Acute kidney injury (AKI) complicates 5–7% of acute care hospital admissions and up to 30% of admissions to the intensive care unit. 1 The tropical zone which is geographically limited by the Tropics of Cancer and Capricorn, is an area with extreme heterogeneity in terms of ethnic composition, as well as the socioeconomic and developmental status. Acute kidney injury remains one of the most enigmatic problems in these regions with high incidence rates (25% - 80%) especially in the intensive care units.

There are very few studies on the etiologic factors of acute kidney injury in the intensive care units in South India. Unlike in the developed nations, were sepsis is a predominant etiologic factor; in high rainfall receiving regions of the tropics - like Kerala, infections like leptospirosis form a major cause of AKI.

To study the etiologic factors, clinical profile and outcome of patients with Acute Kidney Injury admitted in Medical intensive care unit.

Study Design : Cross sectional analytical study

Study Setting : Medical Intensive Care Unit,

Government Medical College, Kottayam

Duration of Study : 12 Months

Study Population : Patients with AKI diagnosed by AKIN Creatinine Criteria

Sample Size

n = (Z α) ² pq/d²

Z α = 1.96 for α at 5% level of significance

p = prevalence of sepsis causing AKI- 38.6%

(Based on Eswarappa et al on spectrum of AKI in critically ill patients2

q = 100 – p

d = absolute precision 3

(4x38.6x61.4)/8x8=148

Minimum required sample size - 150

Study Tool : Performa based on case records

Inclusion Criteria : atients admitted to MICU within the study period with diagnosis of Acute Kidney Injury by AKIN creatinine criteria

Exclusion Criteria : Patients with Chronic Kidney disease

Study Procedure : 150 consecutive patients admitted to Medical Intensive Care Unit during my study period meeting the AKIN creatinine criteria for acute kidney injury was included in this study. All patients or their relatives were interviewed as per the prepared performa and then complete clinical examination was done.

The following investigations were performed: blood routine, random blood sugar, renal function tests, liver function tests, serum electrolytes, serum uric acid, INR, blood pH, urine analysis, ECG, chest X ray, ultrasonography of abdomen for all cases and specific tests for diagnosis in special cases. The data collected was entered in Microsoft excel and analyzed using SPSS software. Associations between various factors were assessed using chi square test for qualitative variables and t test/ANOVA for quantitative variables. Appropriate non parametric tests were applied wherever required. The level of statistical significance was taken as p value less than 0.05.

Ethical consideration: The above mentioned study was conducted in this institution after obtaining due ethical clearance from the Ethical Committee of Government Medical College, Kottayam. Informed consent was obtained from cases in the study.

AKI affects patients worldwide, leading to decreased survival, and sometimes to new onset of CKD. Outcomes in AKI are influenced by the underlying disease, the severity and duration of renal impairment and by the baseline characteristics of the patient. The cause of AKI is often multi-factorial.

The role of sepsis in AKI has been well documented in western literature, causing nearly 50% of the AKI cases in few studies. 4,5,6 Sepsis is associated with high burden of illness, greater abnormality in acute physiology & laboratory findings & greater non-renal organ failure 1,2,3 . Approximately 25-35% of patients with severe sepsis and 40-55% of patients with septic shock die within 30 days1,2.

Other causes include acute gastroenteritis, nephrotoxic drugs, cardiac causes and hepatic causes. Hypertension, diabetes and coronary artery disease are predominant comorbid conditions. Fidel Barrantes et al concluded that the Acute Kidney Injury Network (AKIN) definition of acute kidney injury predicts hospital mortality, need for renal replacement therapy and prolonged hospital stay in critically ill patients.4

The true picture of AKI in the tropics is poorly understood due to the late presentation of patients to tertiary centres. Tropical infectious diseases are an important cause of admission to ICU in Kerala, and diseases like leptospirosis have a high propensity to cause AKI.

CASE DEFINITIONS

Acute Kidney Injury:

Defined by AKIN Creatinine Criteria

Stage 1: Increase of S Cr ≥0.3 mg/dl / increase of S Cr ≥150% to 200% over 48 hrs

Oliguria urine output <0.5 ml/ kg / hr over 6 hrs

Stage 2: Increase of S Cr ≥200% to 300% over 48 hours

Oliguria urine output <0.5 ml/ kg / hr over 12 hrs

Stage 3: Increase of S Cr ≥ 300%

S Cr ≥4 mg/dl and acute rise ≥0.5mg/dl

Initiation of RRT7

A diagnosis of sepsis/septic shock was made where the primary reason for admission was a sepsis related illness, and included sepsis associated with pneumonia, gastrointestinal disease, urinary tract infections, central nervous system infections, soft tissue infections and sepsis of undetermined source based on a qSOFA score >2 in presence of documented infection. A cardiac diagnosis was made where the primary reason for admission was cardiogenic shock, cardiac arrest, congestive cardiac and acute myocardial infarction.

A respiratory diagnosis encompassed aspiration syndrome, exacerbations of chronic obstructive pulmonary disease or asthma. A diagnosis of gastrointestinal diagnosis included chronic liver disease, pancreatitis and acute diarrheal disease. Metabolic/poisoning diagnoses included causes of metabolic coma, diabetic ketoacidosis, drug overdose or other endocrinopathies. Primary neurologic diagnoses included stroke or other neurological causes for coma. A diagnosis of Dengue by IgM Dengue ELISA test. A diagnosis of probable Leptospirosis based on System of scoring using the Modified Faine’s Criteria 8

1. Gender distribution

73.3% of the patients in the study group were males and 26.7% were females.

2. Age distributionTable 1

Mean age was 49.6 years with a standard deviation of 16.5. Youngest and the oldest of the study population had 15 and 88 years of age respectively.

3. Presenting symptomsFigure 1

Fever was the most common presenting symptom seen in 108 (72%) patients, decreased urine output in 72 (48%) patients, breathlessness in 68 (45%) patients, and myalgia in 64(43%) patients.

Other common presenting symptoms were headache, cough, altered sensorium and abdominal pain seen in 35 (23%), 30 (20%), 28 (19%) and 28 (19%) respectively.

4. Nephrotoxic drug intakeTable 2

Only 10 patients that accounts for 7 % had h/o nephrotoxic drug usage. NSAID (2.7%) was the most common one implicated used by 4 patients prior to development of AKI.

5. Etiologic factorsTable 3 Figure 2

Table 3

Etiologic factors	Frequency	Percentage
Leptospirosis	33	22 %
Undifferentiated acute febrile illness	32	21.3 %
Sepsis	26	17.3 %
Respiratory diagnosis	17	11.3 %
Cardiac diagnosis	14	9.3 %
Gastrointestinal	13	8.7 %
Metabolic/poisoning diagnoses	7	4.7 %
Primary neurologic diagnoses	3	2 %
Dengue	3	2 %
Drug	1	0.7 %
Snake bite	1	0.7 %
Total	150	100.0 %

The most common etiology of AKI in the patients in this study group was leptospirosis seen in 33 patients which accounts to 22% followed by undifferentiated fever seen in 32 patients (21%) and sepsis seen in 26 patients (17%).

6. Stage of acute kidney injury by the AKIN creatinine criteriaFigure 3

The most common stage of AKI in the patients in this study group on admission to the medical intensive care unit was stage 3 seen in 97 patients which accounts to 64.7%.

7. Dialysis RequirementFigure 4

68 out of the 150 patients in the study group required dialysis, which accounts to 45%. Among the patients in the study group who underwent dialysis 52 patients (76.5%) underwent intermittent hemodialysis while 16 patients (23.5%) underwent CRRT.

8. Indication For DialysisTable 4

The predominant indication for dialysis in the patients in the study group was metabolic acidosis and anuria in 52.9% of the patients.

9. Ventilation RequirementFigure 5

66 out of the 150 patients which accounts to 56% required invasive ventilation.

10. Inotrope RequirementTable 5

Table 5

	Frequency	Percentage
No	25	16.70%
Yes	125	83.30%
Total	150	100.00%

125 of the 150 patients in the study group which accounts for 83% required inotropic support.

11. Mortality Figure 6

Mortality was 41% in the patients in the study group, 62 patients out of 150 expired.

12. Association Between Aki Stage And MortalityTable 6

53% of those with stage 3 AKI expired compared to no mortality in the stage 1or2.

13. Association Between Inotrope Use And MortalityTable 7

Mortality was more in the patients in this study requiring inotrope support. 48% of those with inotropic support expired.

14. Association Between Ventilation And MortalityTable 8

Mortality was more in the patients in this study requiring ventilator support.

15. Association Between Diagnosis And OutcomeTable 9 Figure 7

Mortality was more in the patients in this study with undifferentiated acute febrile illness, 43.8% and leptospirosis, 36.4%.

16. Association Between Diagnosis And Stage Of AKITable 10 Figure 8

Figure 8

90.9% of those with leptospirosis and 87.5% of those with undifferentiated febrile illness presented in AKI stage 3.

17. Logistic Regression Analysis Of Outcome PredictorsTable 11

Table 11

		S.E.	Wald	df	Sig.	Exp(B)
Basal crepitations	-.416	.654	.405	1	.525	.660
Breathlessness	.989	.643	2.364	1	.124	2.687
AKI stage	20.905	5237.667	.000	1	.997	1199225957.801
Ventilation	3.535	.726	23.717	1	.000	34.291
Inotrope	2.102	1.043	4.059	1	.044	8.186
Indication for dialysis	-.170	.118	2.050	1	.152	.844
Constant	-66.870	15713.000	.000	1	.997	.000

By logistic regression analysis of the factors having association with mortality, requirement of ventilator support and inotropic supports were independent predictors of mortality.

AGE AND GENDER

A trend towards an increasing number of AKI cases among male patients was observed compared to female patients [73% and 27 % respectively] in this study.

Compared with all ICU admissions, patients with AKI were found in male gender in Basu et al, Cruz et al and Daher et al9,10,11 . The reasons for such gender differences are unclear but are likely associated with work activities.

Majority of patients in this study belonged to 55-65 years (21%) followed by 45-55 years (20%). Mean age was 49.6 years with standard deviation 16.5. The population of patients with AKI in developing tropical countries is younger (30–40 years of age) than that reported in developed temperate countries (60–70 years of age) like Angus et al 12

CLINICAL FEATURES

Fever was the most common presenting symptom (72%) followed by decreased urine output (48%), breathlessness (45%), and myalgia (43%). Predominant presenting features included fever (89.6%), oliguria (70.4%), and breathlessness (40.4%) as in Mehta et al 13 .

Only 7 % of the patients in this study had h/o nephrotoxic drug usage. NSAID (2%) being the most common one implicated.

The most common general examination finding in the patients in this study was icterus (52%) followed by conjunctival congestion (22%). The most common systemic examination finding in patients in this study was basal crepitations (40 %). In Eswarappa et al, edema was seen in 28.0%, pallor in 20.0% and icterus in 21.0% of the patients 6

The mean serum creatinine in the patients in this study on admission to the Medical ICU was 3.72 mg/dL with a standard deviation of 1.65. Mean serum creatinine was 3.64 with SD of 3.1 in a study by Maulita et and 10.18 ± 5.19 mg/dl in a study by Ahmed et al 14

DIAGNOSIS

The most common diagnosis in the patients in this study was leptospirosis (22%) followed by undifferentiated fever (21%) and sepsis seen in 17%.

In contrast to sepsis, drugs and cardiogenic shock in the developed world, acute tubular necrosis (ATN) due to community-acquired infections remains the commonest cause of AKI in the tropics. Chug, Medve and Angus et al15,16,17

The main causes of ICU hospitalization were acquired immunodeficiency syndrome (AIDS)-related diseases (28 .6%), pneumonia 13%), leptospirosis (11.6%) in Daher et al 16 It was scrub typhus (51.2%), falciparum malaria (10.4%), and enteric fever (8.7%) in study by Basu et al 14 .Unlike in our study the predominant cause of AKI in most studies in tropical areas was due to malaria.

90.9% of those with leptospirosis and 87.5% of those with undifferentiated febrile illness presented in AKI stage 3. Significant association was identified between the etiology of AKI and the stage of presentation on admission to the medical intensive care unit.

Significant association was identified between etiology of AKI and the need for dialysis. Dialysis was required in 81.3% among those with undifferentiated febrile illness and 78.8% among those with leptospirosis and the association was significant by Chi square test.

Leptospirosis with AKI admitted to ICU had a significantly higher frequency of severe AKI 73.2 % and a higher prevalence of dialysis requirement (57.3 %) in a study by De Francesco et al18

MANAGEMENT

The most common stage of AKI in the patients in this study group on admission to the medical intensive care unit was stage 3 in 64.7%. Unlike in a study by lopes et al in which AKI occurred in 50.4% of patients – 21.1% with Stage 1, 10.1% with Stage 2 and 19.2% with Stage 3 .

45% of the patients in the study group required dialysis, among which 24% underwent CRRT while the 76% underwent hemodialysis.

In a study conducted by Bernieh b et al. 58% of patients were managed conservatively while 42% patient were managed with dialysis which is comparable with present study12

The predominant indication for dialysis in the patients in this study group was metabolic acidosis with anuria. The most common indication for dialysis was refractory fluid overload (89.4 %), followed by uremic signs and symptoms (61.1 %) in study by Ibrahim et al 19 , probably because CKD patients were also included.

MORTALITY

Mortality was 41% in the patients in the study group. The ICU mortality rate was 39.3% (44/112) in study by Medve et al 18 which is comparable with our study.

Maximum mortality was seen in the age group 55-65years - 22.6% and 45-55 years. - 17.7%. Mortality was more, 72.6% in the male patients in this study group compared to females. However, no significant association was identified between age group, gender, patients with comorbidities, and mortality. Age, gender or the presence of co-morbidities was not associated with a higher risk of AKI as in Wijikerma et al 13

Among those with breathlessness as presenting symptom, 58.8% expired compared to only 26.8% among those with no breathlessness on admission to the Medical ICU. Significant association was identified mortality and breathlessness as the presenting symptom. Among those with Basal Crepitations, 53.3% expired compared to only 33.3% among those with no Basal Crepitations on admission to the Medical ICU. Significant association was identified.

Respiratory involvement, along with AKI had statistically significant chances of poor outcome (p<0.05) in a study by Bhadade et al 20

Mortality was more in the patients in this study with undifferentiated acute febrile illness, 43.8% and leptospirosis, 36.4%. No significant association was identified between the etiology of AKI and mortality unlike in study by Eswarappa et al which identified significant difference in mortality due to medical versus surgical etiology of AKI.

53% of those with stage 3 AKI expired compared to no mortality in the stage 1 or 2. Significant association could be identified between stage of AKI and mortality. AKIN stages based on creatinine criteria predicted mortality as in a study by Lopez et al. 21

Mortality was more in the patients in this study who underwent dialysis 48.5% compared to those who did not require it. This may be because of the fact that they are more hemodynamically compromised. But no significant association was identified between dialysis and mortality.

The need for dialysis treatment has been associated with a higher mortality rate (50-70%) than among patients with AKI who did not require RRT in study by Bhadade et al 22

Significant association was identified between indication for dialysis and mortality. 60% of the patients who underwent dialysis and died had both metabolic acidosis and anuria. Significant association was identified between patients with metabolic acidosis and anuria and mortality in Kumar et al. 23

Mortality was more in the patients in this study requiring inotrope support (48%) and ventilator support (70.2%). Significant association was identified between inotropic and ventilator support; and mortality. Vasopressor support and assisted ventilation were predictors for mortality in a study by Mathew George et al.24

By multivariate logistic regression analysis of the factors having association with mortality, requirement of ventilator support and inotropic supports were independent predictors of mortality. Significant association was identified between patients with ventilator support and mortality in Saxena et al ⁴⁷and Liano et al. 25

150 consecutive patients diagnosed to have acute kidney injury admitted to the medical ICU of a tertiary care center in Kerala were studied. The main burden of diseases were due to infective etiologies. The most common was leptospirosis (22%) followed by undifferentiated fever (21%) and sepsis (17%). This is unlike to studies done in developed countries. This reflects mainly the trend in tropical regions with higher rainfall. Mortality was 41% in the study and requirement of ventilator support and inotropic supports were independent predictors of mortality.

In the tropics, diseases like leptospirosis have a high propensity to be the cause of AKI. So a high degree of suspicion should be kept to aid in the diagnosis and early referral to centers with intensive care facilities.

Appendix-1

LIST OF ABBREVIATION