Introduction

Leptospirosis is a zoonotic disease with worldwide distribution [1]. Human leptospirosis is a neglected and underreported disease [2], due to a lack of awareness, misdiagnosis and technically demanding laboratory confirmation tests [2,3]. Recent international initiatives on the assessment of the global disease burden of leptospirosis have estimated the global annual incidence of endemic human leptospirosis exceeding 5 severe cases per 100.000 population, excluding cases from outbreaks [2]. However, this number probably presents a substantial underestimation; as this incidence is derived from underreported, laboratory confirmed cases. The true total incidence of mild and severe leptospirosis most likely presents only a small portion of the true number of mild to severe cases annually.

Leptospirosis is caused by pathogenic spirochaetes of the genus Leptospira [4]. A wide variety of mammalian hosts, both feral and (semi-) domestic, serve as infection reservoirs. In developing countries, urban leptospirosis is an increasing public health hazard due to hygiene and overcrowding issues, however, farmers still represent the major risk group for leptospirosis [5]. In developed, temperate countries leptospirosis is mainly a recreational disease, associated with water-contact activities [1,2].

Leptospirosis is a protean disease ranging from a sub-clinical illness to a potentially fatal disease with haemorrhage and multi-organ failure as its clinical hallmarks. Persistent or chronic symptoms are often variable in character and onset [6]. This complicates the association between acute disease and sequelae [6,7] and makes it difficult to truly measure the burden of disease. Apart from some anecdotal case reports, post-leptospirosis complaints are not well assessed [3]. Data on the impact and occurrence of persistent or chronic complaints, together with data on the period of hospitalization are badly needed for assessing the burden of leptospirosis in terms of disability-adjusted life years (DALYs) [2,3].

Therefore, in this paper we describe the estimate of the burden of hospitalized and outpatient human leptospirosis in The Netherlands with regard to the duration of the acute and chronic illness as well as the incidence of persistent symptoms by active and passive case finding. We estimate the effect of intensive care unit (ICU) stay adjusted for the potential confounding effect of age, gender, infecting serogroup and occurrence of the acute severe complications of lung (acute lung injury, (ALI)) and kidney (acute renal injury (ARI)), liver injury and haemorrhage. As well, several studies have shown that ICU admission is a risk factor for both physical and psychological symptoms up to 12 months after discharge from the hospital [8]. Psychological consequences include depressive symptoms, post-traumatic stress disorders (PTSD), anxiety and chronic fatigue syndrome (CFS). These sequelae were associated with traumatic memories during ICU or hospital stay [9–11]. In the light of these studies regarding persistent symptoms, in this study it was considered of particular importance, to determine whether there is an association between ICU stay and persistent symptoms after the acute phase of leptospirosis.

Methods

Results

Factors associated with Persistent Complaints by Active Case finding (PCAC)

In total 321 questionnaires were distributed of which 256 (79.8%) were returned. Data from 31 questionnaires were excluded because the information provided was incomplete (n=28) or contained conflicting data (n=3). This resulted in 225 eligible participants’ questionnaires. The median age was 38 years (IQR 20-46 years) of which 214 were male (95.5%). This is comparable to those participants excluded, i.e. 33 years (23-46 years) and 88 males (91.7%), respectively.

The number of PCAC was 68 (30.2%), and for 65 (95.6%) cases these persistent complaints were specified. Depression-compatible complaints were most frequently reported, such as extreme fatigue, headache and malaise, or complaints that seemed to present persistence, such as myalgia and joint pains (Table 1). Several participants reported back pain and vertigo (3 (4.6%) patients, each), depression and concentration problems (2 (3.1%) patients, each). One participant reported sleep disorders and one reported severe mental alterations. 27 of 65 participants (39.7%) reported ‘other symptoms’, this included typical thoracic pain, perspiration, painful shoulders, stomach pain, and tinnitus.

The duration of PCAC was reported for 57 cases, 11 (19.3%) cases reported ≤2 months, and 1 case reported 18-24 months (Figure 1). However, 12 (21.1%) cases had chronic complaints for more than 24 months. Difficulties in returning to work after the acute phase of leptospirosis were reported in 9 out of 68 (13.2%) questionnaires, comprising the inability to work in a dusty environment (1/9), disability to work for 2 months (2/9), bed rest for 2 months (1/9), the need of assistance at the farm for 6 months (1/9), increasing inability to do farmers work (1/9), resigning the job because of disability of the associated labour (1/9) and permanent disability to work (2/9).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 1. Persistent complaints by active case finding (PCAC) stratified according to months of duration.

The bars represent the percentage of the total of reported post-leptospirosis complaints separated according to duration, expressed in periods of months as indicated on the X-axis.

https://doi.org/10.1371/journal.pone.0076549.g001

Data in Table 3 show that higher age (≥38 years) was positively associated with the occurrence of PCAC (OR 2.5, 95% CI 1.4-4.5), and males trended towards having more complaints than females (OR 4.1, 95% CI 0.5-33.0), although not significant. There was no association between age groups and duration of PCAC or the type of symptoms, although higher age showed a trend towards a higher proportion of myalgia (OR 3.3, 95% CI 0.7-16.4) and physical condition problems (OR 5.3, 95% CI 0.6-44.9).

Discussion

The main objective of this study was to estimate the burden of human leptospirosis in the Netherlands with regard to the frequency and nature of post leptospirosis symptoms as well as the duration of both the acute and persistent phase. In addition, the effects of age, gender, severity of acute illness, including ICU attendance and infecting serogroups were investigated as potential risk factors on the occurrence of post-leptospirosis complaints.

Dutch hospitalized patients had a median stay in the hospital of 10 days and spend 5 days of illness before hospitalization. The total duration of the acute phase thereby consisted of a median 16 days. This concurs with previous data on the duration of acute leptospirosis in the Netherlands and in other parts of the world [3,15]. A limitation of this data is that it concentrates on acute illness from severely ill patients attending the hospital and therefore excludes less severe cases, for which no hospital data were available. However, hospitalization occurred in almost 80% of cases. Moreover, less severe leptospirosis was recognized as such, suggesting a relative high grade of severity for these cases (obviously, patients did seek medical care and the clinician considered leptospirosis), comparable or exceeding that of an influenza-like disease. Taking into account that recovery from a severe influenza-like disease usually requires one to two weeks, ambulant cases are unlikely to considerably reduce the mean duration of acute leptospirosis. Additionally, it is estimated that at least 30% of severe patients in the Netherlands, are missed due to unawareness or mis-diagnosis [3,16]. Denial of effective treatment because of misdiagnosis probably results in a higher severity of complications and a longer duration of illness [17] and hence missed severe cases might well balance the decrease in hospital days caused by ambulant cases.

The duration of hospital stay showed a declining trend toward recent years. Although this does not exclude the increase of less severe forms of leptospirosis, this more likely reflects a general trend in the Netherlands aiming at the reduction of time of hospital attendance [18–20]. This can be partially explained by economic considerations in the Netherlands and can partially be attributed to developments in medical science and improved health awareness in the population [17]. Thus the decline in duration of hospital stay does not reflect a reduction of the burden of the acute phase of disease but rather indicates that its definition based on discharge of the hospital might become less appropriate.

Significant differences were found between the two age categories, with older patients having a longer hospital stay during the acute phase compared to younger patients. This is in accordance with various reports that indicate that increasing age is associated with increasing severity of leptospirosis [21]. This is not unique for leptospirosis but is commonly seen for a variety of infectious diseases in the Netherlands (CBS).

Significant differences were also found between males and females. Over 90% of the confirmed cases were male patients and they spent more time in the hospital than their females counterparts (11 versus 7.5 days). Similarly, the majority of ICU cases were males (89/93) with a median ICU stay of 7 days. This together indicates that males seem to acquire leptospirosis more frequently and develop more severe complications than females. This further substantiates earlier reports that suggested that males are more susceptible to severe leptospirosis than females [13,22]. The assessment of the economic impact of the disease burden is beyond the scope of this paper. However, it should be noted that the vast majority (86%) of severe cases are male patients between 20 and 65 years of age, which is the current working age in the Netherlands. Since females comprise less than 10% of those presenting with leptospirosis, it is likely they will have little effect on the national economic burden.

In 236 eligible participants from a total of 552 records information was collected on persistent complaints in the first month following the acute disease. This was most likely a representative sample, since the median age and sex ratio was comparable for those included and excluded cases. Nearly half of these patients (45.8%) reported continuing complaints one month after discharge from the hospital. This corroborates well with a previous prospective study in Brazil, where complaints were reported in 49% of former hospitalized patients at a follow up of about 20 days [7].

The assessment of PCAC comprised a sample of 225 eligible participants from a total group of 321 patients, with a comparable age and sex in the included and excluded groups. PCAC occurred for one third of the patients with a declining incidence towards a prolonged duration. However, one-fifth of the PCAC patients had complaints that continued more than two years. Complaints were generally depression-like symptoms such as extreme fatigue, malaise, and headache. Physical complaints included myalgia and joint pains affecting mobility. In this sample, only a few files included information on problems in executing ones job. However, it is well conceivable that these depression-like conditions also impacted on the ability to continue working and hence should be expected to constitute a burden to DALYs.

The PCAC study group was conducted from 1985 to 1993 and was contained within the CPC study group. This allowed for a comparison to investigate the association between the two data sets used for these two groups. The presence or absence of complaints was consistent for two-third of the patients who had both PCAC and CPC data. About half of the CPC group patients apparently had recovered from the post-leptospirosis complaints since PCAC was not indicated. However, for four PCAC group patients no CPC was documented. This can be explained by assuming that either the patients or the clinician considered CPC as normal at the ambulant consultation and did not include the data in the report. On the other hand, it is well conceivable that sequelae develop after apparent complete cure from the acute illness as has been reported for leptospirosis- related uveitis [23].

Evidence has consistently shown that serovars Icterohaemorrhagiae and Copenhageni of the Icterohaemorrhagiae group cause severe leptospirosis more often than serovar Hardjo-bovis (Sejroe group) and serovar Grippotyphosa (Grippotyphosa group) [12,14,24,25]. Assuming that sequelae notably occur after traumatic and stressful acute complications, one might expect to find persistent complaints largely corroborating with Icterohaemorrhagiae as the infecting serogroup. However, our data did not show any association between any infecting serogroup and PCAC, although we found an association compared to the Grippotyphosa serogroup, albeit not significant, between CPC and infections with the Icterohaemorrhagiae group. Surprisingly, however, this association was more pronounced with Sejroe as the infecting serogroup. These Sejroe infections are predominantly due to a dairy fever outbreak caused by serovar Hardjo-bovis in the period 1985-1995 [14]. Since complications requiring ICU attendance apparently were absent, we argue that Sejroe group infections cause less severe forms of leptospirosis. Thus, while there seems to be a correlation between the causative infection serotype and severity of illness on one hand, this is not the case between disease severity and the occurrence of persistent complaints on the other hand. Consistently, delayed physical and psychological recovery after Hardjo infection, taking several months to more than a year have been reported by others [26]. Research among randomly selected dairy farms in 1986 and 1987 showed that 63% of persons seropositive for leptospirosis suffered from prolonged excessive fatigue, which led to a partial or complete disability to work [27]. Taking these arguments together, we conclude that less severely ill dairy fever patients do have a relatively high risk for persistent complaints. The reason for this remains unclear, although it is tempting to speculate that the host-dependence of the serovar Hardjo [28] is associated with a greater invasive power of the central nervous system and/or induces other pathways in the host defence responses.

ICU stay is considered to be a traumatic cause of long lasting sequelae [8,9,11,29–31]. In concordance with this, ICU patients had about a twice-higher risk for persistent complaints that was significant for CPC. The risk even increased nearly a 12-fold when combined with dialysis as another stressful event [32].

Although data on the duration of acute leptospirosis and hospital and ICU stay have been presented in previous studies, these are scarce [3]. Hence, the data on incidence, duration of disease and the occurrence of persistent complaints assessed in this paper present a valuable contribution to the current initiatives to estimate the global burden of leptospirosis (WHO-LERG) [2].

This study has some caveats. Numbers, notably of ICU stay and of female cases are low and affect the statistical analyses robustness. Additionally, too few data on the length of ICU stay were available for adequately examining the association ICU stay and PCAC. Although the severity of illness was assessed by analysing several reliable indicators, such as ICU stay, dialysis, ALI and ARI, this study did not use a ‘severity of illness scoring system’ such as APACHE II score or SAPS score [31]. Finally, and probably most importantly, data on sequelae are based on persistent complaints after acute leptospirosis. This does not necessarily imply that these can be fully attributed to leptospirosis. Indeed, persistent complaints also have been reported for other infectious diseases and thus are not unique to leptospirosis. Uniqueness of persistent or chronic symptoms can only be made plausible when including a proper control group, which was not part of this study. Moreover, sequelae, generally consisting of depression-like symptoms, need to be corrected for similar complaints in the whole population in order to clarify that a proportion of the sequelae indeed can be exclusively attributed to the acute disease. We have addressed this point by consulting national databases on population-wide trends in health and lifestyle [33].

Compared to the persistent leptospirosis (CPC and PCAP of 45.8% and 30.2% respectively) the baseline data were much lower, i.e. 18.5% of the general Dutch population did not have a good health with several minor complaints summarized as malaise and likely only for part constitute complaints with a severity that has been reported for persistent leptospirosis. Additionally, there is little correlation between malaise complaints and age in the general population [33], while for leptospirosis such complaints increase with increasing age. Therefore, it is concluded that acute leptospirosis induces persistent and late sequelae that contribute to the disease burden. Reports on chronic leptospirosis usually concern anecdotic case reports. This is one of the few, if not first, systematic reports on the burden of both acute and chronic leptospirosis and as such forms a highly valuable contribution to the assessment of its global burden.

Supporting Information

Acknowledgments

We are appreciative of the contributions of Drs. P. Speelman and WJ Terpstra in the active case finding study. We also wish to thank the participants in the meetings of the World Health Organization, Leptospirosis Epidemiology Reference Group (LERG) and the Global Leptospirosis Environmental Action Network (GLEAN) for their fruitful discussions.