Leptospirosis: Prognostic Model for Patient Mortality in the Transcarpathian Region,Ukraine

Introduction

Leptospirosis is a major threat to public health and one of the most important and widely distributed zoonoses in the world (Adler and Moctezuma, 2010; Ullmann and Langoni, 2011). Every year, it is estimated that nearly 1 million people are infected with leptospirosis, with 58,000 of them dying (Costa et al, 2015). This disease causes an annual loss of 2.9 million disability-adjusted life years (Torgerson et al, 2015).

In the Transcarpathian region of Ukraine leptospirosis is a significant public health concern. The incidence rate in this region is more than three times higher than the relative incidence in the remainder of Ukraine and the case fatality ratio in leptospirosis in Transcarpathia averages 12.5%, whereas the national level is 9.8% (Hopko, 2017; Markovych et al, 2019).

The symptom spectrum is broad, and leptospirosis shares clinical signs with many other acute febrile diseases, such as Congo-Crimean hemorrhagic fever and hemorrhagic fever with renal syndrome. Severe manifestations occur in 10–15% of human infections and are typified as Weil's syndrome (a triad of jaundice, hemorrhage, and acute renal failure), which has a 15–20% case fatality rate and severe pulmonary hemorrhage syndrome, which may present as acute respiratory distress. Weil's syndrome has an elevated case fatality rates of >50% in several studies (Gouveia et al, 2008).

Early detection of severe leptospirosis may be useful in reducing mortality (Daher et al, 2015; Dupont et al, 1997). Identifying these prognostic factors associated with an increased risk of death is important to establish early hospitalization so that increased monitoring and more aggressive therapeutic measures can be targeted to those most in need (Daher et al, 2015; Dupont et al, 1997).

Leptospirosis presentations vary in different geographical areas around the world (Spichler et al, 2008). This may due to difference in Leptospira species and serovars, as well as socioeconomic factors, and environmental factors. Variation in intrinsic virulence among serovars and species also have been proposed to explain some of the differences in disease severity between mild and severe forms of leptospirosis (Philip et al, 2021).

The factors that lead to severe forms to appear to have not been clearly identified (Herrmann-Storck et al, 2010; Paganin et al, 2007). This study sought to examine the risk and prognostic factors associated with laboratory-confirmed cases of severe leptospirosis in a localized region of the Transcarpathian region. This was done to help control for various other factors such as differences in care and treatment protocols, as well as obvious differences in Leptospira variants that might also impact results.

Materials and Methods

Results

There were 102 leptospirosis patients with confirmed positive MAT that were included in the study, 76 patients (74.5%) survived and 26 (25.5%) died.

Demographic data such as age and gender did not differ significantly between fatal and surviving patients. The groups were predominantly middle-aged males (Table 1). Clinical signs and symptoms associated with death only differed significantly for the presence of oliguria being more frequent among those who died (OR = 13.5; [2.56–71.12] p < 0.001).

Among patients dying from leptospirosis all (n = 26, 100%) were jaundiced although this did not differ significantly from those who survived (OR = 1.7; [1.39–2.12] p = 0.13). Similarly, both groups did not differ in the likelihood of developing a fever (OR = 0.30; [0.07–1.23] p = 0.12), nausea (OR = 2.0; [0.52–7.72] p = 0.31), myalgias (OR = 1.0; [0.23–4.59] p = 1.00), arthralgias (OR = 0.7; [0.62–0.86] p = 1.00), vomiting (OR = 1.2; [0.31–4.95] p = 0.73) or abdominal pain (OR = 3.27; [0.41–26.01] p = 0.26. The only sign that differed significantly was the occurrence of oliguria. A total of 22 (84.6%) fatal cases had oliguria, whereas a significantly lower proportion of surviving patients (28.9%) were oliguric (OR = 13.5; [2.56–71.12] p < 0.001).

In contrast with the paucity of signs and symptoms among fatal and surviving cases (oliguria being the single exception) (Table 1), laboratory results provided a greater distinction between those likely to survive and patients who died (Table 2).

Fatal cases had significantly higher serum creatinine levels compared with those who survived, namely 452.35 (349.87–527.50) μmol/L versus 126.00 (77.80–281.80) μmol/L, respectively (p = 0.001). Statistically significant changes were also found in the levels of urea (p = 0.001), total bilirubin (p = 0.001), direct bilirubin (p = 0.001), platelet level (p = 0.005), and white blood cell (WBC) count (p = 0.003). Alanine aminotransferase (ALT), the rate of erythrocyte sedimentation, and the percentage of granulocytes in the blood were not associated with the lethality of leptospirosis (p > 0.05).

The best model of laboratory results predicting fatal outcomes, using logistic regression included two laboratory findings: creatinine and direct bilirubin (with an included intercept).

The model had a very good ability to anticipate lethal outcomes and was especially successful at identifying patients that would survive. The sensitivity (true positive rate) of the prognostic model is 75%. The specificity (true negative rate) is 90.9%. The model had an area under receiver operating characteristic of 0.936 (95% CI, 0.89–0.98) to predict the lethality (Table 3; Fig. 1).

OPEN IN VIEWER

FIG. 1.

ROC curve based on the created regression model. ROC, receiver operator characteristic.

Discussion

Leptospirosis presents with a wide range of presentations from mild disease to fatal outcomes. Identifying early markers of a poor prognosis is critical in clinical settings for improved patient care.

This study sought to identify predictive factors for lethality among hospitalized patients with leptospirosis in the Transcarpathian region. Our findings suggest that oliguria is a significant risk factor for fatal leptospirosis in this population. This sign can be recognized early after hospitalization, and early identification and control of renal dysfunction can alert the treating physician to the possibility that the patient will die. Although not a predictive factor, as some fatal cases were not oliguric and a minority of surviving patients did suffer renal dysfunction, it was the best predictor of death. Consequently, aggressive treatment can be started immediately for patients with this sign.

The kidney is one of the main targets of Leptospira, with kidney damage occurring in 20–85% of patients (Meneses et al, 2020). Renal failure, marked by oliguria as a predictor of death especially when associated with pulmonary involvement, has been reported in some studies (Bharti et al, 2003; Silva Junior et al, 2011). Histologically, spirochetes are found in the renal tubules, associated with interstitial nephritis, and glomerular damage with tubular necrosis (Sellors et al, 2021). The mechanisms are unknown but is thought to be a mix of direct toxic injury, immune-mediated responses, and circulatory collapse (Cerqueira et al, 2008).

Our data suggest that risk stratification for patients should not rely on icterus. Although jaundice occurred in all fatal cases this was not significantly more common than in surviving patients. Jaundice results from damage to the vessels of the hepatic capillaries but without hepatocellular necrosis. There is disagreement among studies about the utility of jaundice as a sinister marker with some retrospective studies that confirm (Daher et al, 2009; Durmaz Cetin et al, 2004), and others deny (Daher et al, 2015; Spichler et al, 2008) its utility as a predictor of death.

Laboratory findings, either alone or in combination were especially helpful to identify at risk patients. Tests indicated renal and hepatic derangements. The best measures of outcome were a combination of elevated direct bilirubin and creatinine levels. Elevated greater total and direct bilirubin levels have been also reported, elsewhere (Daher et al, 2015; Dupont et al, 1997; Spichler et al, 2008), and an aspartate aminotransferase–ALT ratio of >3 has been reported as an indicator of a poorer prognosis (Chang et al, 2005).

In leptospirosis, liver enzymes such as ALT and AST are usually considered only moderately elevated and were not different between our groups. This suggests that liver impairment is generally mild and resolves with time (Bharti et al, 2003). The role of hepatic transaminases as a factor in disease severity or lethality also needs to be further researched (Daher et al, 2015; Hochedez et al, 2015; Sandhu et al, 2020).

When elevated direct bilirubin was combined with elevated creatinine levels there was the best predictor of patient outcome. Increased blood creatinine and urea levels indicate kidney damage and the potential for acute renal failure, one of the most common and crucial predictors of death in leptospirosis (Panaphut et al, 2002). The presence of renal failure in leptospirosis patients should be given special attention when it is accompanied by jaundice (Weil's syndrome) because it is the clinical syndrome most associated with the risk of death (Tantitanawat and Tanjatham, 2003).

Thrombocytopenia, which is frequent in leptospirosis, was also more severe and more common with fatal outcomes. It is another key risk factor for a severe course of the disease (Daher et al, 2014). It has been proposed that thrombocytopenia is indicative of certain strains of Leptospira that directly activate platelets (Wagenaar et al, 2007). However, we did not have data on strains in this region. Thrombocytopenia in the acute phase of disease can play a role in hemorrhagic disorders. In many studies, thrombocytopenia has been identified as one of the most common causes of serious courses and deaths. (Dupont et al, 1997; Panaphut et al, 2002; Spichler et al, 2008; Tantitanawat and Tanjatham, 2003). Another factor contributing to bleeding during the acute stage of the disease may be uremia. The pathophysiology of bleeding in uremia is multifaceted, nevertheless, alterations in platelet–platelet and platelet–vessel wall interaction play a crucial role. Platelet dysfunction is partly caused by uremic toxins in the circulatory blood (Boccardo et al, 2004).

Although gender and age have been reported to influence health outcomes with leptospirosis our study did not find these effects. Neither some studies (Dupont et al, 1997; Spichler et al, 2008) have observed an increased risk of death among older patients. Most studies (Daher et al, 2015; Dupont et al, 1997; Spichler et al, 2008), including ours, have discovered that gender was not a significant determinant in leptospirosis lethality. Men, however, were overall, more likely to be hospitalized compared with women (Spichler et al, 2008). Whether this reflects gender-based differences in activities and overall exposure risk requires further study with population-based surveys.

To reduce mortality, important predictors were identified. Patients presenting with oliguria on or shortly after admission should receive special attention. Once laboratory results are obtained serum creatinine and urea levels, as well as levels of direct and total bilirubin, platelets, and WBC count should be performed and these “red flag” laboratory and clinical characteristics should aid medical personnel in rapidly identifying a patient at risk of death. These patients need early intensive care, monitoring, and therapy adjustment.