Abstract
Objectives
Acute pancreatitis (AP) is a severe complication of leptospirosis. This review focuses on the current evidence of AP in patients with leptospirosis.
Methods
Data on clinical characteristics, biochemical parameters, diagnosis, complications, critical care, fluid management, operative management, and outcomes were analyzed. This study was registered in PROSPERO (CRD42022360802).
Results
We included 35 individual case reports and 4 case series involving 79 patients. Sex was reported for 48 (60.7%) patients; 38 (48.1%) were male and 10 (12.6%) were female. The patients’ mean age was 45.13 (15–83 years). Acute kidney injury, thrombocytopenia, hypotension, and liver injury were the most common complications reported. Complete recovery was reported for 36 (45.5%) patients. Biochemical and radiological recovery was reported for 10 (12.6%) and 9 (11.3%) patients, respectively. Death was reported in 18 (22.7%) patients.
Conclusion
A high degree of clinical suspicion and different modalities of investigations are essential in the diagnosis of AP in leptospirosis. AP can be easily missed in leptospirosis because both conditions share similar clinical presentations and complications. Because of the high prevalence of acute kidney injury, judicious fluid management and close monitoring are mandatory.
Keywords
Introduction
Leptospirosis is a common zoonotic illness caused by Leptospira icterohaemorrhagiae and other subspecies. 1 It has a significant impact on health expenditures in developing countries. 2 Leptospirosis has a wide spectrum of presentations ranging from asymptomatic self-limiting illness to fatal multiorgan involvement. 3 The incubation period of leptospirosis is approximately 1 to 2 weeks. 4 The bacterium is present in water contaminated by excreta of chronically infected rodents, and it enters the human body through mucous membrane and skin aberrations. 5 People at risk include farmers and slum dwellers. 6 Complications are mainly caused by vasculitis involving medium and small vessels. 1 Widespread activation and damage of endothelial cells in these vessels occurs by either direct injury by the microorganism or immune-mediated destruction following cytokine storm. 7 Leptospirosis often occurs in tropical and subtropical countries because the climate favors the transmission. 7 Because most tropical regions contain developing countries, leptospirosis continues to have a negative impact on the healthcare systems in these countries with each outbreak that occurs. 4
The mild form of leptospirosis has two distinctive phases: the septicemic and immune phases. 8 The acute septicemic phase lasts for 5 to 7 days and is followed by transient symptomatic improvement. 7 The disease then either progresses to the severe form or regresses to an asymptomatic illness. 7 The severity of leptospirosis is not predictable at symptom onset 7 ; it is influenced by host-related factors and the pathogenicity of the microorganism. 7 Weil syndrome is a severe form leptospirosis manifesting as renal failure, hepatic dysfunction, pulmonary hemorrhage, and multiple hemorrhagic diathesis.9–12
The mortality rate is higher in patients with Weil syndrome, reaching 10% to 15%, 12 and exceeds 50% when severe pulmonary hemorrhagic syndrome occurs.5,13 Acute pancreatitis (AP) is an uncommon but known complication of leptospirosis. 12 However, very few studies have focused on the incidence, pathogenesis, and risk factors of AP in leptospirosis. AP has a wide spectrum of presentations ranging from mild symptomatic illness to severe hemorrhagic pancreatitis causing peripancreatic necrotic collections (PPNC) and fatal multiorgan failure (MOF). 6
Diagnosis of AP in patients with leptospirosis is important because both AP and leptospirosis can lead to MOF, and early diagnosis with prompt management may reduce the risk of mortality. 14 The diagnosis of AP is reached after correlating the biochemical and radiological findings with the clinical presentation (CP). 10 Serological testing of leptospirosis is also important to confirm the etiology of AP. 15 The modified Atlanta criteria are used to diagnose AP, and two of the following three criteria must be fulfilled 16 : clinical features including abdominal pain, nausea, and vomiting; biochemical features including a serum lipase or amylase concentration higher than three times the upper limit of normal; and characteristic imaging findings on computed tomography (CT) or magnetic resonance imaging. 16
The severity of AP is classified as mild, moderate, or severe based on local and systemic complications. 17 Various scoring systems are available to classify the severity of AP. 16 The revised Atlanta classification is one such system that is widely accepted for defining the severity of AP. 16 Persistent organ failure is the hallmark of severe AP and is associated with an increased mortality rate (20%) and admission to the intensive care unit (ICU). 18
It should be noted that clinical manifestations are not reliable because both AP and leptospirosis have common symptoms and signs. 19 Early identification and treatment, including both operative and nonoperative measures, are important to minimize complications. 20 A synopsis of existing evidence on leptospirosis complicated by AP is needed. The primary objective of this review was to systematically examine the clinical characteristics, biochemistry, imaging, complications, management, and outcomes of AP in the setting of leptospirosis.
Methods
Search strategy
We systematically searched the electronic databases of PubMed, Scopus, EMBASE, Cochrane CENTRAL, and Latin American and Caribbean Health Sciences Literature (LILACS) from database inception to December 2021 with no language restriction. Key words related to AP and its complications as well as leptospirosis were searched in the title and abstract fields. The detailed search strategy is shown in the supplementary file. Additionally, the reference list of each eligible article was manually searched to identify more publications (Figure 1). This study was registered in PROSPERO (Number: CRD42022360802).

PRISMA flow chart.
Eligibility criteria and screening of articles
All types of observational studies (e.g., cohort studies, case-control studies, descriptive cross-sectional studies, case reports, and series) were included in the review. Articles were screened using three key criteria:
The article must describe AP as a complication of leptospirosis. The article must be based on primary data of actual patients. The article must include an interim or full analysis and not be restricted to a description of a protocol.
In the first round, two investigators independently performed the initial screening based on the titles and abstracts. In the second round, the full texts of relevant records were assessed based on the eligibility criteria. In doubtful situations, a consensus was reached after discussion with a third investigator.
Data extraction
Two reviewers independently performed the data extraction using a predesigned template. All data pertaining to the CP, investigations, treatment, and outcomes were extracted, categorized, and tabulated. The extracted data were cross checked for any discrepancy by a third reviewer.
Data analysis
The risk-of-bias assessment of eligible studies was performed using the Downs and Black checklist, and the findings are shown in Table S1. A narrative synthesis was performed with the available data. A meta-analysis of the quantitative parameters (e.g., outcome) was not performed because of the heterogeneity in the reporting and the limited number of studies.
Data analysis
Ethics approval and consent to participate are not applicable because of the nature of this study (systematic review with no patient involvement).
Results
Clinical characteristics
The screening process resulted in the inclusion of 35 individual case reports and 4 case series involving a total of 79 patients. Sex was reported for 48 (60.7%) patients and was not available for 31 (39.2%) patients. Of the 48 patients with available data on sex, 38 (48.1%) were male and 10 (12.6%) were female. The patients’ mean age was 45.13 (range, 15–83) years.
Abdominal pain (n = 51, 64.5%), fever (n = 45, 56.9%), vomiting (n = 30, 37.9%), and oliguria (n = 11, 13.9%) were the most common symptoms. Other manifestations included bleeding (n = 12, 15.1%), arthralgia (n = 8, 10.1%), lethargy (n = 8, 10.1%), diarrhea (n = 6, 7.5%), and chills and rigors (n = 5, 6.3%). Headache (n = 4, 5.0%), dyspnea (n = 3, 3.7%), occipital headache (n = 2, 2.5%), cough (n = 2, 2.5%), and back pain (n = 2, 2.5%) were also reported as common symptoms. Icterus (n = 34, 43.0%), tachycardia (n = 19, 24.0%), and hypotension (n = 17, 21.5%) were the most common examination findings described. Other clinical signs were abdominal tenderness (n = 15, 18.9%), conjunctival suffusion (n = 13, 16.4%), tachypnea (n = 10, 12.6%), and basal crepitation (n = 9, 11.3%) (Table 1). 3 ,6,8,11–15,17,19,21–47
General and clinical characteristics of patients with leptospirosis and acute pancreatitis.
SN, serial number; M, male; F, female; NA, not available; MA, mean age; CP, clinical presentation; BP, blood pressure; PR, pulse rate; RR, respiratory rate; CT, computed tomography; USS, ultrasound scan; GCS, Glasgow Coma Scale.
Biochemical and electrocardiographic findings
Among all 79 patients, hyperamylasemia (n = 56, 70.8%), leukocytosis (n = 38, 48.1%), and thrombocytopenia (n = 31, 39.2%) were the most common findings. Other common findings were hyperbilirubinemia (n = 20, 25.3%), altered liver enzymes (n = 23, 29.1%), hyperlipasemia (n = 20, 25.3%), and anemia (n = 10, 12.6%). Metabolic acidosis (n = 9, 11.3%), hypocalcemia (n = 9, 11.3%), microscopic hematuria (n = 6, 7.5%), an increased alkaline phosphatase concentration (n = 8, 10.1%), and proteinuria (n = 6, 7.5%) were the remaining findings. Electrocardiographic findings were available for seven (8.8%) patients. The most common electrocardiographic findings noted were atrial fibrillation (n = 3, 3.7%), bradycardia (n = 3, 3.7%), and atrioventricular block (n = 1, 1.2%) (Table 2). 3 ,6,8,11–15,17,19,21–47
Biochemical findings of patients with leptospirosis and acute pancreatitis
SN, serial number; WBC, white blood cell count; PCV, packed cell volume; NA, not available; Hb, hemoglobin; CRP, C-reactive protein; AST, aspartate transaminase; ALT, alanine transaminase; TBr, total bilirubin; SCr, serum creatinine.
Radiological findings
At least one modality of imaging was reported for 48 (60.7%) patients. An abdominal ultrasound scan (USS), abdominal contrast-enhanced CT (CECT), chest X-ray (CXR), and abdominal X-ray (AXR) were used as imaging modalities. USS was performed in 41 (51.8%) patients. Abnormalities consistent with pancreatitis (n = 19, 24.0%) and ascites (n = 4, 5.0%) were the most common findings. CECT findings were reported for 21 (26.5%) patients.
The most common findings in CECT were a large bulky pancreas (n = 5, 6.3%), peripancreatic fat stranding (n = 5, 6.3%), an edematous pancreatitis (n = 5, 6.3%), hepatosplenomegaly (n = 4, 5.0%), and PPNC (n = 4, 5.0%). The remaining findings were ascites (n = 3, 3.7%), bilateral pleural effusion (n = 2, 2.5%), and isolated hepatomegaly (n = 1, 1.2%). CXR was performed examine the concomitant lung involvement in six (7.5%) patients, and diffuse pulmonary hemorrhage (n = 1, 1.2%) was the most common finding. AXR was performed in five (6.3%) patients, and the presence of dilated intestinal loops (n = 2, 2.5%) was the most common finding (Table 3). 3 ,6,8,11–15,17,19,21–47
Microbiological and radiological investigations of patients with leptospirosis and acute pancreatitis
Lepto IgM, Leptospira immunoglobulin M; Lepto IgG, Leptospira immunoglobulin G; SN, serial number; AP, acute pancreatitis; ARDS, acute respiratory distress syndrome; CKD, chronic kidney disease; CLCD, chronic liver cell disease; DFM, dark field microscopy; PCR, polymerase chain reaction; MAT, microscopic agglutination test; CXR, chest X-ray; AXR, abdominal X-ray; USS, ultrasound scan; GB, gallbladder; NA, not available; CECT, contrast-enhanced computed tomography.
Diagnosis of AP and leptospirosis
The serum concentration of either amylase or lipase was reported for 43 (54.4%) patients. Only the enzyme concentrations and CP were reported for 18 (22.7%) patients. Amylase with CP (n = 3, 3.7%), lipase with CP (n = 4, 5.0%), and both amylase and lipase with CP (n = 9, 11.3%) were reported. Radiological findings were used in the diagnosis of AP in 27 (34.1%) patients. USS, amylase, and CP were reported for 15 (18.9%) patients. Only CECT with CP was reported for six (7.5%) patients. CECT findings, both amylase and lipase, and CP were reported for five (6.3%) patients. CECT findings, amylase, and CP were reported for three (3.7%) patients. CECT findings, lipase, and CP were reported for one (1.2%) patient.
At least one type of leptospirosis diagnostic workup was performed in 54 (68.1%) patients. Measurement of immunoglobulin M (IgM) (n = 39, 49.3%) and performance of a microscopic agglutination test (MAT) (n = 18, 22.7%) were the most common diagnostic modalities. Urine polymerase chain reaction (n = 5, 6.3%), blood polymerase chain reaction (n = 4, 5.0%), dark field microscopy (n = 2, 2.5%), and measurement of immunoglobulin G (n = 3, 3.7%) were the other reported modalities.
Various serovars of Leptospira species were mentioned. Leptospira interrogans Icterohaemorrhagiae (n = 7, 8.8%) and L. interrogans unspecified (n = 5, 6.3%) were the most common types mentioned. Leptospira interrogans Autumnalis and (n = 3, 3.7%) L. interrogans Bratislava (n = 3, 3.7%) were the other common strains reported.
Local and systemic complications
Hemorrhagic pancreatitis (n = 2, 2.5%) and necrotizing pancreatitis (n = 2, 2.5%) were the most common local complications noted. Others included extensive PPNC (n = 1, 1.2%), paralytic ileus (n = 1, 1.2%), and acalculous cholecystitis (n = 4, 5.0%). Acute kidney injury (AKI) (n = 55, 69.6%), thrombocytopenia (n = 32, 40.0%), hypotension (n = 19, 24.0%), liver injury (n = 13, 14.4%), acidosis (n = 10, 12.6%), cardiac involvement (n = 10, 12.6%), and sepsis (n = 8, 10.1%) were the most common complications reported. MOF (n = 6, 7.5%) and acute respiratory distress syndrome (n = 4, 5.0%) were other common complications. Atrial fibrillation (n = 4, 5.0%), bradycardia (n = 4, 5.0%), and meningitis (n = 2, 2.5%) were other notable complications (Table 4). 3 ,6,8,11–15,17,19,21–47
Complications and management of patients with leptospirosis and acute pancreatitis.
SN, serial number; NA, not applicable; AKI, acute kidney injury; ARDS, acute respiratory distress syndrome; AV block, atrioventricular block; ICU, intensive care unit; IV, intravenous; CVA, cerebrovascular accident; CVP, central venous pressure; UOP, urine output.
Management and outcomes of AP in patients with leptospirosis
Admission to the ICU was needed for 17 (21.5%) patients. Noninvasive ventilation and intubation were performed in 4 (5.0%) and 11 (13.9%) patients, respectively. Ionotropic support was needed for four (5.0%) patients. Among patients with AKI, hemodialysis (HD) was needed for 20 (25.3%) patients. Urine output and central venous pressure were used for objective assessment of the fluid status in four (5.0%) and one (1.2%) patient, respectively. A judicious fluid regimen was used in six (7.5%) patients, and a liberal fluid regimen was used in one (1.2%) patient (Table 4). 3 ,6,8,11–15,17,19,21–47
Among all 79 patients, intravenous (IV) penicillin G (n = 22, 27.8%), IV ceftriaxone (n = 18, 22.7%), and doxycycline (n = 9, 11.3%) were the most commonly used drugs. Other commonly used drugs were IV hydrocortisone (n = 8, 10.1%), IV meropenem (n = 3, 3.7%), IV imipenem (n = 3, 3.7%), and IV ampicillin sulbactam (n = 3, 3.7%).
Complete recovery was reported for 36 (45.5%) patients. Biochemical and radiological recovery was reported for 10 (12.6%) and 9 (11.3%) patients, respectively. Death was reported in 18 (22.7%) patients. Respiratory failure due to pulmonary hemorrhage (n = 6, 7.5%) and MOF (n = 4, 5.0%) were the most common causes of death. Septic shock (n = 3, 3.7%) and bleeding (n = 2, 2.5%) were the other causes. The cause was not established in two patients.
Operative management was mentioned in only three (3.7%) patients, all of whom underwent exploratory laparotomy. All three patients had features of acute abdomen due to necrotizing pancreatitis, and two patients also had ascites. Two patients underwent necrosectomy and drainage of PPNC. One patient also had acute cholecystitis for which cholecystectomy and bile duct exploration were performed. The histologic findings were compatible with acute cholecystitis. Two patients achieved complete recovery. The outcome of the remaining patient was not reported.
Histological findings were obtained through autopsy and were available in only 13 (17.4%) patients. Among these 13 patients, edema (n = 9, 11.3%) and inflammatory lymphocytic infiltration (n = 8, 10.1%) were the most commonly reported findings. Hemorrhage (n = 5, 6.3%), fat necrosis (n = 3, 3.7%), congestion (n = 3, 3.7%), and calcification (n = 1, 1.2%) were the other findings.
Discussion
This systematic review revealed the clinical characteristics and outcomes of AP associated with leptospirosis. In general, the most common symptoms in patients with leptospirosis included fever, myalgia, occipital headache, red eye, and jaundice. 7 Compared with the aforementioned symptoms, abdominal pain was reported less commonly (30%–40%). 7 In patients with AP, however, abdominal pain was a predominant symptom that was observed in 97% of patients, and it has diagnostic significance according to the modified Atlanta criteria.16,48 Interestingly, abdominal pain in patients with concurrent AP and leptospirosis was not as common in the present review (64.5%). Thus, a high degree of clinical suspicion is required for a definitive diagnosis.
Leptospirosis occurs in two distinct phases: icteric and nonicteric. 49 Most patients with leptospiral infections are asymptomatic and have a nonicteric presentation. 50 Icterus was identified as one of the most common signs in the present review (43%). Frieden 51 reported that the incidence of icterus in patients with AP was 41.3% (n = 75) and that icterus originated from common bile duct enlargement secondary to the inflamed head of the pancreas as evidenced by autopsy. However, icterus is unlikely to be present solely due to pancreatitis in the absence of duct obstruction. 51
Goswami et al. 52 reported that 61.4% of patients with leptospirosis had icterus. Another study showed that among patients with leptospirosis, those with icterus had a higher mortality rate (5%–10%) than those without icterus (1%). 53 In the present review of patients with AP secondary to leptospirosis, the mortality rate was higher among patients with icterus (n = 16, 89%) than among those without icterus (n = 2, 11%). The sensitivity and specificity of an elevated amylase concentration are high on the first day of the CP of AP but decrease thereafter. 54 Smith et al. 55 reported a 76.8% sensitivity and 92.6% specificity for amylase in patients with AP. In the present review, hyperamylasemia was seen in 70.8% of patients with AP secondary to leptospirosis. In previous studies, however, hyperamylasemia has been noted in association with nonpancreatic etiologies. Furthermore, Edwards 49 reported the presence of hyperamylasemia in 65% of patients diagnosed with leptospirosis in the absence of pancreatitis. Daher et al. 17 reported that the presence of AKI in patients with leptospirosis may increase the amylase concentration and make the diagnosis of AP difficult. Therefore, hyperamylasemia must be carefully interpreted to diagnose AP when AKI is present in patients with leptospirosis. 17 A combined assay of elevated lipase and amylase is superior in achieving a diagnosis of AP. 56 This combined assay can be used when hyperamylasemia cannot be interpreted in the context of AP associated with leptospirosis. 56
In this review, IgM enzyme-linked immunosorbent assay (49.3%) and MAT (22.7%) were the most common methods of diagnosing leptospirosis. MAT is widely used in the diagnosis of leptospirosis because of its high sensitivity and remains the gold standard. 7 Furthermore, long-term persistence of IgM in the serum may interfere with the identification of acute infection and give rise to false positives. 7
In the present review, AKI (70%), hemodynamic instability (24%), and liver injury (14%) were the most common complications reported. The reported incidence of AKI in patients with leptospirosis ranges from 40% to 60%. 17 The presence of AKI increases the risk of mortality in patients with leptospirosis. 57 Daher et al. 17 reported a 22% mortality rate in patients with leptospirosis complicated by AKI. AKI also occurs in patients with severe AP. 58 Devani et al. 59 reported that the incidence of AKI in patients with AP was 7.9% (n = 3,466,493). The mortality rate is increased in patients with concurrent AKI and AP. 60 Thus, the presence of AKI carries a poor prognosis in patients with both leptospirosis and AP. 59 Furthermore, oliguria has been found to be an important prognostic factor in patients with leptospirosis.7,61 Therefore, patients with AP and leptospirosis should be carefully monitored and managed for AKI, and care should be taken to avoid iatrogenic causes of AKI such as drugs and contrast-induced nephropathy. In our review, we found that most patients with AKI needed HD (n = 20, 25.3%). Previous studies have shown that HD can be delayed up to 72 hours following the identification of AKI in patients with leptospirosis. 61 Low-volume HD is preferred to prevent pulmonary hemorrhage in patients with concurrent leptospirosis and AKI. 5 In our review, the mortality rate among patients with AKI was considerably high (29%).
Thrombocytopenia is an important hematological marker in the diagnosis and prognostication of leptospirosis and has been identified in more than 50% of patients.49,62 Furthermore, thrombocytopenia is a significant predictor of AKI in patients with leptospirosis and has been shown to correlate with mortality.63,64 However, various platelet abnormalities may be noted in patients with severe AP, including thrombocytopenia, thrombotic thrombocytopenic purpura, and disseminated intravascular coagulation.65–67 The recorded incidence of thrombocytopenia in patients with AP ranges from 36% to 43%. 66 Furthermore, the presence of thrombocytopenia has been shown to correlate with severe complications and ICU admission in patients with AP. 65 In the present review, thrombocytopenia was present in 50% of patients with concurrent AP and leptospirosis and could be attributable to the combined influence of both AP and leptospirosis. Furthermore, approximately 16% of patients with thrombocytopenia were admitted to the ICU. However, only one (3.2%) patient died among all patients with thrombocytopenia.
In our review, the most common radiological modalities used to diagnose pancreatitis were CECT (26.5%) and abdominal USS (21.0%). USS has a sensitivity of 62% to 90% in diagnosing pancreatitis in general, whereas CECT has an accuracy of 82% to 90% with 100% specificity.68–70 Furthermore, less than 3% of PPNC is missed in CECT. 69 USS is the preferred initial imaging modality to exclude biliary stones, which are among the most common etiologies of AP. 68 Silverstein et al. 71 conducted a study on pancreatic imaging and concluded that pancreatic CECT was superior to abdominal USS because CECT had an increased diagnostic yield (38%). However, CECT has the disadvantage of contrast administration to patients with leptospirosis who have impaired renal function. 68 Magnetic resonance imaging can be a good alternative in such patients. 68 In patients with the severe form of leptospirosis (Weil disease), the mortality rate ranges from 10% to 15%.23,72,73 Severe AP is associated with a mortality rate of 20%.74,75 In our review, death was reported in 18 (31.03%) patients who had concurrent AP and leptospirosis. Furthermore, respiratory failure due to pulmonary hemorrhage was the most common cause of death found in our review.
To overcome the challenges in diagnosing AP in patients with leptospirosis, we recommend a high degree of clinical suspicion combined with routine biochemical assays, including measurement of the serum amylase and lipase concentrations, and a basic USS. CECT should be reserved for select patients with diagnostic uncertainty and assessment of complications, and extreme caution is needed in the presence of AKI. Among patients who live in or have traveled to an area endemic for leptospirosis, those who have AP and jaundice without bile duct dilation should be promptly screened for leptospirosis. Although management includes antibiotics and supportive care, judicious fluid management is also required, preferably with monitoring of the central venous pressure or inferior vena cava filling and close examination for AKI. Surgical management should be the last option. Minimally invasive approaches are more advisable than exploratory laparotomy to avoid the morbidity induced by major surgery in physiologically compromised patients. A consensus statement or guidelines to manage severe pancreatitis in the context of leptospirosis should be developed by experts.
Limitations
This review had several limitations. Heterogeneity in the reporting of case reports and case series was noted, and a meta-analysis was therefore not feasible. Relatively few patients with concurrent AP and leptospirosis have been reported in the literature. Many data were missing and the reporting was incomplete, leading to inaccuracies in the data extraction and analysis. There were also considerable variations in the choice of investigations and management because of the lack of standard guidelines for management of AP in the setting of leptospirosis. Furthermore, the disease outcomes are likely to have been influenced by other complications of leptospirosis.
Conclusion
AP is uncommon but may give rise to severe complications in patients with leptospirosis. A high degree of clinical suspicion and different modalities of investigations are essential to achieve a correct diagnosis. AP can be easily missed in patients with leptospirosis because both conditions share similar CPs and complications. Judicious fluid management with monitoring for AKI is an essential component of supportive management. Mortality and morbidity are considerably higher when both AP and leptospirosis are present. A consensus statement or guidelines to manage severe pancreatitis in the context of leptospirosis is warranted.
Footnotes
Authors’ contributions
UJ, JCC, and DS conceived and designed the study, acquired and analyzed the data, and drafted the article. UJ, JCC, and DS collected, analyzed, and interpreted the data and wrote the article. UJ, JCC, and DS contributed to the design and conception of the study, revised it critically for important intellectual content, and approved the final version to be published. All authors have read and approved the final version of the manuscript.
Availability of data and materials
The data used in the above analysis will be available on reasonable request from the corresponding author.
Declaration of competing interests
The authors declare that there is no conflict of interest.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
