured on a scale from 1(low fatigue) to 7 (high fatigue). The FSS has beensuccessfully used in clinical research to distinguish the

frequency and severity of fatigue between healthysubjects and patients with various infectious and non-communicable diseases [21].

Ethical approval

The study was approved by the Regional Ethical ReviewBoard in Umeå, Sweden. Written informed consent wasobtained from the participants.

Statistical analyses

Descriptive analyses of the study population were calcu-lated as frequencies and percentages for categorical var-iables, and means and standard deviations forcontinuous variables, separately for the patients and thecomparison group. Patients’ disease-related characteris-tics were described, including hospitalisation, co-morbid-ity during the acute disease and recovery time.

The relationship between NE exposure and thefatigue score was explored via simple and multiple lin-ear regression analyses. Multivariate models wereadjusted for potential confounders such as current ill-ness, body mass index (BMI), smoking and residence. Allstatistical analyses were stratified by gender, consideringprevious evidence of a higher frequency of post-infec-tious fatigue in women [15]. All analyses were per-formed at the 5% significance level using SPSS (v.24).

Results

Study population

In total, 1132 NE patients and 915 subjects in the compari-son group returned the questionnaire, giving an overallresponse rate of 59.9% (65.8% among patients and 53.2%among the comparison group), making a total sample of2047. The study population is described in Table 1. Themean age was 60 years; no significant differences betweenthe groups appeared regarding gender and age distribu-tion, socioeconomic status or exercise habits. More NEpatients had a current disease at the time of participation,and reported poor health to a greater extent than partici-pants from the comparison group. There were significantlymore current smokers among patients.

Disease-related characteristics of NE patients

The vast majority had received their NE infection diag-nosis during the large outbreak in 2006–2007, with nodifference by gender. More than 90% of the patientswere bedridden during the acute NE episode, but fewer

INFECTIOUS DISEASES 521

than half had to be hospitalised. More women than menwere in bed or hospital, although the difference was sig-nificant only for bed confinement. Twenty-three percent of the patients reported that they had another dis-ease when contracting NE; significantly more womenthan men had some concurrent disease (Table 2).

Recovery time

For 32% of the patients, 6 months or longer wererequired for a full recovery, nearly half, 47%, experienceda recovery time of at least 3 months. In this group, thevast majority named fatigue as the main symptom pre-venting recovery; one-fifth of the 47% (98 individuals)spontaneously used the same, very specific Swedishword for ‘feebleness’ (‘orkesl€oshet’) to describe the pre-dominant symptom that prevented a complete recovery(data not shown).

Recovery time differed by gender and age. More menthan women recovered within the first month, butno gender difference could be identified thereafter(Table 2). Among women (but not men), young and oldpatients exhibited contrasting recovery times, withyounger women (<60 years) tending to report longerrecovery times than elderly women (60 years or older),who more often recovered within the first 3 months(Figure 2).

Association between fatigue and NE infection

Among both men and women, NE patients exhibitedhigher FSS scores than the comparison group. One-fifth(21.6%) of the former NE patients reported high fatiguelevels (score 5 or higher) whereas only 10.6% of thecomparison group reported such a high score. The dif-ference was more pronounced among women than men(Figure 3).

A statistically significant difference in the FSS scorewas identified between NE patients and the comparisongroup among both men and women, with larger effectsfor women. Adjustment for current diseases, BMI, smok-ing habits and current residence reduced the effect onlyslightly (Table 3). Additional models adjusting for otherpotential confounders (age and physical activity) changedthe effect estimates only marginally (data not shown).

Discussion

This study on the long-term outcomes of NE has shownthat almost half of the NE patients reported a recoverytime of at least 3 months. This is in line with the pro-spective study by Settergren et al., which reported a fullrecovery after 3 months for 46 out of 74 NE patients(62%). After 6 months, all 66 evaluated patients hadrecovered fully (6 were lost to follow up) [2]. In 1971,L€ahdevirta reported fatigue lasting 1–5 months for 17%(6 out of 36) of NE patients evaluated [13]. To contrastthat, in the present study, 6 months following the acute

Table 2. Disease-related characteristics of NE patients.Total Men Women p-value

Time since diagnosis, N (%) .1613 to <6 months 27 (2.4) 17 (2.8) 10 (2.0)6 to <12 months 52 (4.6) 25 (4.1) 27 (5.3)1 to <2 years 120 (10.7) 64 (10.4) 56 (11.1)2–3 years 202 (18.0) 106 (17.2) 96 (19.0)�4 years 721 (64.3) 404 (65.6) 317 (62.6)Total 1122 (100) 616 (100) 506 (100)

Confinement in bed, N (%) .015Yes 1000 (91.2) 537 (89.4) 463 (93.5)No 96 (8.8) 64 (10.6) 32 (6.5)Total 1096 (100) 601 (100) 495 (100)

Hospitalisation, N (%) .116Yes 454 (41.2) 235 (39.0) 219 (43.7)No 648 (58.8) 367 (61) 281(56.3)Total 1102 (100) 602 (100) 500 (100)

Concurrent disease during NE, N (%) .012Yes 247 (23.4) 118 (20.5) 129 (27.0)No 808 (76.6) 458 (79.5) 350 (73)Total 1055 (100) 576 (100) 479 (100)

Recovery time, N (%) <.0011–4 weeks 282 (26.2) 185 (31.5) 97 (19.9)1 to <3 months 290 (27.0) 149 (25.3) 141 (28.9)3 to <6 months 159 (14.8) 71 (12.1) 88 (18.0)6 to <12 months 166 (15.4) 86 (14.6) 80 (16.4)�1 year 179 (16.6) 97 (16.5) 82 (16.8)Total 1076 (100) 588 (100) 488 (100)

NE: Nephropathia epidemica. Differences between sums of subgroups and totalnumbers are due to missing values. P-values are for gender differences(Chi2 test).

Table 1. Study population.NE patients Comparison group p-value

Total, N (%) 1132 (55.3) 915 (44.7) .827Gender, N (%) .827Men 623 (55.0) 508 (55.5)Women 509 (45.0) 407 (44.5)

Age, mean (SD) 58.5 (14.1) 58.3 (13.9) .682Residence, N (%) <.001Rural (�10,000) 863 (77.8) 595 (67.3)Urban (>10,000) 246 (22.2) 289 (32.7)

Education, N (%) .171Primary/elementary 338 (31.2) 261 (29.5)Senior high school 346 (32.0) 274 (30.9)College/university 272 (25.1) 261 (29.5)Other higher education 126 (11.6) 90 (10.2)

Employment status, N (%) .191Working 528 (47.2) 464 (51.1)Not working 509 (45.5) 387 (42.6)Other 82 (7.3) 57 (6.3)

Current illness, N (%) 432 (40.6) 311 (35.1) .012Smoking habits, N (%) <.001Current smoker 189 (16.9) 95 (10.5)Ex-smoker 408 (36.5) 329 (36.2)Never smoker 521 (46.6) 484 (53.3)

BMI, mean (SD) 26.4 (4.1) 26.2 (4.0) .146Poor health, N (%) 336 (29.8) 184 (20.2) <.001

NE: nephropathia epidemica; SD: standard deviation; BMI: body mass index.Difference between sums of subgroups and total numbers are due to missingvalues. P-values are for difference between NE patients and the comparisongroup (Chi-squared test for categorical variables, t-test for continuous variables).

522 M. FURBERG ET AL.

infection one-third of the NE patients still did not feelcompletely recovered.

Fatigue appears to be a common problem preventingrecovery. In the Dubbo prospective cohort study theprevalence of post-infectious fatigue syndrome was 27%at 3 months and 12% at 6 months following the acuteepisode [16]. Another prospective cohort study found aprevalence of physical fatigue in 40% of patients6 months after acute EBV infection [17]. Previous workon EBV-associated post-infectious fatigue has demon-strated a relationship with the severity of the initialinfection [16]. Such an association was not found in ourdata, using hospitalisation as a proxy for severity (datanot shown), nor in a study of Dengue fever [15].

The usual treatment of NE patients at the Clinic ofInfectious Diseases, Umeå University Hospital inV€asterbotten, involves signing off NE patients for up to1 month and longer sick leave is unusual (personal com-munication). Our study results indicate a far longer-last-ing effect on the wellbeing of a large proportion of NEpatients. During major outbreak years such as2006–2007, when the incidence rate reached 313reported cases/100,000 inhabitants in V€asterbottenCounty, and the true incidence is believed to be eighttimes as high, the potential public health and healtheconomic effects can be substantial [3,8].

The present study was conducted in order to fill aknowledge gap regarding the association between

Figure 2. Self-reported recovery time in men and women by age group.

Figure 3. Fatigue scores in the study population by gender.

Table 3. Fatigue score differences by NE status and co-variates (regression coefficient with 95% confidence intervals).Simple model Adjusted model 1 Adjusted model 2

Men Women Men Women Men Women

NE exposure 0.37 (0.20; 0.54) 0.49 (0.28; 0.69) 0.34 (0.18; 0.50) 0.41 (0.22; 0.61) 0.37 (0.20; 0.53) 0.44 (0.24; 0.64)Current illness – – 0.81 (0.64; 0.98) 0.89 (0.69; 1.09) 0.80 (0.62; 0.97) 0.83 (0.62; 1.04)BMI – – – – 0.02 (–0.01; 0.04) 0.03 (0.007; 0.05)Smoking – – – – 0.04 (–0.17; 0.20) 0.13 (–0.07; 0.33)Urban residence 0.02 (–0.17; 0.20) 0.06 (–0.17; 0.29)

Model 1, adjusted for current illness. Model 2, adjusted for current illness, BMI, smoking (current or former smokers; reference: never smokers) andcurrent urban residence (reference: rural). NE: nephropathia epidemica; BMI: body mass index.

INFECTIOUS DISEASES 523

fatigue and NE infection. The results show that NEpatients displayed significantly higher FSS scores thanthe comparison group in both men and women. Theproportion with high scores (5 or above) was twice ashigh in the NE patients as in the comparison group(22% versus 11%). Even when adjusting for potentialconfounders, this association was not altered. Moreover,the effect was stronger among women than men, withexposure to NE infection increasing the FSS score (rang-ing from 1 to 7) by 0.37 units in men and by 0.44 unitsin women, adjusted for current illness, BMI, smoking andrecidence. A prospective study following acute NE infec-tion found chronic overt hormonal deficiencies in 17%(9 out of 54) of the patients. Hormonal alterations couldprovide some explanation to the higher levels of fatigue;however, the hormonal changes were more commonamong men (22.9%) than women (12%) [22].

In the present study, women generally displayedhigher FSS scores than men in both groups. A perform-ance test of FSS in a general Swedish setting has yet tobe conducted, but other fatigue assessment tools havealso indicated significantly higher fatigue scores amongwomen than men in Sweden [23]. A Norwegian valid-ation of FSS suggested a cut-off of 5 or higher (on thefatigue scale ranging from 1 to 7) for high fatigue andthe same cut-off was used in the present study. In theirsample of 1893 respondents from the general popula-tion, high fatigue scores were found in 23.1%. In ourcomparison group, only 10.6% reported a fatigue scoreof 5 or higher, whereas 21.6% of the former NE patientsreported high fatigue. The Norwegian study reportedthat more women (26.2%) than men (19.8%) in the gen-eral population experienced high fatigue [24]. In ourstudy, gender differences were more pronounced in thepatient than the comparison group, suggesting an inter-action between gender and NE exposure. In a study onDengue fever, fatigue was significantly more commonamong women [15].

Differences between men and women did not consti-tute the focus of our study, but became evident in thedata. Gender differences in prolonged fatigue (as dem-onstrated for a number of viral infections) pose ques-tions regarding potential mechanisms. For hantavirusesas a group, research has shown that even though suchinfections are more prevalent among males, mortality ishigher in females [25]. In our study, younger women(under 60 years) were more prone to long recoverytimes than were older women and men. One couldhypothesise that sex hormones represent a contributorto the recovery time differences seen in our data. It is

likely that the pathogenesis of post-infectious fatigue ismultifactorial and that the onset may be a consequenceof immune alterations triggered by the viruses.However, further research is required for clarification.

Some findings from the descriptive analysis may par-tially explain the higher fatigue scores among former NEpatients. The patient group comprised significantly morecurrent smokers than the comparison group. This findingis to be expected, because previous studies have identi-fied smoking as a risk factor for NE infection [26,27].However, adjusting for smoking in the analyses reducedthe effect of NE exposure on fatigue to a negligibleextent, suggesting the involvement of other pathogenicmechanisms. The patient group reported poor healthand current diseases to a greater extent than the com-parison group. In both groups, as expected, fatiguescores were higher among those reporting current ill-ness, but the difference was larger in the patient groupthan in the comparison group; hence, NE infectionappears to have an additional effect on fatigue.

This study had several limitations. The most importantone is the retrospective design, prone to both recall biasand selection bias. The participation rate was low (60%in total), and was lower in the comparison group than inthe patient group (53% and 66%, respectively). One rea-son may have been individuals’ reluctance to participatewhere they felt high levels of fatigue, resulting in selec-tion bias. Furthermore, a large number of patients con-tracted their disease during the outbreak of 2006–2007,several years before the study was conducted, and sorecall bias (in particular regarding disease characteristicssuch as recovery time) may have affected our data.Individuals in the comparison group were checked forreported previous NE infection diagnosed in the two lab-oratories included in the study, but did not undergolaboratory tests to rule out any undiagnosed NE. Thus,undiagnosed infections might have been included in thecomparison group, posing the potential for bias due tomisclassification. Their number is however likely to besmall. Finally, one major limitation is the lack of informa-tion regarding patient participants’ health status or levelsof fatigue prior to the NE infection.

The major strength of this study is its large studygroup from a well-defined geographical study area oftwo counties in Sweden. The involvement of a matchedcomparison group enabled us to compare formerpatients and their fatigue scores with individuals from asimilar context.

The findings highlight the importance of furtherinvestigation into fatigue-affected former NE patients. A

524 M. FURBERG ET AL.

prospective study to reduce recall bias and examine theextent and properties of post-infectious fatigue follow-ing acute NE infection would thus be valuable.

Conclusions

A prolonged state of post-infectious fatigue is very com-mon following acute NE infection. People with previousNE infection show higher fatigue scores than non-infected individuals, even 5 years after the infection. Theunderlying mechanisms are unclear and merit furtherinvestigation into the nature and extent of fatigue inthis population.

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

V€asterbotten County Council This study was financially supportedby V€asterbotten County Council.

ORCID

Maria Furberg http://orcid.org/0000-0002-7580-6485Cynthia Anticona http://orcid.org/0000-0001-9321-6174Barbara Schumann http://orcid.org/0000-0002-9722-0370

References

[1] Leduc JW. Epidemiology of hemorrhagic fever viruses. Rev

Infect Dis. 1989;11:S730.[2] Settergren B, Juto P, Trollfors B, et al. Clinical characteristics

of nephropathia epidemica in Sweden: prospective study

of 74 cases. Rev Infect Dis. 1989;11:921–927.[3] Pettersson L, Boman J, Juto P, et al. Outbreak of Puumala

virus infection, Sweden. Emerg Infect Dis. 2008;14:808–810.[4] Hjertqvist M, Klein SL, Ahlm C, et al. Mortality rate patterns

for hemorrhagic fever with renal syndrome caused by

Puumala virus. Emerg Infect Dis. 2010;16:1584–1586.[5] Settergren B, Ahlm C, Juto P, et al. Specific Puumala IgG

virus half a century after haemorrhagic fever with renal

syndrome. Lancet. 1991;338:66.[6] Folkh€alsomyndigheten Public Health Agency of Sweden

2018. Notifiable diseases 2018 [20181207]. Available from:

https://www.folkhalsomyndigheten.se/the-public-health-

agency-of-sweden/communicable-disease-control/surveil-

lance-of-communicable-diseases/notifiable-diseases/[7] Olsson GE, Dalerum F, Hornfeldt B, et al. Human hantavirus

infections, Sweden. Emerg Infect Dis. 2003;9:1395–1401.[8] Ahlm C, Linderholm M, Juto P, et al. Prevalence of serum

IgG antibodies to Puumala virus (haemorrhagic fever with

renal syndrome) in northern Sweden. Epidemiol Infect.1994;113:129–136.

[9] Miettinen MH, Makela SM, Ala-Houhala IO, et al. Ten-yearprognosis of Puumala hantavirus-induced acute interstitialnephritis. Kidney Int. 2006;69:2043–2048.

[10] Miettinen MH, Makela SM, Ala-Houhala IO, et al. Tubularproteinuria and glomerular filtration 6 years after puumalahantavirus-induced acute interstitial nephritis. Nephron ClinPract. 2009;112:c115–c120.

[11] Makary P, Kanerva M, Ollgren J, et al. Disease burden ofPuumala virus infections, 1995–2008. Epidemiol Infect.2010;138:1484–1492.

[12] Latus J, Schwab M, Tacconelli E, et al. Clinical course andlong-term outcome of hantavirus-associated nephropathiaepidemica, Germany. Emerg Infect Dis. 2015;21:76–83.

[13] L€ahdevirta J. Nephropathia epidemica in Finland. Aclinical histological and epidemiological study. Ann ClinRes. 1971;3:1.

[14] Petersen I, Thomas JM, Hamilton WT, et al. Risk and predic-tors of fatigue after infectious mononucleosis in a largeprimary-care cohort. QJM. 2006;99:49–55.

[15] Seet RC, Quek AM, Lim EC. Post-infectious fatigue syn-drome in dengue infection. J Clin Virol. 2007;38:1–6.

[16] Hickie I, Davenport T, Wakefield D, et al. Post-infective andchronic fatigue syndromes precipitated by viral and non-viral pathogens: prospective cohort study. BMJ. 2006;333:575–578.

[17] White PD, Thomas JM, Amess J, et al. Incidence, risk andprognosis of acute and chronic fatigue syndromes and psy-chiatric disorders after glandular fever [Research Support,Non-U.S.] Br J Psychiatry. 1998;173:475–481.

[18] Fukuda K, Straus SE, Hickie I, et al. The chronic fatigue syn-drome: a comprehensive approach to its definition andstudy. International Chronic Fatigue Syndrome StudyGroup. Ann Intern Med. 1994;121:953–959.

[19] Parviainen M, Kelo E, Sirola H, et al. Detection of acutehantavirus infections using novel instrument-readable rapidtests. Eur Infect Dis. 2011;5:35–37.

[20] Elgh F, Wadell G, Juto P. Comparison of the kinetics ofPuumala virus specific IgM and IgG antibody responsesin nephropathia epidemica as measured by a recombin-ant antigen-based enzyme-linked immunosorbent assayand an immunofluorescence test. J Med Virol. 1995;45:146–150.

[21] Valko PO, Bassetti CL, Bloch KE, et al. Validation of thefatigue severity scale in a Swiss cohort. Sleep. 2008;31:1601.

[22] M€akel€a S, Jaatinen P, Miettinen M, et al. Hormonal deficien-cies during and after Puumala hantavirus infection. Eur JClin Microbiol Infect Dis. 2010;29:705–713.

[23] Engberg I, Segerstedt J, Waller G, et al. Fatigue in the gen-eral population- associations to age, sex, socioeconomicstatus, physical activity, sitting time and self-rated health:the northern Sweden MONICA study. BMC Public Health.2017;2014:17.

[24] Lerdal A, Wahl AK, Rustoen T, et al. Fatigue in thegeneral population: a translation and test of thepsychometric properties of the Norwegian version of

INFECTIOUS DISEASES 525

the fatigue severity scale. Scand J Public Health. 2005;33:123–130.

[25] Klingstr€om J, Ahlm C. Sex, gender, and hemorrhagic feverviruses. In Klein SL, Roberts CW, editors. Sex and genderdifferences in infection and treatments for infectious dis-eases. New York: Springer International Publishing; 2015. p.211–230.

[26] Vapalahti K, Virtala AM, Vaheri A, et al. Case-control studyon Puumala virus infection: smoking is a risk factor.Epidemiol Infect. 2010;138:576–584.

[27] Gherasim A, Hjertqvist M, Lundkvist Å, et al. Risk factorsand potential preventive measures for nephropatia epide-mica in Sweden 2011–2012: a case–control study. InfectEcol Epidemiol. 2015;5:27698.

526 M. FURBERG ET AL.