- Email: [email protected]
Hypocalcemia, hypochloremia, and eosinopenia as clinical predictors of leptospirosis: A retrospective study
G Model
ARTICLE IN PRESS
JIPH-1151; No. of Pages 5
Journal of Infection and Public Health xxx (2019) xxx–xxx
Contents lists available at ScienceDirect
Journal of Infection and Public Health journal homepage: http://www.elsevier.com/locate/jiph
Hypocalcemia, hypochloremia, and eosinopenia as clinical predictors of leptospirosis: A retrospective study Cheng-Yee Fish-Low a , Ahmed D. Balami b , Leslie T.L. Than a , King-Hwa Ling c , Niazlin Mohd Taib a , Anim Md. Shah d , Zamberi Sekawi a,∗ a Department of Medical Microbiology and Parasitology, Faculty of Medicine and Health Sciences, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia b Department of Community Health, Faculty of Medicine and Health Sciences, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia c Department of Biomedical Sciences, Faculty of Medicine and Health Sciences, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia d Department of Medicine, Faculty of Medicine and Health Sciences, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia
a r t i c l e
i n f o
Article history: Received 1 December 2018 Received in revised form 15 July 2019 Accepted 27 July 2019 Keywords: Leptospirosis Diagnostic scoring Hypocalcemia Hypochloremia Eosinopenia
a b s t r a c t Background: Underestimation of leptospirosis cases is happening in many countries. The most common factor of underreporting is misdiagnosis. Considering the limitations of direct detection of pathogen and serological diagnosis for leptospirosis, clinical features and blood tests though non-specific are usually referred in making presumptive diagnosis to decide disease management. Methods: In this single-centre retrospective study, comparative analysis on clinical presentations and laboratory findings was performed between confirmed leptospirosis versus non-leptospirosis cases. Results: In multivariate logistic regression evidenced by a Hosmer-Lemeshow significance value of 0.979 and Nagelkerke R square of 0.426, the predictors of a leptospirosis case are hypocalcemia (calcium <2.10 mmol/L), hypochloremia (chloride <98 mmol/L), and eosinopenia (absolute eosinophil count <0.040 × 109 /L). The proposed diagnostic scoring model has a discriminatory power with area under the curve (AUC) 0.761 (p < 0.001). A score value of 6 reflected a sensitivity of 0.762, specificity of 0.655, a positive predictive value of 0.38, negative predictive value of 0.91, a positive likelihood ratios of 2.21, and a negative likelihood ratios of 0.36. Conclusion: With further validation in clinical settings, implementation of this diagnostic scoring model is helpful to manage presumed leptospirosis especially in the absence of leptospirosis confirmatory tests. © 2019 The Authors. Published by Elsevier Limited on behalf of King Saud Bin Abdulaziz University for Health Sciences. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).
Introduction A broad array of clinical manifestations has been documented for human leptospirosis [1,2]. Leptospiral infections may be asymptomatic, or with symptoms ranging from mild and self-limiting febrile illnesses to multi-organ dysfunction with fatal outcomes. Leptospirosis may mimic other febrile illnesses, which makes the clinical manifestations not fully reliable for the diagnosis of leptospirosis. Underreporting due to misdiagnosis of leptospirosis is therefore common. Reliance on laboratory diagnostic tests to differentiate leptospirosis from other diseases is not entirely practicable considering the limitations of currently available leptospirosis diagnostic tests.
∗ Corresponding author. E-mail address: [email protected] (Z. Sekawi).
The microscopic agglutination test (MAT) for detection of agglutinating antibodies [3] serves as the gold standard of serological diagnosis for leptospirosis [2]. MAT is not an ideal test for leptospirosis diagnosis especially during the bacteremia phase, and in regions where the diagnostic facilities are poorly accessible or not available. Although there are many other commercially available rapid serological tests, the problem of delayed diagnosis due to late sero-conversion and undetectable antibodies titre still exists. On the other hand, direct detection of leptospiral DNA by polymerase chain reaction (PCR) has been proposed to be used routinely in leptospirosis diagnosis as it overcomes the limitations of MAT especially during the early acute phase [4]. However, it may have more stringent requirements on sampling timing and type of specimen. This is because leptospiremia may end very soon after the onset of symptoms [5]. The success of molecular detection of leptospiral DNA depends not only on the bacterial load in clinical materials, but also on the specificity
https://doi.org/10.1016/j.jiph.2019.07.021 1876-0341/© 2019 The Authors. Published by Elsevier Limited on behalf of King Saud Bin Abdulaziz University for Health Sciences. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Please cite this article in press as: Fish-Low C-Y, et al. Hypocalcemia, hypochloremia, and eosinopenia as clinical predictors of leptospirosis: A retrospective study. J Infect Public Health (2019), https://doi.org/10.1016/j.jiph.2019.07.021
G Model JIPH-1151; No. of Pages 5
ARTICLE IN PRESS C.-Y. Fish-Low et al. / Journal of Infection and Public Health xxx (2019) xxx–xxx
2
and sensitivity of the primers or probes that are used in the detection. To address the non-availability of a rapid confirmatory diagnostic test for leptospirosis, several scoring models have been developed as guidelines in diagnosing presumptive leptospirosis. Faine’s criteria [6] were first developed based on clinical data (Part A), epidemiological factors (Part B), and bacteriological and laboratory findings based on only MAT titre (Part C). A score of 26 or more in Part A or in Part A + Part B, or 25 or more as the total of three parts may suggest a presumptive diagnosis. Faine’s criteria were then modified, hence known as modified Faine’s criteria [7] to include other formats of serological tests under Part C. The modified Faine’s criteria were later amended (known as modified Faine’s criteria with amendment) to improve the usefulness in clinical settings [8]. Despite several amendments of the guidelines, a scoring section based on blood test profile has never been incorporated. This study aimed to determine the clinical predictors for leptospirosis based on comparative analysis of blood test parameters between confirmed leptospirosis versus non-leptospirosis cases. Then a scoring model is developed for the determined predictors. Material and methods Ethics statement The study protocol was approved by the Institutional Review Board of Faculty of Medicine and Health Sciences, Universiti Putra Malaysia, and the Medical Research Ethics Committee (MREC), Ministry of Health Malaysia (NMRR-15-2148-27536). Study design and study subjects This study utilized a cross-sectional study design among clinically suspected leptospirosis patients [2] admitted in Hospital Serdang, Selangor, Malaysia. Since leptospirosis was gazetted as a notifiable disease in Malaysia in December 2010, patients admitted from 2011 to 2017 constituted the sampling frame. Patients in the paediatric age group (below 18 years old) and patients with past history of autoimmune diseases or any known comorbidities were excluded. The one proportion formula was used to calculate the minimum required sample size [9]. The expected prevalence (P) of leptospirosis (8.4%) was taken from a previous study in Malaysia [10]. The Z-statistic and precision level (d) were substituted with 1.96 and 0.05 respectively [11], and an adjustment for 90% eligibility to obtain a minimum sample size requirement of 133 respondents. Case notes of all suspected leptospirosis patients admitted between 2011 and 2017 were extracted from the hospital’s medical patients’ electronic database, using the ‘Filter’ function. The file serial numbers of these patients was then copied and transferred to Excel spreadsheet from where 137 out of these were randomly selected using the ‘Random’ function.
of total antibodies in MAT [2] using a 20-serovar panel, which consisted of Australis, Autumnalis, Bataviae, Canicola, Celledoni, Djasiman, Grippotyphosa, Hardjoprajitno, Icterohaemorrhagiae, Javanica, Patoc, Pomona, Pyrogenes, Tarassovi, IMR LEP/1, IMR LEP/115, IMR LEP/175, IMR LEP 803/11, IMR LEP/27, and IMR LEP/22. A confirmed leptospirosis case (CLC) is defined as single serum titre of 1:400 or at least a four-fold rise in titre of paired sera in the standard MAT, a positive culture of Leptospira in blood or urine sample, or positive detection of Leptospira DNA in PCR [12]. All cases of confirmed leptospirosis are tested negative in both aerobic and anaerobic blood culture, and negative in dengue IgM and NS1 detection. The cases otherwise are non-leptospirosis (NLC). Statistical analysis The data collected was analysed by using IBM Statistical Package for Social Sciences (SPSS) version 22. The laboratory findings were categorized as normal, low or high, based on the hospital’s pre-defined cut-off values. Comparative analysis between CLC (n = 30) and NLC (n = 107) was performed. Bivariate analysis using the Chi-square test was run on all the variables, to determine any difference between the two groups. Variables significant in the bivariate analysis along with other variables with p-value <0.25 were entered into a multivariate logistic regression using the Backward-LR method, and the best model was selected to obtain the predictors of leptospirosis infection. Development of diagnostic scoring model In line with a previous similar study [14], the beta coefficients of the predictor variables were used to develop a practical scoring system. All the coefficients were divided by the smallest coefficient, and then multiplied by 4, the results of which were rounded up to the nearest whole number. This scoring system was then used to generate the receiver operating characteristics (ROC) curve. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and likelihood ratios (LR) were also computed. Results A total of 137 subjects were included in the study, 72.3% of whom were males. Their ages ranged from 18 to 86 years, and a mean (SD) age of 35.5 (14.8) years. Their ethnicities were: Malay (68.6%), Indian (13.1%), Chinese (6.6%) and others (11.7%). Around one-fifth (21.9%) had a diagnosis of confirmed leptospirosis. Out of the 30 cases of confirmed leptospirosis, 53.3% (n = 16) gave positive PCR result, 36.7% (n = 11) was positive by MAT, 3.3% (n = 1) was positive in blood cultures, and 6.7% (n = 2) of the cases were positive in both urine cultures and MAT. Bivariate analysis
Data collection and case definition A structured proforma was used to collect data from the selected case notes. Data on clinical presentations were collected on a binary scale (Yes or No), while the values for the laboratory findings were collected in their raw numerical forms as continuous variables. The cases were categorized into confirmed leptospirosis versus non-leptospirosis. The confirmatory diagnosis of leptospirosis was performed by the Institute for Medical Research (IMR), Ministry of Health Malaysia [12]. The tests performed were pathogen isolation [2], detection of leptospiral DNA by real-time polymerase chain reaction [13], and and detection
Table 1 shows the comparison of clinical and laboratory characteristics of CLC and NLC. There was a significant difference between the two groups in their proportions of temperature above 39 ◦ C, low calcium (<2.10 mmol/L), high urea (>9.2 mmol/L), low sodium (<136 mmol/L), low chloride (<98 mmol/L), and low absolute eosinophil count (<0.040 × 109 /L). Multivariate analysis For the multivariate logistic regression, the model fitted the sample, evidenced by a Hosmer-Lemeshow significance
Please cite this article in press as: Fish-Low C-Y, et al. Hypocalcemia, hypochloremia, and eosinopenia as clinical predictors of leptospirosis: A retrospective study. J Infect Public Health (2019), https://doi.org/10.1016/j.jiph.2019.07.021
G Model
ARTICLE IN PRESS
JIPH-1151; No. of Pages 5
C.-Y. Fish-Low et al. / Journal of Infection and Public Health xxx (2019) xxx–xxx
3
Table 1 Comparison of clinical and laboratory features of confirmed (CLC) and non-confirmed leptospirosis cases (NLC). NLC (n = 107)
Characteristic Headache Fever Temperature >39◦ C Conjunctival suffusion Myalgia Jaundice Haemoptysis / shortness of breath Skin rash Gastrointestinal symptoms Haemorrhages Creatinine kinase <29 U/L Creatinine kinase >200 U/L C-reactive protein >5 mg/L Calcium <2.10 mmol/L Calcium >2.55 mmol/L Magnesium <0.85 mmol/L Magnesium >1.05 mmol/L Albumin <35 g/L Albumin >50 g/L Bilirubin, total >20.5 mol/L Alkaline phosphatase <40 U/L Alkaline phosphatase >150 U/L Aspartate transferase >34 U/L Urea <2.5 mmol/L Urea >9.2 mmol/L Sodium <136 mmol/L Potassium <3.5 mmol/L Potassium >5.1 mmol/L Chloride <98 mmol/L Chloride >107 mmol/L Creatinine <53 mol/L Creatinine >115 mol/L Red blood cell <3.80 × 1012 /L Red blood cell >5.80 × 1012 /L Haemoglobin <11.5 g/dL Haemoglobin >18.0 g/dL Platelet <150 × 109 /L Platelet >400 × 109 /L White blood cell <4.0 × 109 /L White blood cell >11.0 × 109 /L Absolute neutrophil <2.00 × 109 /L Absolute neutrophil >7.50 × 109 /L Absolute lymphocyte <1.50 × 109 /L Absolute lymphocyte >4.00 × 109 /L Absolute monocyte <0.200 × 109 /L Absolute monocyte >0.800 × 109 /L Absolute eosinophil <0.040 × 109 /L Absolute eosinophil >0.400 × 109 /L *
CLC (n = 30)
p-value
Frequency
(%)
Frequency
(%)
48 99 17 14 72 11 14 15 83 6 4 63 66 43 1 51 8 57 1 32 2 14 83 8 17 66 23 5 26 6 1 38 6 19 13 3 24 6 11 33 7 42 65 1 9 43 24 9
(44.9) (92.5) (15.9) (13.1) (67.3) (10.3) (13.1) (14.0) (77.6) (5.6) (3.7) (58.9) (81.5) (62.3) (1.4) (72.9) (11.4) (54.3) (1.0) (30.2) (1.9) (13.1) (77.6) (7.5) (15.9) (61.7) (21.5) (4.7) (24.3) (5.6) (0.9) (35.8) (5.6) (17.8) (12.1) (2.8) (22.4) (5.6) (10.3) (30.8) (6.5) (39.9) (60.7) (0.9) (8.4) (40.2) (22.4) (8.4)
16 28 10 3 20 3 4 4 27 2 0 19 18 20 0 15 2 18 0 10 1 4 23 2 12 26 8 0 16 1 0 15 1 7 1 1 9 0 5 10 5 12 23 0 1 10 14 1
(53.3) (93.3) (33.3) (10.0) (66.7) (10.0) (13.3) (13.3) (90.0) (6.7) (0.0) (63.3) (90.0) (90.9) (0.0) (68.2) (9.1) (60.0) (0.0) (33.3) (3.3) (13.3) (76.7) (6.7) (40.0) (86.7) (26.7) (0.0) (53.3) (3.3) (0.0) (50.0) (3.3) (23.3) (3.3) (3.3) (30.0) (0.0) (16.7) (33.3) (16.7) (40.0) (76.7) (0.0) (3.3) (33.3) (46.7) (3.3)
0.411 0.880 0.034* 0.651 0.949 0.964 0.972 0.924 0.130 0.827 0.657 0.830 0.362 0.014* 1.000 0.476 0.529 0.614 0.614 0.742 0.526 1.000 0.917 0.767 0.004* 0.010* 0.640 0.364 0.004* 0.972 1.000 0.172 0.664 0.527 0.160 1.000 0.490 0.204 0.285 0.601 0.069 0.588 0.069 0.588 0.439 0.367 0.014* 0.577
Statistically significant at p < 0.05.
Table 2 Predictors of presumptive leptospirosis. Factors
B
SE
Wald
df
p-value
Adjusted OR
95% CI
Calcium Normal Low (<2.10 mmol/L)
– 2.04
– 1.00
– 4.17
– 1
– 0.041
1.00 7.72
– 1.09–54.98
Chloride Normal Low (<98 mmol/L)
– 1.70
– 0.77
– 4.82
– 1
– 0.028
1.00 5.45
– 1.20–24.79
Absolute eosinophil Normal Low (<0.040 × 109 /L)
– 1.76
– 0.73
– 5.75
– 1
– 0.017
1.00 5.80
– 1.38–24.41
B = beta coefficient; SE = standard error; df = degrees of freedom; OR = odd ratio; CI = confidence interval.
value of 0.979. The Nagelkerke R square was 0.426, and the predictors of being a case of leptospirosis are presented in Table 2. Table 3 illustrates the proposed score for diagnosis of leptospirosis, based on the multivariate analysis results.
Fig. 1 shows the receiver operating characteristics (ROC) curve to differentiate between leptospirosis and non-leptospirosis. The area under the curve (AUC) was 0.761 (p < 0.001). As presented in Table 4, score value of 6 reflected a sensitivity of 0.762, specificity of 0.655, a positive predictive value (PPV) of 0.38, negative predictive
Please cite this article in press as: Fish-Low C-Y, et al. Hypocalcemia, hypochloremia, and eosinopenia as clinical predictors of leptospirosis: A retrospective study. J Infect Public Health (2019), https://doi.org/10.1016/j.jiph.2019.07.021
G Model
ARTICLE IN PRESS
JIPH-1151; No. of Pages 5
C.-Y. Fish-Low et al. / Journal of Infection and Public Health xxx (2019) xxx–xxx
4
Table 3 Proposed score for diagnosis of leptospirosis, based on multivariate analysis. Factor
Beta coefficient
Calculated score
Score (rounded to nearest integer)
Calcium <2.10 mmol/L Chloride <98 mmol/L Eosinophil <0.040 × 109 /L
2.044 1.696 1.758
4.80 4.00 4.15
5 4 4
Maximum possible score
13
value (NPV) of 0.91, a positive likelihood ratios (PLR) of 2.21, and a negative likelihood ratios (NLR) of 0.36. Discussion As observed in the present study and other similar studies that compared serum biochemistry between leptospirosis and non-leptospirosis patients, the variables proposed in the respective diagnostic score models may be cohort- or region-specific. A Sri Lankan cohort-based study [14] reveals high creatinine (>150 mol/L), high bilirubin (>30 mol/L), neutrophil >80%, low platelet count (<85,000/mm3 ), and exposure to possible source of leptospirosis are positively associated with the confirmed diagnosis. A recent Thai cohort-based study [15] reported the presence of hypotension, jaundice, muscle pain, acute kidney failure, low haemoglobin (<12 g/dL), and hypokalemia (potassium <3.5 mEq/L) with hyponatremia (sodium <135 mEq/L) are significant predictors for leptospirosis. On the other hand, hypocalcemia (calcium <2.10 mmol/L), hypochloremia (chloride <98 mmol/L), and eosinopenia (<0.040 × 109 /L) are proposed as the scoring criteria for leptospirosis in this Malaysian cohort-based study. Despite none of the predictors in any of the abovementioned studies are present in another study, abnormal renal panel is common across these studies. Perturbations of electrolytes: chloride, potassium, sodium, and creatinine (waste product), and calcium (minerals) may serve as indications of renal involvement during leptospirosis. Hypokalemia is regularly seen in leptospirosis [16,17] whereas hyponatremia resolves quickly as soon as the infection is controlled [18]. Leptospirosis-induced acute kidney injury (AKI) causes rapid increment of serum creatinine and blood urea nitrogen [19]. Though not proven in the context of leptospirosis, dyschloremia is proposed as a risk factor for development of AKI, and hypochloremic patients are particularly having higher mortality rate [20]. Mild hypocalcemia usually develops asymptomatically and unnoticed or otherwise may show cardiac arrest, seizures, numbness, and confusion [21,22]. It can be caused by kidney failure, rhabdomyolysis, overdose of calcium channel blocker, and medications [21]. Hypocalcemia in leptospirosis is recognized as one of the reasons leading to arrhythmias [23,24]. The biphasic calcium kinetic in leptospirosis that was reported by Zakout R et al. [23] is found consistent with that seen in rhabdomyolysis-induced AKI [25], which hypocalcemia occurred in oliguric phase and then hypercalcemia in diuretic phase during recovery. This sequential
Fig. 1. Coordinates of the ROC curve for outcome—positive leptospirosis.
observation may be explained by calcium deposition in damaged muscle followed by resolving calcification once the renal functions have recovered [26]. Phenomenon of low eosinophil count in leptospirosis is either has not been evaluated or lacks of emphasis in previous studies. It has once been reported in canine leptospirosis [27]. Eosinopenia is evidenced to be associated with peritonitis [28], sepsis [29], and enteric fever [30]. Bass DA et al. [31] speculated that the migration of eosinophils to the inflammatory site, which was triggered by chemotactic factors during acute inflammation has resulted in eosinopenia. Further analysis on renal profile of leptospirosis patients would be beneficial and have diagnostic values. Kidney is the known target organ where leptospires may colonize and persist [32,33], possibly related to their ability of forming biofilm [34]. In addition, leptospiral outer membrane proteins have been demonstrated to induce renal inflammation and interstitial nephritis [35]. Collectively, bacterial invasion, inflammation, haemodynamic perturbations and, pathogenicity of bacterial products contribute to the development of leptospiral nephropathy [19]. Varying clinical spectrum of renal involvement are documented in leptospirosis, ranging from mild course of proteinuria [36], serum electrolyte changes [16] to severe acute renal failure associated with rhabdomyolysis [37]. Conclusion Confirmatory diagnosis of leptospirosis is challenging and currently limited by many well recognized factors. A scoring model
Table 4 Diagnostic model performance (sensitivity, specificity, positive and negative predictive value and likelihood ratios). Score
Sensitivity
Specificity
PPV
NPV
PLR
NLR
4 6 10
0.810 0.762 0.238
0.483 0.655 0.966
0.29 0.38 0.38
0.96 0.91 0.89
1.45 2.21 2.18
0.14 0.36 0.43
PPV = positive predictive value; NPV = negative predictive value; PLR = positive likelihood ratios; NLR = negative likelihood ratios.
Please cite this article in press as: Fish-Low C-Y, et al. Hypocalcemia, hypochloremia, and eosinopenia as clinical predictors of leptospirosis: A retrospective study. J Infect Public Health (2019), https://doi.org/10.1016/j.jiph.2019.07.021
G Model JIPH-1151; No. of Pages 5
ARTICLE IN PRESS C.-Y. Fish-Low et al. / Journal of Infection and Public Health xxx (2019) xxx–xxx
based on clinical predictors of blood test parameter, particularly the renal profile will efficiently improve leptospirosis diagnosis. It may be of great diagnostic potential if the proposed scoring model to be combined with both epidemiological risk factors and clinical manifestations. Case history such as contact with possible infected animals or exposure to leptospire-containing water is especially useful in the diagnosis of leptospirosis as leptospires are mostly transmitted via direct or indirect contact with infected animals. We acknowledge the limitations of this study as the enrolled cases may not be representative of all leptospirosis cases in Malaysia. Until the proposed scoring model is validated in larger cohort size and involves multicentre, it shows potential to be incorporated into the modified Faine’s criteria with amendment, or can be applied as a standalone diagnostic guideline for leptospirosis. Funding This work was funded by the Ministry of Higher Education, Malaysia Long Term Research Grant Scheme (LRGS) [UPM/7002/1/LRGS/5526400]. Competing interest None declared. Ethical approval Not required. Acknowledgements We would like to thank the Director General of Health Malaysia for his permission to publish this article. The authors are indebted to all the clinicians and supporting staff from Hospital Serdang, Malaysia. The authors also express their appreciation to all the members of Malaysia Leptospirosis Research Network (MyLepto) and to all the study participants. References [1] Centers for Disease Control and Prevention. Infectious diseases related to travel. New York: Oxford University Press; 2017. [2] World Health Organization. Human leptospirosis: guidance for diagnosis, surveillance and control. World Health Organization; 2003, http://dx.doi.org/ 10.1590/S0036-46652003000500015. [3] Turner LH. Leptospirosis II: serology. Trans R Soc Trop Med Hyg 1968;62:880–99, http://dx.doi.org/10.1016/0035-9203(68)90017-5. [4] Mullan S, Panwala TH. Polymerase chain reaction: an important tool for early diagnosis of leptospirosis cases. J Clin Diagn Res 2016;10:DC08–11, http://dx. doi.org/10.7860/JCDR/2016/22462.9010. [5] Turner LH, Leptospirosis I. Trans R Soc Trop Med Hyg 1967;61:842–55, http:// dx.doi.org/10.1016/0035-9203(67)90045-4. [6] Faine S. Guidelines for the control of leptospirosis. Geneva: WHO offset publication; 1982, no. 67. [7] Shivakumar S, Shareek PS. Diagnosis of leptospirosis utilizing modified Faine’s criteria. J Assoc Phys India 2004;52:678–9. [8] Shivakumar S. Indian guidelines for the diagnosis and management of human leptospirosis. Med Update 2013;23:23–9. [9] Daniel WW. Biostatistics: a foundation for analysis in the health sciences. 7th ed. New York: Wiley; 1999. [10] Rafizah AAN, Aziah BD, Azwany YN, Imran MK, Rusli AM, Nazri SM, et al. A hospital-based study on seroprevalence of leptospirosis among febrile cases in northeastern Malaysia. Int J Infect Dis 2013;17:e394–7, http://dx.doi.org/10. 1016/j.ijid.2012.12.012. [11] Naing L, Winn T, Rusli BN. Practical issues in calculating the sample size for prevalence studies. Arch Orofac Sci 2006;1:9–14. [12] Ministry of Health Malaysia. Guidelines for diagnosis management, prevention and control of leptospirosis. 1st ed. BKPKKM; 2011. [13] Ahmed A, Engelberts MFM, Boer KR, Ahmed N, Hartskeerl RA. Development and validation of a real-time PCR for detection of pathogenic Leptospira species in clinical materials. PLoS One 2009;4:e7093, http://dx.doi.org/10.1371/journal. pone.0007093.
5
[14] Rajapakse S, Weeratunga P, Niloofa R, Fernando N, de Silva NL, Rodrigo C, et al. A diagnostic scoring model for leptospirosis in resource limited settings. PLoS Negl Trop Dis 2016;10:e0004513, http://dx.doi.org/10.1371/journal.pntd. 0004513. [15] Sukmark T, Lumlertgul N, Peerapornratana S, Khositrangsikun K, Tungsanga K, Sitprija V, et al. Thai-Lepto-on-admission probability (THAI-LEPTO) score as an early tool for initial diagnosis of leptospirosis: result from Thai-Lepto AKI study group. PLoS Negl Trop Dis 2018;12:e0006319, http://dx.doi.org/10.1371/ journal.pntd.0006319. [16] Seguro AC, Lomar AV, Rocha AS. Acute renal failure of leptospirosis: nonoliguric and hypokalemic forms. Nephron 1990;55(146), http://dx.doi.org/10. 1159/000185943. [17] Abdulkader RCRM, Seguro AC, Malheiro PS, Burdmann EA, Marcondes M. Peculiar electrolytic and hormonal abnormalities in acute renal failure due to leptospirosis. Am J Trop Med Hyg 1996;54(1), http://dx.doi.org/10.4269/ajtmh. 1996.54.1. [18] Sitprija V. Altered fluid, electrolyte and mineral status in tropical disease, with an emphasis on malaria and leptospirosis. Nat Clin Pract Nephrol 2008;4(91), http://dx.doi.org/10.1038/ncpneph0695. [19] Visith S, Kearkiat P. Nephropathy in leptospirosis. J Postgrad Med Educ Res 2005;51:184–8. [20] Shao M, Li G, Sarvottam K, Wang S, Thongprayoon C, Dong Y, et al. Dyschloremia is a risk factor for the development of acute kidney injury in critically ill patients. PLoS One 2016;11:e0160322, http://dx.doi.org/10.1371/journal.pone. 0160322. [21] Soar J, Perkins GD, Abbas G, Alfonzo A, Barelli A, Bierens JJLM, et al. European resuscitation council guidelines for resuscitation 2010 section 8. Cardiac arrest in special circumstances: electrolyte abnormalities, poisoning, drowning, accidental hypothermia, hyperthermia, asthma, anaphylaxis, cardiac surgery, trauma, pregna. Resuscitation 2010;81:1400–33, http://dx.doi.org/10.1016/j. resuscitation.2010.08.015. [22] Fong J, Khan A. Hypocalcemia: updates in diagnosis and management for primary care. Can Fam Physician 2012;58:158–62. [23] Zakout R, do Carmo M, Freitas P. Reversible myocardial calcification following severe leptospirosis complicated with rhabdomyolysis-induced acute kidney injury and magnesium-wasting nephropathy. J Med Cases 2013;4:92–8. [24] de Sousa Soares D, Galdino GS, Rodrigues BC, Silva Jr G, Daher EF. Arrhythmias in leptospirosis-associated acute kidney injury: a case series. Braz J Infect Dis 2017;21(209), http://dx.doi.org/10.1016/j.bjid.2016.11.010. [25] Pakkyara A, Mohsin N, Budruddin M, Mohammad AJE, Mohammad K. Calcium kinetic in a patient with acute renal failure due to rhabdomyolysis. A case report and review of literature. Oman Med J 2010;25, http://dx.doi.org/10.5001/omj. 2010.93, e013–e013. [26] Akmal M, Goldstein DA, Telfer N, Wilkinson E, Massry SG. Resolution of muscle calcification in rhabdomyolysis and acute renal failure. Ann Intern Med 1978;89:928–30, http://dx.doi.org/10.7326/0003-4819-89-6-928. [27] Rentko VT, Clark N, Ross LA, Schelling SH. Canine leptospirosis: a retrospective study of 17 cases. J Vet Intern Med 1992;6:235–44, http://dx.doi.org/10.1111/ j.1939-1676.1992.tb00345.x. [28] Jagdeesh TS, Mishra A, Saxena A, Sharma D. Eosinopenia as a prognostic marker in patients with peritonitis. ISRN Infect Dis 2013;2013:8, http://dx.doi.org/10. 5402/2013/540948. [29] Lavoignet CE, Le Borgne P, Slimani H, Forato M, Kam C, Kauffmann P, et al. Relevance of eosinopenia as marker of sepsis in the emergency department. La Rev Médecine Interne 2016;37(730), http://dx.doi.org/10.1016/j.revmed.2016. 02.018. [30] Matono T, Kutsuna S, Kato Y, Katanami Y, Yamamoto K, Takeshita N, et al. Role of classic signs as diagnostic predictors for enteric fever among returned travellers: relative bradycardia and eosinopenia. PLoS One 2017;12:e0179814, http://dx.doi.org/10.1371/journal.pone.0179814. [31] Bass DA, Gonwa TA, Szejda P, Cousart MS, DeChatelet LR, McCall CE. Eosinopenia of acute infection: production of eosinopenia by chemotactic factors of acute inflammation. J Clin Invest 1980;65:1265–71, http://dx.doi.org/10.1172/ JCI109789. [32] Athanazio DA, Silva EF, Santos CS, Rocha GM, Vannier-Santos MA, McBride AJA, et al. Rattus norvegicus as a model for persistent renal colonization by pathogenic Leptospira interrogans. Acta Trop 2008;105(176), http://dx.doi.org/ 10.1016/j.actatropica.2007.10.012. [33] Ratet G, Veyrier FJ, Fanton d’Andon M, Kammerscheit X, Nicola M-A, Picardeau M, et al. Live imaging of bioluminescent Leptospira interrogans in mice reveals renal colonization as a stealth escape from the blood defenses and antibiotics. PLoS Negl Trop Dis 2014;8:e3359, http://dx.doi.org/10.1371/journal. pntd.0003359. [34] Ristow P, Bourhy P, Kerneis S, Schmitt C, Prevost M-C, Lilenbaum W, et al. Biofilm formation by saprophytic and pathogenic leptospires. Microbiology 2008;154:1309–17, http://dx.doi.org/10.1099/mic.0.2007/014746-0. [35] Yang CW, Wu MS, Pan MJ. Leptospirosis renal disease. Nephrol Dial Transplant 2001;16(Suppl):73–7. [36] Thresiamma KC, Asha B, Reetika C, Manjula S, Chithra J, Sunil T, et al. Proteinuria in early detection of human leptospirosis. Int J Res Med Sci 2017;5:646–52, http://dx.doi.org/10.18203/2320-6012.ijrms20170168. [37] Droulias J, Moutzouris D, Kassimatis T, Kollia K, Apostolou T, Hadjikonstantinou V. Rhabdomyolysis-induced acute renal failure in a patient with leptospirosis. Saudi J Kidney Dis Transpl 2007;18:430–1.
Please cite this article in press as: Fish-Low C-Y, et al. Hypocalcemia, hypochloremia, and eosinopenia as clinical predictors of leptospirosis: A retrospective study. J Infect Public Health (2019), https://doi.org/10.1016/j.jiph.2019.07.021
ARTICLE IN PRESS
JIPH-1151; No. of Pages 5
Journal of Infection and Public Health xxx (2019) xxx–xxx
Contents lists available at ScienceDirect
Journal of Infection and Public Health journal homepage: http://www.elsevier.com/locate/jiph
Hypocalcemia, hypochloremia, and eosinopenia as clinical predictors of leptospirosis: A retrospective study Cheng-Yee Fish-Low a , Ahmed D. Balami b , Leslie T.L. Than a , King-Hwa Ling c , Niazlin Mohd Taib a , Anim Md. Shah d , Zamberi Sekawi a,∗ a Department of Medical Microbiology and Parasitology, Faculty of Medicine and Health Sciences, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia b Department of Community Health, Faculty of Medicine and Health Sciences, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia c Department of Biomedical Sciences, Faculty of Medicine and Health Sciences, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia d Department of Medicine, Faculty of Medicine and Health Sciences, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia
a r t i c l e
i n f o
Article history: Received 1 December 2018 Received in revised form 15 July 2019 Accepted 27 July 2019 Keywords: Leptospirosis Diagnostic scoring Hypocalcemia Hypochloremia Eosinopenia
a b s t r a c t Background: Underestimation of leptospirosis cases is happening in many countries. The most common factor of underreporting is misdiagnosis. Considering the limitations of direct detection of pathogen and serological diagnosis for leptospirosis, clinical features and blood tests though non-specific are usually referred in making presumptive diagnosis to decide disease management. Methods: In this single-centre retrospective study, comparative analysis on clinical presentations and laboratory findings was performed between confirmed leptospirosis versus non-leptospirosis cases. Results: In multivariate logistic regression evidenced by a Hosmer-Lemeshow significance value of 0.979 and Nagelkerke R square of 0.426, the predictors of a leptospirosis case are hypocalcemia (calcium <2.10 mmol/L), hypochloremia (chloride <98 mmol/L), and eosinopenia (absolute eosinophil count <0.040 × 109 /L). The proposed diagnostic scoring model has a discriminatory power with area under the curve (AUC) 0.761 (p < 0.001). A score value of 6 reflected a sensitivity of 0.762, specificity of 0.655, a positive predictive value of 0.38, negative predictive value of 0.91, a positive likelihood ratios of 2.21, and a negative likelihood ratios of 0.36. Conclusion: With further validation in clinical settings, implementation of this diagnostic scoring model is helpful to manage presumed leptospirosis especially in the absence of leptospirosis confirmatory tests. © 2019 The Authors. Published by Elsevier Limited on behalf of King Saud Bin Abdulaziz University for Health Sciences. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).
Introduction A broad array of clinical manifestations has been documented for human leptospirosis [1,2]. Leptospiral infections may be asymptomatic, or with symptoms ranging from mild and self-limiting febrile illnesses to multi-organ dysfunction with fatal outcomes. Leptospirosis may mimic other febrile illnesses, which makes the clinical manifestations not fully reliable for the diagnosis of leptospirosis. Underreporting due to misdiagnosis of leptospirosis is therefore common. Reliance on laboratory diagnostic tests to differentiate leptospirosis from other diseases is not entirely practicable considering the limitations of currently available leptospirosis diagnostic tests.
∗ Corresponding author. E-mail address: [email protected] (Z. Sekawi).
The microscopic agglutination test (MAT) for detection of agglutinating antibodies [3] serves as the gold standard of serological diagnosis for leptospirosis [2]. MAT is not an ideal test for leptospirosis diagnosis especially during the bacteremia phase, and in regions where the diagnostic facilities are poorly accessible or not available. Although there are many other commercially available rapid serological tests, the problem of delayed diagnosis due to late sero-conversion and undetectable antibodies titre still exists. On the other hand, direct detection of leptospiral DNA by polymerase chain reaction (PCR) has been proposed to be used routinely in leptospirosis diagnosis as it overcomes the limitations of MAT especially during the early acute phase [4]. However, it may have more stringent requirements on sampling timing and type of specimen. This is because leptospiremia may end very soon after the onset of symptoms [5]. The success of molecular detection of leptospiral DNA depends not only on the bacterial load in clinical materials, but also on the specificity
https://doi.org/10.1016/j.jiph.2019.07.021 1876-0341/© 2019 The Authors. Published by Elsevier Limited on behalf of King Saud Bin Abdulaziz University for Health Sciences. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Please cite this article in press as: Fish-Low C-Y, et al. Hypocalcemia, hypochloremia, and eosinopenia as clinical predictors of leptospirosis: A retrospective study. J Infect Public Health (2019), https://doi.org/10.1016/j.jiph.2019.07.021
G Model JIPH-1151; No. of Pages 5
ARTICLE IN PRESS C.-Y. Fish-Low et al. / Journal of Infection and Public Health xxx (2019) xxx–xxx
2
and sensitivity of the primers or probes that are used in the detection. To address the non-availability of a rapid confirmatory diagnostic test for leptospirosis, several scoring models have been developed as guidelines in diagnosing presumptive leptospirosis. Faine’s criteria [6] were first developed based on clinical data (Part A), epidemiological factors (Part B), and bacteriological and laboratory findings based on only MAT titre (Part C). A score of 26 or more in Part A or in Part A + Part B, or 25 or more as the total of three parts may suggest a presumptive diagnosis. Faine’s criteria were then modified, hence known as modified Faine’s criteria [7] to include other formats of serological tests under Part C. The modified Faine’s criteria were later amended (known as modified Faine’s criteria with amendment) to improve the usefulness in clinical settings [8]. Despite several amendments of the guidelines, a scoring section based on blood test profile has never been incorporated. This study aimed to determine the clinical predictors for leptospirosis based on comparative analysis of blood test parameters between confirmed leptospirosis versus non-leptospirosis cases. Then a scoring model is developed for the determined predictors. Material and methods Ethics statement The study protocol was approved by the Institutional Review Board of Faculty of Medicine and Health Sciences, Universiti Putra Malaysia, and the Medical Research Ethics Committee (MREC), Ministry of Health Malaysia (NMRR-15-2148-27536). Study design and study subjects This study utilized a cross-sectional study design among clinically suspected leptospirosis patients [2] admitted in Hospital Serdang, Selangor, Malaysia. Since leptospirosis was gazetted as a notifiable disease in Malaysia in December 2010, patients admitted from 2011 to 2017 constituted the sampling frame. Patients in the paediatric age group (below 18 years old) and patients with past history of autoimmune diseases or any known comorbidities were excluded. The one proportion formula was used to calculate the minimum required sample size [9]. The expected prevalence (P) of leptospirosis (8.4%) was taken from a previous study in Malaysia [10]. The Z-statistic and precision level (d) were substituted with 1.96 and 0.05 respectively [11], and an adjustment for 90% eligibility to obtain a minimum sample size requirement of 133 respondents. Case notes of all suspected leptospirosis patients admitted between 2011 and 2017 were extracted from the hospital’s medical patients’ electronic database, using the ‘Filter’ function. The file serial numbers of these patients was then copied and transferred to Excel spreadsheet from where 137 out of these were randomly selected using the ‘Random’ function.
of total antibodies in MAT [2] using a 20-serovar panel, which consisted of Australis, Autumnalis, Bataviae, Canicola, Celledoni, Djasiman, Grippotyphosa, Hardjoprajitno, Icterohaemorrhagiae, Javanica, Patoc, Pomona, Pyrogenes, Tarassovi, IMR LEP/1, IMR LEP/115, IMR LEP/175, IMR LEP 803/11, IMR LEP/27, and IMR LEP/22. A confirmed leptospirosis case (CLC) is defined as single serum titre of 1:400 or at least a four-fold rise in titre of paired sera in the standard MAT, a positive culture of Leptospira in blood or urine sample, or positive detection of Leptospira DNA in PCR [12]. All cases of confirmed leptospirosis are tested negative in both aerobic and anaerobic blood culture, and negative in dengue IgM and NS1 detection. The cases otherwise are non-leptospirosis (NLC). Statistical analysis The data collected was analysed by using IBM Statistical Package for Social Sciences (SPSS) version 22. The laboratory findings were categorized as normal, low or high, based on the hospital’s pre-defined cut-off values. Comparative analysis between CLC (n = 30) and NLC (n = 107) was performed. Bivariate analysis using the Chi-square test was run on all the variables, to determine any difference between the two groups. Variables significant in the bivariate analysis along with other variables with p-value <0.25 were entered into a multivariate logistic regression using the Backward-LR method, and the best model was selected to obtain the predictors of leptospirosis infection. Development of diagnostic scoring model In line with a previous similar study [14], the beta coefficients of the predictor variables were used to develop a practical scoring system. All the coefficients were divided by the smallest coefficient, and then multiplied by 4, the results of which were rounded up to the nearest whole number. This scoring system was then used to generate the receiver operating characteristics (ROC) curve. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and likelihood ratios (LR) were also computed. Results A total of 137 subjects were included in the study, 72.3% of whom were males. Their ages ranged from 18 to 86 years, and a mean (SD) age of 35.5 (14.8) years. Their ethnicities were: Malay (68.6%), Indian (13.1%), Chinese (6.6%) and others (11.7%). Around one-fifth (21.9%) had a diagnosis of confirmed leptospirosis. Out of the 30 cases of confirmed leptospirosis, 53.3% (n = 16) gave positive PCR result, 36.7% (n = 11) was positive by MAT, 3.3% (n = 1) was positive in blood cultures, and 6.7% (n = 2) of the cases were positive in both urine cultures and MAT. Bivariate analysis
Data collection and case definition A structured proforma was used to collect data from the selected case notes. Data on clinical presentations were collected on a binary scale (Yes or No), while the values for the laboratory findings were collected in their raw numerical forms as continuous variables. The cases were categorized into confirmed leptospirosis versus non-leptospirosis. The confirmatory diagnosis of leptospirosis was performed by the Institute for Medical Research (IMR), Ministry of Health Malaysia [12]. The tests performed were pathogen isolation [2], detection of leptospiral DNA by real-time polymerase chain reaction [13], and and detection
Table 1 shows the comparison of clinical and laboratory characteristics of CLC and NLC. There was a significant difference between the two groups in their proportions of temperature above 39 ◦ C, low calcium (<2.10 mmol/L), high urea (>9.2 mmol/L), low sodium (<136 mmol/L), low chloride (<98 mmol/L), and low absolute eosinophil count (<0.040 × 109 /L). Multivariate analysis For the multivariate logistic regression, the model fitted the sample, evidenced by a Hosmer-Lemeshow significance
Please cite this article in press as: Fish-Low C-Y, et al. Hypocalcemia, hypochloremia, and eosinopenia as clinical predictors of leptospirosis: A retrospective study. J Infect Public Health (2019), https://doi.org/10.1016/j.jiph.2019.07.021
G Model
ARTICLE IN PRESS
JIPH-1151; No. of Pages 5
C.-Y. Fish-Low et al. / Journal of Infection and Public Health xxx (2019) xxx–xxx
3
Table 1 Comparison of clinical and laboratory features of confirmed (CLC) and non-confirmed leptospirosis cases (NLC). NLC (n = 107)
Characteristic Headache Fever Temperature >39◦ C Conjunctival suffusion Myalgia Jaundice Haemoptysis / shortness of breath Skin rash Gastrointestinal symptoms Haemorrhages Creatinine kinase <29 U/L Creatinine kinase >200 U/L C-reactive protein >5 mg/L Calcium <2.10 mmol/L Calcium >2.55 mmol/L Magnesium <0.85 mmol/L Magnesium >1.05 mmol/L Albumin <35 g/L Albumin >50 g/L Bilirubin, total >20.5 mol/L Alkaline phosphatase <40 U/L Alkaline phosphatase >150 U/L Aspartate transferase >34 U/L Urea <2.5 mmol/L Urea >9.2 mmol/L Sodium <136 mmol/L Potassium <3.5 mmol/L Potassium >5.1 mmol/L Chloride <98 mmol/L Chloride >107 mmol/L Creatinine <53 mol/L Creatinine >115 mol/L Red blood cell <3.80 × 1012 /L Red blood cell >5.80 × 1012 /L Haemoglobin <11.5 g/dL Haemoglobin >18.0 g/dL Platelet <150 × 109 /L Platelet >400 × 109 /L White blood cell <4.0 × 109 /L White blood cell >11.0 × 109 /L Absolute neutrophil <2.00 × 109 /L Absolute neutrophil >7.50 × 109 /L Absolute lymphocyte <1.50 × 109 /L Absolute lymphocyte >4.00 × 109 /L Absolute monocyte <0.200 × 109 /L Absolute monocyte >0.800 × 109 /L Absolute eosinophil <0.040 × 109 /L Absolute eosinophil >0.400 × 109 /L *
CLC (n = 30)
p-value
Frequency
(%)
Frequency
(%)
48 99 17 14 72 11 14 15 83 6 4 63 66 43 1 51 8 57 1 32 2 14 83 8 17 66 23 5 26 6 1 38 6 19 13 3 24 6 11 33 7 42 65 1 9 43 24 9
(44.9) (92.5) (15.9) (13.1) (67.3) (10.3) (13.1) (14.0) (77.6) (5.6) (3.7) (58.9) (81.5) (62.3) (1.4) (72.9) (11.4) (54.3) (1.0) (30.2) (1.9) (13.1) (77.6) (7.5) (15.9) (61.7) (21.5) (4.7) (24.3) (5.6) (0.9) (35.8) (5.6) (17.8) (12.1) (2.8) (22.4) (5.6) (10.3) (30.8) (6.5) (39.9) (60.7) (0.9) (8.4) (40.2) (22.4) (8.4)
16 28 10 3 20 3 4 4 27 2 0 19 18 20 0 15 2 18 0 10 1 4 23 2 12 26 8 0 16 1 0 15 1 7 1 1 9 0 5 10 5 12 23 0 1 10 14 1
(53.3) (93.3) (33.3) (10.0) (66.7) (10.0) (13.3) (13.3) (90.0) (6.7) (0.0) (63.3) (90.0) (90.9) (0.0) (68.2) (9.1) (60.0) (0.0) (33.3) (3.3) (13.3) (76.7) (6.7) (40.0) (86.7) (26.7) (0.0) (53.3) (3.3) (0.0) (50.0) (3.3) (23.3) (3.3) (3.3) (30.0) (0.0) (16.7) (33.3) (16.7) (40.0) (76.7) (0.0) (3.3) (33.3) (46.7) (3.3)
0.411 0.880 0.034* 0.651 0.949 0.964 0.972 0.924 0.130 0.827 0.657 0.830 0.362 0.014* 1.000 0.476 0.529 0.614 0.614 0.742 0.526 1.000 0.917 0.767 0.004* 0.010* 0.640 0.364 0.004* 0.972 1.000 0.172 0.664 0.527 0.160 1.000 0.490 0.204 0.285 0.601 0.069 0.588 0.069 0.588 0.439 0.367 0.014* 0.577
Statistically significant at p < 0.05.
Table 2 Predictors of presumptive leptospirosis. Factors
B
SE
Wald
df
p-value
Adjusted OR
95% CI
Calcium Normal Low (<2.10 mmol/L)
– 2.04
– 1.00
– 4.17
– 1
– 0.041
1.00 7.72
– 1.09–54.98
Chloride Normal Low (<98 mmol/L)
– 1.70
– 0.77
– 4.82
– 1
– 0.028
1.00 5.45
– 1.20–24.79
Absolute eosinophil Normal Low (<0.040 × 109 /L)
– 1.76
– 0.73
– 5.75
– 1
– 0.017
1.00 5.80
– 1.38–24.41
B = beta coefficient; SE = standard error; df = degrees of freedom; OR = odd ratio; CI = confidence interval.
value of 0.979. The Nagelkerke R square was 0.426, and the predictors of being a case of leptospirosis are presented in Table 2. Table 3 illustrates the proposed score for diagnosis of leptospirosis, based on the multivariate analysis results.
Fig. 1 shows the receiver operating characteristics (ROC) curve to differentiate between leptospirosis and non-leptospirosis. The area under the curve (AUC) was 0.761 (p < 0.001). As presented in Table 4, score value of 6 reflected a sensitivity of 0.762, specificity of 0.655, a positive predictive value (PPV) of 0.38, negative predictive
Please cite this article in press as: Fish-Low C-Y, et al. Hypocalcemia, hypochloremia, and eosinopenia as clinical predictors of leptospirosis: A retrospective study. J Infect Public Health (2019), https://doi.org/10.1016/j.jiph.2019.07.021
G Model
ARTICLE IN PRESS
JIPH-1151; No. of Pages 5
C.-Y. Fish-Low et al. / Journal of Infection and Public Health xxx (2019) xxx–xxx
4
Table 3 Proposed score for diagnosis of leptospirosis, based on multivariate analysis. Factor
Beta coefficient
Calculated score
Score (rounded to nearest integer)
Calcium <2.10 mmol/L Chloride <98 mmol/L Eosinophil <0.040 × 109 /L
2.044 1.696 1.758
4.80 4.00 4.15
5 4 4
Maximum possible score
13
value (NPV) of 0.91, a positive likelihood ratios (PLR) of 2.21, and a negative likelihood ratios (NLR) of 0.36. Discussion As observed in the present study and other similar studies that compared serum biochemistry between leptospirosis and non-leptospirosis patients, the variables proposed in the respective diagnostic score models may be cohort- or region-specific. A Sri Lankan cohort-based study [14] reveals high creatinine (>150 mol/L), high bilirubin (>30 mol/L), neutrophil >80%, low platelet count (<85,000/mm3 ), and exposure to possible source of leptospirosis are positively associated with the confirmed diagnosis. A recent Thai cohort-based study [15] reported the presence of hypotension, jaundice, muscle pain, acute kidney failure, low haemoglobin (<12 g/dL), and hypokalemia (potassium <3.5 mEq/L) with hyponatremia (sodium <135 mEq/L) are significant predictors for leptospirosis. On the other hand, hypocalcemia (calcium <2.10 mmol/L), hypochloremia (chloride <98 mmol/L), and eosinopenia (<0.040 × 109 /L) are proposed as the scoring criteria for leptospirosis in this Malaysian cohort-based study. Despite none of the predictors in any of the abovementioned studies are present in another study, abnormal renal panel is common across these studies. Perturbations of electrolytes: chloride, potassium, sodium, and creatinine (waste product), and calcium (minerals) may serve as indications of renal involvement during leptospirosis. Hypokalemia is regularly seen in leptospirosis [16,17] whereas hyponatremia resolves quickly as soon as the infection is controlled [18]. Leptospirosis-induced acute kidney injury (AKI) causes rapid increment of serum creatinine and blood urea nitrogen [19]. Though not proven in the context of leptospirosis, dyschloremia is proposed as a risk factor for development of AKI, and hypochloremic patients are particularly having higher mortality rate [20]. Mild hypocalcemia usually develops asymptomatically and unnoticed or otherwise may show cardiac arrest, seizures, numbness, and confusion [21,22]. It can be caused by kidney failure, rhabdomyolysis, overdose of calcium channel blocker, and medications [21]. Hypocalcemia in leptospirosis is recognized as one of the reasons leading to arrhythmias [23,24]. The biphasic calcium kinetic in leptospirosis that was reported by Zakout R et al. [23] is found consistent with that seen in rhabdomyolysis-induced AKI [25], which hypocalcemia occurred in oliguric phase and then hypercalcemia in diuretic phase during recovery. This sequential
Fig. 1. Coordinates of the ROC curve for outcome—positive leptospirosis.
observation may be explained by calcium deposition in damaged muscle followed by resolving calcification once the renal functions have recovered [26]. Phenomenon of low eosinophil count in leptospirosis is either has not been evaluated or lacks of emphasis in previous studies. It has once been reported in canine leptospirosis [27]. Eosinopenia is evidenced to be associated with peritonitis [28], sepsis [29], and enteric fever [30]. Bass DA et al. [31] speculated that the migration of eosinophils to the inflammatory site, which was triggered by chemotactic factors during acute inflammation has resulted in eosinopenia. Further analysis on renal profile of leptospirosis patients would be beneficial and have diagnostic values. Kidney is the known target organ where leptospires may colonize and persist [32,33], possibly related to their ability of forming biofilm [34]. In addition, leptospiral outer membrane proteins have been demonstrated to induce renal inflammation and interstitial nephritis [35]. Collectively, bacterial invasion, inflammation, haemodynamic perturbations and, pathogenicity of bacterial products contribute to the development of leptospiral nephropathy [19]. Varying clinical spectrum of renal involvement are documented in leptospirosis, ranging from mild course of proteinuria [36], serum electrolyte changes [16] to severe acute renal failure associated with rhabdomyolysis [37]. Conclusion Confirmatory diagnosis of leptospirosis is challenging and currently limited by many well recognized factors. A scoring model
Table 4 Diagnostic model performance (sensitivity, specificity, positive and negative predictive value and likelihood ratios). Score
Sensitivity
Specificity
PPV
NPV
PLR
NLR
4 6 10
0.810 0.762 0.238
0.483 0.655 0.966
0.29 0.38 0.38
0.96 0.91 0.89
1.45 2.21 2.18
0.14 0.36 0.43
PPV = positive predictive value; NPV = negative predictive value; PLR = positive likelihood ratios; NLR = negative likelihood ratios.
Please cite this article in press as: Fish-Low C-Y, et al. Hypocalcemia, hypochloremia, and eosinopenia as clinical predictors of leptospirosis: A retrospective study. J Infect Public Health (2019), https://doi.org/10.1016/j.jiph.2019.07.021
G Model JIPH-1151; No. of Pages 5
ARTICLE IN PRESS C.-Y. Fish-Low et al. / Journal of Infection and Public Health xxx (2019) xxx–xxx
based on clinical predictors of blood test parameter, particularly the renal profile will efficiently improve leptospirosis diagnosis. It may be of great diagnostic potential if the proposed scoring model to be combined with both epidemiological risk factors and clinical manifestations. Case history such as contact with possible infected animals or exposure to leptospire-containing water is especially useful in the diagnosis of leptospirosis as leptospires are mostly transmitted via direct or indirect contact with infected animals. We acknowledge the limitations of this study as the enrolled cases may not be representative of all leptospirosis cases in Malaysia. Until the proposed scoring model is validated in larger cohort size and involves multicentre, it shows potential to be incorporated into the modified Faine’s criteria with amendment, or can be applied as a standalone diagnostic guideline for leptospirosis. Funding This work was funded by the Ministry of Higher Education, Malaysia Long Term Research Grant Scheme (LRGS) [UPM/7002/1/LRGS/5526400]. Competing interest None declared. Ethical approval Not required. Acknowledgements We would like to thank the Director General of Health Malaysia for his permission to publish this article. The authors are indebted to all the clinicians and supporting staff from Hospital Serdang, Malaysia. The authors also express their appreciation to all the members of Malaysia Leptospirosis Research Network (MyLepto) and to all the study participants. References [1] Centers for Disease Control and Prevention. Infectious diseases related to travel. New York: Oxford University Press; 2017. [2] World Health Organization. Human leptospirosis: guidance for diagnosis, surveillance and control. World Health Organization; 2003, http://dx.doi.org/ 10.1590/S0036-46652003000500015. [3] Turner LH. Leptospirosis II: serology. Trans R Soc Trop Med Hyg 1968;62:880–99, http://dx.doi.org/10.1016/0035-9203(68)90017-5. [4] Mullan S, Panwala TH. Polymerase chain reaction: an important tool for early diagnosis of leptospirosis cases. J Clin Diagn Res 2016;10:DC08–11, http://dx. doi.org/10.7860/JCDR/2016/22462.9010. [5] Turner LH, Leptospirosis I. Trans R Soc Trop Med Hyg 1967;61:842–55, http:// dx.doi.org/10.1016/0035-9203(67)90045-4. [6] Faine S. Guidelines for the control of leptospirosis. Geneva: WHO offset publication; 1982, no. 67. [7] Shivakumar S, Shareek PS. Diagnosis of leptospirosis utilizing modified Faine’s criteria. J Assoc Phys India 2004;52:678–9. [8] Shivakumar S. Indian guidelines for the diagnosis and management of human leptospirosis. Med Update 2013;23:23–9. [9] Daniel WW. Biostatistics: a foundation for analysis in the health sciences. 7th ed. New York: Wiley; 1999. [10] Rafizah AAN, Aziah BD, Azwany YN, Imran MK, Rusli AM, Nazri SM, et al. A hospital-based study on seroprevalence of leptospirosis among febrile cases in northeastern Malaysia. Int J Infect Dis 2013;17:e394–7, http://dx.doi.org/10. 1016/j.ijid.2012.12.012. [11] Naing L, Winn T, Rusli BN. Practical issues in calculating the sample size for prevalence studies. Arch Orofac Sci 2006;1:9–14. [12] Ministry of Health Malaysia. Guidelines for diagnosis management, prevention and control of leptospirosis. 1st ed. BKPKKM; 2011. [13] Ahmed A, Engelberts MFM, Boer KR, Ahmed N, Hartskeerl RA. Development and validation of a real-time PCR for detection of pathogenic Leptospira species in clinical materials. PLoS One 2009;4:e7093, http://dx.doi.org/10.1371/journal. pone.0007093.
5
[14] Rajapakse S, Weeratunga P, Niloofa R, Fernando N, de Silva NL, Rodrigo C, et al. A diagnostic scoring model for leptospirosis in resource limited settings. PLoS Negl Trop Dis 2016;10:e0004513, http://dx.doi.org/10.1371/journal.pntd. 0004513. [15] Sukmark T, Lumlertgul N, Peerapornratana S, Khositrangsikun K, Tungsanga K, Sitprija V, et al. Thai-Lepto-on-admission probability (THAI-LEPTO) score as an early tool for initial diagnosis of leptospirosis: result from Thai-Lepto AKI study group. PLoS Negl Trop Dis 2018;12:e0006319, http://dx.doi.org/10.1371/ journal.pntd.0006319. [16] Seguro AC, Lomar AV, Rocha AS. Acute renal failure of leptospirosis: nonoliguric and hypokalemic forms. Nephron 1990;55(146), http://dx.doi.org/10. 1159/000185943. [17] Abdulkader RCRM, Seguro AC, Malheiro PS, Burdmann EA, Marcondes M. Peculiar electrolytic and hormonal abnormalities in acute renal failure due to leptospirosis. Am J Trop Med Hyg 1996;54(1), http://dx.doi.org/10.4269/ajtmh. 1996.54.1. [18] Sitprija V. Altered fluid, electrolyte and mineral status in tropical disease, with an emphasis on malaria and leptospirosis. Nat Clin Pract Nephrol 2008;4(91), http://dx.doi.org/10.1038/ncpneph0695. [19] Visith S, Kearkiat P. Nephropathy in leptospirosis. J Postgrad Med Educ Res 2005;51:184–8. [20] Shao M, Li G, Sarvottam K, Wang S, Thongprayoon C, Dong Y, et al. Dyschloremia is a risk factor for the development of acute kidney injury in critically ill patients. PLoS One 2016;11:e0160322, http://dx.doi.org/10.1371/journal.pone. 0160322. [21] Soar J, Perkins GD, Abbas G, Alfonzo A, Barelli A, Bierens JJLM, et al. European resuscitation council guidelines for resuscitation 2010 section 8. Cardiac arrest in special circumstances: electrolyte abnormalities, poisoning, drowning, accidental hypothermia, hyperthermia, asthma, anaphylaxis, cardiac surgery, trauma, pregna. Resuscitation 2010;81:1400–33, http://dx.doi.org/10.1016/j. resuscitation.2010.08.015. [22] Fong J, Khan A. Hypocalcemia: updates in diagnosis and management for primary care. Can Fam Physician 2012;58:158–62. [23] Zakout R, do Carmo M, Freitas P. Reversible myocardial calcification following severe leptospirosis complicated with rhabdomyolysis-induced acute kidney injury and magnesium-wasting nephropathy. J Med Cases 2013;4:92–8. [24] de Sousa Soares D, Galdino GS, Rodrigues BC, Silva Jr G, Daher EF. Arrhythmias in leptospirosis-associated acute kidney injury: a case series. Braz J Infect Dis 2017;21(209), http://dx.doi.org/10.1016/j.bjid.2016.11.010. [25] Pakkyara A, Mohsin N, Budruddin M, Mohammad AJE, Mohammad K. Calcium kinetic in a patient with acute renal failure due to rhabdomyolysis. A case report and review of literature. Oman Med J 2010;25, http://dx.doi.org/10.5001/omj. 2010.93, e013–e013. [26] Akmal M, Goldstein DA, Telfer N, Wilkinson E, Massry SG. Resolution of muscle calcification in rhabdomyolysis and acute renal failure. Ann Intern Med 1978;89:928–30, http://dx.doi.org/10.7326/0003-4819-89-6-928. [27] Rentko VT, Clark N, Ross LA, Schelling SH. Canine leptospirosis: a retrospective study of 17 cases. J Vet Intern Med 1992;6:235–44, http://dx.doi.org/10.1111/ j.1939-1676.1992.tb00345.x. [28] Jagdeesh TS, Mishra A, Saxena A, Sharma D. Eosinopenia as a prognostic marker in patients with peritonitis. ISRN Infect Dis 2013;2013:8, http://dx.doi.org/10. 5402/2013/540948. [29] Lavoignet CE, Le Borgne P, Slimani H, Forato M, Kam C, Kauffmann P, et al. Relevance of eosinopenia as marker of sepsis in the emergency department. La Rev Médecine Interne 2016;37(730), http://dx.doi.org/10.1016/j.revmed.2016. 02.018. [30] Matono T, Kutsuna S, Kato Y, Katanami Y, Yamamoto K, Takeshita N, et al. Role of classic signs as diagnostic predictors for enteric fever among returned travellers: relative bradycardia and eosinopenia. PLoS One 2017;12:e0179814, http://dx.doi.org/10.1371/journal.pone.0179814. [31] Bass DA, Gonwa TA, Szejda P, Cousart MS, DeChatelet LR, McCall CE. Eosinopenia of acute infection: production of eosinopenia by chemotactic factors of acute inflammation. J Clin Invest 1980;65:1265–71, http://dx.doi.org/10.1172/ JCI109789. [32] Athanazio DA, Silva EF, Santos CS, Rocha GM, Vannier-Santos MA, McBride AJA, et al. Rattus norvegicus as a model for persistent renal colonization by pathogenic Leptospira interrogans. Acta Trop 2008;105(176), http://dx.doi.org/ 10.1016/j.actatropica.2007.10.012. [33] Ratet G, Veyrier FJ, Fanton d’Andon M, Kammerscheit X, Nicola M-A, Picardeau M, et al. Live imaging of bioluminescent Leptospira interrogans in mice reveals renal colonization as a stealth escape from the blood defenses and antibiotics. PLoS Negl Trop Dis 2014;8:e3359, http://dx.doi.org/10.1371/journal. pntd.0003359. [34] Ristow P, Bourhy P, Kerneis S, Schmitt C, Prevost M-C, Lilenbaum W, et al. Biofilm formation by saprophytic and pathogenic leptospires. Microbiology 2008;154:1309–17, http://dx.doi.org/10.1099/mic.0.2007/014746-0. [35] Yang CW, Wu MS, Pan MJ. Leptospirosis renal disease. Nephrol Dial Transplant 2001;16(Suppl):73–7. [36] Thresiamma KC, Asha B, Reetika C, Manjula S, Chithra J, Sunil T, et al. Proteinuria in early detection of human leptospirosis. Int J Res Med Sci 2017;5:646–52, http://dx.doi.org/10.18203/2320-6012.ijrms20170168. [37] Droulias J, Moutzouris D, Kassimatis T, Kollia K, Apostolou T, Hadjikonstantinou V. Rhabdomyolysis-induced acute renal failure in a patient with leptospirosis. Saudi J Kidney Dis Transpl 2007;18:430–1.
Please cite this article in press as: Fish-Low C-Y, et al. Hypocalcemia, hypochloremia, and eosinopenia as clinical predictors of leptospirosis: A retrospective study. J Infect Public Health (2019), https://doi.org/10.1016/j.jiph.2019.07.021