Acute Kidney Injury of Infectious Etiology in Monsoon Season: A Prospective Study Using Acute Kidney Injury Network Criteria

Inclusion criteria

All patients, with age ≥18 years and hospitalized with fever, in medicine or nephrology wards including intensive care units (ICUs), with AFI, suspected of infectious etiology and AKI, were included in the study. AKI was defined as an absolute increase in serum creatinine concentration of 0.3 mg/dl or greater as per the AKI Network (AKIN) consensus definition. Written informed consent was obtained from all cases.

Exclusion criteria

All known chronic kidney disease patients confirmed by ultrasound studies or blood or urine abnormalities of renal function of 3 months and pregnant females were excluded from the study.

Detailed history taking and clinical examination of patients were carried out and noted. Biochemical investigations were carried out on Olympus 400 Full Autoanalyzer. All were evaluated with complete blood count, renal function tests, electrolytes, liver function tests, urine examination, blood culture, urine culture, stool culture, X-ray chest, ultrasonography of kidney, ureter, and bladder, peripheral smear for malarial parasite, rapid diagnostic test for malaria (for the detection of histidine-rich protein II, specific to Plasmodium falciparum and Plasmodium lactate dehydrogenase specific to Plasmodium vivax in human blood sample), serum IgM antibody for leptospirosis, and serum NS1 antigen for dengue. Widal test was performed in all to identify infection. Anti-hepatitis A virus and hepatitis E virus IgM antibodies, anti-HIV, and hepatitis B virus and hepatitis C virus antibodies were done as per indication.

The severity of acute kidney injury was defined by the Acute Kidney Injury Network staging criteria as follows[9]

• Stage I, serum creatinine increase to 1.5–2 fold of baseline

• Stage II, serum creatinine increase to >2–3 fold of baseline

• Stage III, serum creatinine increase to >3 fold of baseline or serum creatinine of >4 mg/dl

• All patients requiring dialysis were categorized in Stage III.

Patients were followed up regularly during the period of hospitalization. Follow-up investigations, mainly renal function tests, were noted. Recovery or progression of illness with involvement of other organ systems, mainly pulmonary, cardiovascular, hematology, and central nervous system, was noted. Patients were assessed for the requirement of RRT and were treated accordingly as per indications.

Treatment

Statistical data analysis methods

Categorical variables were summarized as frequency and percentage. Continuous variables were represented using mean, standard deviation, and median. Association between categorical independent and dependent variables was determined using Chi-square test and Fisher's exact test for all 2 × 2 tables where P value of Chi-square test was not valid due to small counts (e.g., sex of the patient vs. mortality). Predictiveness of factors for mortality (outcome of AKI) was assessed using binary logistic regression analysis. Results were graphically represented where deemed necessary. All tests of significance were two sided, with P < 0.05 indicating statistical significance. The data were analyzed using the Statistical Package for the Social Sciences software version 17.0 [SPSS Inc, Chicago IL].

Incidence and clinical demography of acute kidney injury

It is generally believed that tropical infections contribute to more than 2/3^rd of AKI cases in tropics which may account up to 800 per million population according to certain calculations.[24] More than twenty tropical infections may be associated with AKI; the most widely known are malaria, leptospirosis, cholera, shigellosis, dengue, and HIV.[24]

In the present study, the incidence of AKI in monsoon due to infectious etiology was 13.21% [Table 1]. Incidence of AKI of infectious etiology by Daher et al.[25] was 17.7% [Table 6]. The main causes of ICU admission were AIDS-related diseases, pneumonia, leptospirosis, meningitis, and tetanus. Basu et al.[4] from CMC Vellore in South India in 2007–2008 reported an incidence of 41.1% of AKI in infectious diseases, one of the highest incidences noted among tropical infections. Most of the patients in the study done by Basu et al. were in risk class and had milder degree of renal dysfunction. Our study included patients of AKI due to infections in monsoon season, in prominent metropolitan city of a tropical country. The enormous burden of infectious diseases is reflected in the number of patients being admitted in the present study in 8 months which is more than four times the number of patients admitted in 1 year in the study done by Basu et al.[4]

Table 6 Acute kidney injury in tropical infections

Out of 230 patients of AKI in our study, 183 (79.59%) were males and 47 (20.43%) were females. Reason for gender difference is unclear but it is likely related to work activities. Mean age of study population was 40.95 ± 16.55 years. It implies the occurrence of community-acquired AKI in younger age group in tropics as compared with developed countries. Relatively young patients aged 37–47 years develop AKI of tropical diseases as a result of factors encountered in their environment.[10]

No combination of symptoms, signs, or demographic data [Table 1 and Figures 2, 3] can reliably distinguish between the different causes of fever. The common presentation with multiorgan involvement also highlights the widespread inflammation with endothelial dysfunction in common tropical infections such as malaria, leptospirosis, and dengue.

Etiology of acute kidney injury

In our study, malaria was the most common etiology of AKI affecting 65 patients (28.30%) [Table 3]. Malaria is estimated to account for 60% of outpatient visits, 30%–50% of hospital admissions, and 40% of public health expenditure in countries with a high burden of malarial disease.[26] India contributes 80% of all the cases of malaria in Southeast Asia.[1617] Basu et al.[4] reported malaria as the second most common tropical illness of their study accounting for 25.8% of cases of AKI of infectious etiology while Kaul et al.[27] noted that malaria was responsible for 18.8% of medical causes of community-acquired AKI. The changes in epidemiology of infections causing community-acquired AKI were highlighted by Prakash et al.,[28] with a significant increase in the number of AKI cases due to malaria from 4.7% in 1983–1995 to 17.09% in 1996–2008. Naqvi et al.[16] reported that 5.9% of community-acquired AKI was related to malaria.

AKI due to diarrheal diseases was the second most common cause of AKI accounting for 23% of patients in our cohort. RRT was necessary in 32.1% of these patients.

The incidence of AKI due to dengue was 16.50% in the present study. In Mumbai, dengue is a fast emerging important cause of AKI. As per the data obtained from the Satellite Disease Surveillance Unit, there is serial increase in the occurrence of number of cases of dengue in Mumbai along with consequent increase in the incidence of AKI (unpublished data). In our study, 19 patients (50%) with dengue presented in AKIN Class I, but 44.70% of the total patients with dengue required dialysis. Mortality was maximum with dengue in our cohort, at 23.7%. AKI is a complication of severe dengue infection and is typically associated with hypotension, rhabdomyolysis, or hemolysis.[29] Incidence of 16.5% of AKI in dengue in our group is in sharp contrast to 4% and 7.6% as reported by Dahr et al. and Basu et al., respectively. The prevalence of AKI was 1.6% among 617 children with dengue hemorrhagic fever (DHF) in Colombia, 3.3% in hospitalized adults with DHF, 4.9% in 81 Chinese patients with DHF/dengue shock syndrome, and 5% in DHF patients in Qatar.[30]

Leptospirosis is another emerging cause of AKI in Mumbai. There is a sharp increase in its incidence during rainy season or after floods. AKI occurs in 20–85% of cases of leptospirosis.[31] In the present study, the incidence of AKI due to leptospirosis was 13%. Major factors involved in pathogenesis are direct nephrotoxic action of leptospires, toxin-induced immune response, hyperbilirubinemia, rhabdomyolysis, and hemodynamic alterations due to decrease in peripheral vascular resistance, diarrhea, vomiting, and fever. Basu et al.[4] in their study had 3.3% incidence of AKI due to leptospirosis. Kaul et al.[27] in their study had 8% incidence of AKI due to leptospirosis. Prakash et al.[28] did not note any case of AKI due to leptospirosis during 26 years. Leptospirosis is the leading cause of AKI in Thailand and Singapore. Daher et al.[2325] in their study of AKI in infectious disease in ICU noted a 11% incidence of leptospiral AKI. Thus, the incidence of AKI due to leptospirosis varies as per the region.

In 10.4% of patients with AKI, the etiology could not be determined and were labeled as cases of AKI due to undifferentiated etiology. AKI in most of these patients was suspected to be secondary to viral infection. The diagnosis is difficult requiring expensive antibody-based tests which are not available in a tertiary care charitable government hospital. These are diseases caused by RNA viruses belonging to Flaviviridae, Arenaviridae, Bunyaviridae, and Filoviridae groups.[32] In the present study, patients were not evaluated for scrub typhus, hantavirus, and melioidosis which are other emerging causes of tropical illness causing AKI. In addition, the rapid antibody-based tests for malaria, leptospirosis, and dengue are done at admission. These tests may report negative if done before antibodies are developed against the causative infection. These tests should ideally be repeated postadmission since they are not both 100% sensitive and specific. However, due to enormous workload during monsoon along with limited resources, these are usually not repeated if patient shows response to therapy. In the study done by Basu et al.,[4] there were 8.4% cases, of undifferentiated AFI with AKI.

In the present study, 12 patients were affected with mixed infections [Table 3]. About 58.3% of patients with AKI due to mixed etiology presented in more severe stages of AKI requiring dialytic support [Figure 6].

In our cohort, eight patients (3.5%) of typhoid were diagnosed with AKI. Basu et al.[4] reported 8.7% cases of AKI due to typhoid whereas Daher et al.[23] reported none.

Distribution of study group as per the Acute Kidney Injury Network stages and requirement of renal replacement therapy

In our cohort, maximum number of patients were seen in Class III: 111 patients (48.26%) [Table 1]. In the study done by Yang et al.[33] in China, 271 patients were classified as per the AKIN criteria with maximum number of patients seen in Class III: 141 patients (52%), followed by Class I: 86 patients (33.7%), and Class II: 44 patients (16.2%). Basu et al. classified AKI according to RIFLE criteria with maximum number of patients of AKI, 64 (42.4%) in risk class followed by 53 patients (35.1%) in failure class, and 34 patients (22.5%) in injury class. Thus, a significant number of patients are seen in more severe class of AKI irrespective of criteria used for classification in cases of community-acquired AKI.

Leptospirosis, mixed etiology, malaria, undifferentiated, etiology and dengue were associated with severe AKI in our study group [Figure 5], with more than 40% of each etiology needed RRT. Widespread inflammation with endothelial dysfunction in common tropical infections such as malaria, leptospirosis, and dengue along with multiorgan involvement is the culprit for severe AKI. Leptospirosis in the early 1980s was seen as benign infection with nonoliguric AKI, not requiring RRT. Over the past 30 years, it has become a more severe disease with AKI necessitating RRT and significant mortality. Basu et al.[4] in the year 2011 in South India reported an incidence of 12% for leptospirosis among which 6% developed AKI, but none of them required RRT and there was no mortality. The present scenario with higher mortality and morbidity may be related to severe strain development or drug resistance.

What is worrisome is dengue emerging as a cause of severe AKI. In the present study, 44.70% of total patients with dengue presented with Stage III AKIN and required dialysis. In the study by Basu et al.,[4] out of ten patients with dengue-related AKI, two required dialysis but mortality was maximum at 25% in patients with dengue. In the study done by Daher et al.,[23] 4% of total admissions of AKI in ICU were due to dengue. Acute renal failure is a potential complication of severe dengue infection and is typically associated with hypotension, rhabdomyolysis, or hemolysis.[29] Development of AKI in patients with dengue infection is associated with increased mortality. In Thailand, the prevalence of AKI in fatal DHF was 33.3%, compared with 0.3% in all DHF cases.[34]

Patients with AGE or typhoid presented with less severe class of AKI and lesser need for hemodialysis. Out of the 230 patients involved in the study, 103 patients (44.78%) required dialysis. In AKIN Stage III (92.80%), patients needed dialysis. In the present study, correlation of etiology with both, AKIN class depicting the severity of AKI and need for RRT, was statistically significant (P = 0.001).

The number of patients requiring dialysis varies in different studies of infection-related AKI carried out in India and other tropical countries. In the study by Basu et al.,[4] 19.2% of patients received dialysis in South India and 27.6% and 35% of patients received dialysis in two separate studies done by Daher et al.[2325] in Brazil. Eighty-three percentage of patients received dialysis in the study done by Kaul et al.[27] in North India. Ninety-two percentage of patients received dialysis in the study done by Mehta et al.[5] in Mumbai. Requirement of dialysis in tropical infections varies from 19% to 92% depending on the underlying etiology and severity of illness. Studies by Daher et al.[2325] enrolled ICU patients comprising mainly HIV/AIDS-related cases, pneumonia, meningitis, and tuberculosis. Infections such as leptospirosis and dengue comprise small proportion of total patient base which differs from patient base of the present study which involves infections such as malaria, leptospirosis, dengue, and more than one etiology comprising a combination of the same infections.

Predictors of severity of acute kidney injury

Association between severity of AKI with etiology, multiorgan failure, vasopressor support, hyperbilirubinemia, ARDS, and need for assisted ventilation was statistically significant in this study [Table 4].

Mortality and prognostic factors significantly associated with mortality

In the present study, maximum mortality [Figure 7] was seen in patients with dengue: 9 patients (23.7%) followed by leptospirosis: 5 patients (16.7%), and those with mixed etiology: 2 patients (16.7%) and undifferentiated etiology 3 patients (12.5%). Seven patients (10.8%) died due to malaria. Although malaria was the causative agent for the highest number of AKIs in our study, dengue and leptospirosis and those with multiple etiologies were responsible for higher mortality. Etiology as a risk factor for mortality was statistically significant in the present study [Table 5]. The development of AKI in patients with dengue infection is associated with increased mortality.[30] The mortality of AKI associated with a specific disease such as severe malaria, that is often complicated with multiple organ failure, is very high.[35] Leptospirosis presenting with AKI is a severe disease, frequently leading to multiorgan failure and death in one-third of the patients.[36] Significant morbidity and mortality has been reported by Yang et al.[37] in patients with AKI due to leptospirosis. Mortality with diarrheal illness is 3.8% which has decreased due to practices of early, rapid, and careful hydration with definitive treatment. No mortality was seen with typhoid. In two separate studies in Brazil, by Daher et al.[2325] in AKI in infectious diseases ICU, the mortality rate was 62.8% and 66%, respectively, reflecting the more critical illness of patients admitted in ICU.

The maximum number of deaths involving 25 patients (22.5%) was seen in AKIN Stage III which correlates with the severity of AKI. Association of mortality with increasing severity of AKIN stages is statistically significant. Risk for mortality increased with increasing delay in seeking medical attention (P = 0.001).

More deaths were observed in patients requiring vasopressors (P = 0), thus implying the role of hemodynamic instability, hypoperfusion-related cellular injury, and multiorgan involvement increasing the risk for mortality.[38] In multivariate analysis [Table 5], need for vasopressor support (odds ratio [OR] = 16.29) and assisted ventilation (OR = 27.05) was the significant risk factor for mortality.

It would have been ideal to analyze the following parameters: time of diagnosis of AKI after admission, time of nephrology consult after admission, use of diuretics, RRT requirement, and type of RRT with respect to incidence of CKD after an episode of AKI. However, these patients once discharged after recovery are not on regular follow-up.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.