Outcome of Dengue Hemorrhagic FeverCaused Acute Kidney Injury in Thai Children

Kamolwish Laoprasopwattana, MD, Pornpimol Pruekprasert, MD, Pornsak Dissaneewate, MD, Alan Geater, PhD, and Prayong Vachvanichsanong, MD Objectives To examine the outcome of acute kidney injury (AKI) in children with dengue hemorrhagic fever (DHF), the cause(s) of AKI, and the risk of AKI and fatality. Study design The medical records of patients age <15 years during 1989 to 2007 were reviewed. DHF-caused AKI and patients with DHF with no AKI were matched 1:2 by age. Results DHF-caused AKI was clinically estimated to be 0.9% (25/2893) of admissions, with a high mortality rate of 64.0%. Risk factors of AKI were DHF grade IV and obesity (odds ratio, 16.9; 95% CI, 4.2 to 68.5, and odds ratio, 6.3; 95% CI, 1.4 to 28.8, respectively). Respiratory failure, hepatic failure, and massive bleeding were complications found in 80.0%, 96.0%, and 84.0% of cases with AKI, respectively. Fatality was more likely in cases with DHF grade IV, oliguric AKI, respiratory failure, or prolongation of prothrombin or activated partial thromboplastin time more than twice that of reference specimens. Among the survivors, none had chronic kidney disease, and serum creatinine levels returned to normal in 32 (1 to 48) days. Conclusions Patients with DHF and AKI had a high mortality rate, although those who survived had a full return to normal function within 1 month. DHF grade IV and obesity were the major risk factors of AKI. (J Pediatr 2010;157:303-9).
engue virus is a mosquito-borne virus with 4 serotypes; infection with 1 serotype provides lifelong homotypic immunity, but there is only short-term cross-protective immunity against heterotypic serotypes. Epidemiological evidence has demonstrated that dengue hemorrhagic fever/dengue shock syndrome (DHF/DSS), the severest form of dengue viral infection (DVI), is associated more commonly with secondary DVI.1 More than 100 million cases of dengue fever (DF) and several hundred thousand cases of DHF, with an overall fatality rate of approximately 2%, occur each year in tropical countries.2 DVI can present with a wide variety of clinical manifestations including asymptomatic infection, undifferentiated fever DF, DHF, and the life-threatening DSS. All are more common in children than in adults.3,4 Acute liver injury, respiratory distress, and hemorrhagic complications are the major risk factors for lethal DHF/DSS.5-7 Mild serum creatinine elevation is common in DHF,8 but acute kidney injury (AKI) is rare, so most studies in DVI with AKI have been only small series or case reports. A recent study in adults found DHF-caused AKI in 3.3% of cases (10/304), but with a mortality rate of 60% in these cases.9 Hypotension, rhabdomyolysis, acute hemolysis, and direct kidney injury causing AKI in DVI have been reported.9-13 A previous study from our institute found that besides sepsis, DVI and leptospirosis were the most common infectious causes of AKI in children.14 There are no published cohort studies on DVI with AKI in children, thus this study from the major tertiary care institution in southern Thailand aimed to determine the outcome and clinical course of AKI in a large group of children with DVI, the actual cause of AKI in these children, and the risk of AKI and fatality in DHF-caused AKI.

Methods
We retrospectively reviewed the medical records of all children (<15 years of age) diagnosed with AKI caused by DVI admitted from January 1989 to December 2007 in Songklanagarind Hospital, the major tertiary care center for the 14 provinces of southern Thailand. Permission from the institution review board of the university was ALT Serum alanine aminotransferase obtained. DVI was diagnosed according APTT Activated partial thromboplastin time to the criteria of the World Health
AST BUN Cr DB DHF PT TB WHO WSDS Serum aspartate aminotransferase Blood urea nitrogen Serum creatinine Direct bilirubin Dengue hemorrhagic fever Prothrombin time Total bilirubin World Health Organization Weight standard deviation score

From the Department of Pediatrics (K.L., P.P., P.D., P.V.) and the Epidemiology Unit (A.G.), Faculty of Medicine, Prince of Songkla University, Hat Yai, Songkhla,Thailand The authors declare no conicts of interest.
0022-3476/$ - see front matter. Copyright 2010 Mosby Inc. All rights reserved. 10.1016/j.jpeds.2010.02.008

303

THE JOURNAL OF PEDIATRICS

www.jpeds.com

Vol. 157, No. 2 Student t-test or Mann-Whitney U test for normally distributed and non-normally distributed continuous variables, respectively. Fisher exact test was used for comparisons of categorical data. Multivariate logistic regression analyses were used to identify risk factors for AKI but not for risk factors for death as the outcome numbers were too low. Stata version 7 (StataCorp, College Station, Texas) was used for statistical analysis.

Organization (WHO),15 primarily based on the presence of dengue IgM and the hemagglutination inhibition test (HAI). Primary and secondary DVI were diagnosed if there was a 4-fold increase of HAI and the titers were #1:1280 and $1:2560, respectively. DHF was diagnosed if the patient fullled all of the following criteria: acute febrile illness; hemorrhagic manifestation; thrombocytopenia (<100 000 platelets/mm3); and evidence of plasma leakage as determined by hemoconcentration (hematocrit increased above baseline by $20%), pleural or abdominal effusion (as revealed by radiography or another imaging method) or hypoalbuminemia. DHF grade I was diagnosed if the patient met all of the DHF criteria without evidence of circulatory failure. DHF grade II was diagnosed if the patient had evidence of a bleeding disorder. DHF grade III or IV (DSS) was diagnosed if the patient met all of the DHF criteria and there was also evidence of impending (narrow pulse pressure, <20 mmHg) or profound circulatory failure. To determine the risk factors of AKI, the patients who had DHF-caused AKI were matched with patients with DHF by age at a ratio of 1:2. Demographic characteristics and known potential risk factors for disease severity were recorded, including age, sex, underlying diseases, weight standard deviation score (WSDS), obesity (WSDS >2), and severity of DVI according to the WHO criteria.15 AKI was dened by a sudden increase in serum creatinine (Cr) levels >2 mg/dL or a serum creatinine concentration that was >2 times previous or subsequent values and that was also higher than the upper limit of normal values for the patients age.16 Urine output lower than 0.5 mL/kg/h was classied as oliguric AKI. Hepatic failure was dened by the rapid development of severe acute liver injury with impaired synthetic function and encephalopathy in a patient with no history of liver disease. Respiratory failure was dened by severe hypoxemia requiring a mechanical ventilator. Multiple organ failure was deemed in patients who had failure of $2 organs, including AKI, respiratory failure or hepatic failure. Dual infection was dened as infections detected within 3 days after admission to the hospital. Nosocomial infection was dened as a positive bacterial/fungal culture >72 hours after admission. To determine the severity of plasma leakage, coagulopathy, hepatitis, and renal impairment, the highest and lowest levels of hematocrit, the lowest platelet count, the highest prothrombin (PT) and activated partial thromboplastin times (APTT), highest levels of aspartate aminotransferase (AST), alanine aminotransferase (ALT), total bilirubin (TB), direct bilirubin (DB), blood urea nitrogen (BUN), and Cr, and the lowest levels of albumin were obtained from medical records. To determine the differences of oliguric and nonoliguric AKI, urine output and urine ndings (protein, blood, WBC, RBC) were recorded. Statistical Analysis Data were evaluated using descriptive statistics (mean and standard deviation, median, and interquartile range [IQR], or frequency and percentage, as appropriate). Comparisons between patients with and without AKI were made using
304

Results
Characteristics and Clinical Course of AKI in DHF During the 19-year period, 2893 patients <15 years of age needed hospitalization due to clinically diagnosed DVI. Of these, 25 (0.9%) patients had or later developed AKI and 16 of 25 (64.0%) of these patients died. Fourteen of 25 (56.0%) were male and the median age of all was 10.0 years (range, 6 months to 13.1 years). The mean serum Cr level was 4.9 2.9 mg/dL. Most (18/25, 72.0%) were referred from other hospitals, and 9 of these 18 (50.0%) had AKI before being referred to our hospital. Only 21 of 25 had a recorded weight, and 9 of these 21 (42.9%) were obese (Table I). Secondary DVI was diagnosed in 17 patients and primary DVI in a 6-month-old infant. Of the 7 patients who had only a rst serum test available because of early death due to profound shock, dengue antibodies were detected in all tests, but it could not be determined whether these patients had primary or secondary DVI. Of the 25 patients with AKI, 24 had a nal diagnosis of DSS (DHF grades III and IV in 5 and 19 patients, respectively); the 25th patient was classied as DHF grade II (Table I). Of the 24 patients who had DSS, AKI developed within 24 hours of the appearance of shock in 18 of 24 (75.0%) patients. One patient who had DHF grade II also had status epilepticus, but we could not conrm whether this patient had rhabdomyolysis because a serum creatinine kinase test had not been performed. Renal replacement therapy was performed in 11 of 25 (44.0%) patients. Hepatic failure, respiratory failure, and serious bleeding requiring blood transfusion were found in 24 of 25 (96.0%), 20 of 25 (80.0%), and 21 of 25 (84.0%) of the cases, respectively (Table I). Four patients had an underlying disease, one each with asthma, vesicoureteral reux grade III, angiobromatosis, and glucose-6-phosphate dehydrogenase (G-6-PD) deciency. All 4 of these patients developed profound shock and all except the one who had G-6-PD deciency died. The patient who had angiobromatosis also had massive epistaxis from the original tumor. The patient who had asthma was asymptomatic during the DSS, and the patient who had vesicoureteral reux had normal Cr levels and no urinary tract infection before developing DSS. The patient who had G-6-PD deciency also had intravascular hemolysis. Risk Factors of Acute Kidney Injury in DHF To determine the risk factors of AKI in DHF, we compared the clinical characteristics and possible risk factors of AKI
Laoprasopwattana et al

August 2010

ORIGINAL ARTICLES

Table I. Comparison of patients with DHF with and without AKI

Factors Male, n, (%) Age, years, mean SD WSDS, median (IQR) Referred, n, (%) Initial diagnosis Undifferentiated fever, n, (%) Dengue fever, n, (%) DHF grade I-II, n, (%) DHF grade III, n, (%) DHF grade IV, n, (%) Final diagnosis DHF grade II, n, (%) DHF grade III, n, (%) DHF grade IV, n, (%) Respiratory failure, n, (%) Liver failure, n (%) Bleeding requiring transfusion, n (%) Nosocomial-acquired infection, n (%) Death, n (%) Highest hematocrit, %, mean SD Lowest hematocrit, %, mean SD Lowest platelet, cells/mm3, median (IQR) Highest PT, seconds, mean SD Highest APTT, seconds, mean SD Highest DB, mg/dL, median (IQR) Highest TB, mg/dL, median (IQR) Highest ALT, units/L, median (IQR) Highest AST, units/L, median (IQR) Lowest albumin, mg/dL, mean SD Highest BUN, mg/dL, mean SD Highest Cr levels, mg/dL, mean SD AKI (n = 25) 14 (56.0) 9.1 3.6 1.9 (0.4-4.2) (n = 21) 18 (72.0) 0 2 (8.0) 2 (8.0) 3 (12.0) 18 (72.0) 1 (4.0) 5 (20.0) 19 (76.0) 20 (80.0) 24 (96.0) 21 (84.0) 8 (32.0) 16 (64.0) 45.1 7.3 23.9 5.4 (n=21) 15 000 (9500-38 500) 39.9 27.9 (n = 23) 68.6 23.9 (n = 23) 3.4 (2.0-8.0) (n = 23) 5.7 (3.6-11.3) (n = 23) 2530 (1470-3252) (n = 24) 9565 (6060-14 099) (n = 24) 2.7 0.6 (n = 24) 64.5 42.2 4.9 2.9 No AKI (n = 50) 23 (46.0) 8.8 3.2 0.2 (-1.5-1.6) 15 (30.0) 7 (14.0) 5 (10.0) 13 (26.0) 19 (38.0) 6 (12.0) <.01 0 40 (80.0) 10 (20.0) 2 (4.0) 3 (6.0) 6 (12.0) 4 (8.0) 0 46.1 5.7 34.2 6.0 30 000 (17 000-43 000) 14.4 4.5 (n = 45) 51.5 20.6 (n = 45) 0.2 (0.1-0.4) (n = 46) 0.4 (0.2-0.7) (n = 46) 117 (33-264) (n = 48) 249 (105-734) (n = 48) 2.8 0.8 (n = 44) 14.7 5.9 0.7 0.3 P* .41 .74 <.01 <.01 <.01

<.01 <.01 <.01 .02 <.01 .52 <.01 .06 <.01 <.01 <.01 <.01 <.01 <.01 .65 <.01 <.01

WSDS, weight standard deviation score; DHF, dengue hemorrhagic fever; PT, prothrombin time; APTT, activated partial thromboplastin time; DB, direct bilirubin; TB, total bilirubin; ALT, serum alanine aminotransferase; AST, serum aspartate aminotransferase; Cr, serum creatinine; BUN, blood urea nitrogen. *Fisher exact test, t test, or Mann-Whitney U test, as appropriate

between patients with DHF with and without AKI with matching for both age and sex. Patients who had AKI had higher WSDS and more obesity (WSDS >2), were more likely to have been referred, and more likely to have had either an initial or nal diagnosis of DHF grade IV. Patients who had AKI had more associated conditions with respiratory failure, liver failure, bleeding requiring blood transfusion, and nosocomial infection and also had a higher mortality rate than DSS patients who had no AKI. Patients with AKI were more likely to have abnormal laboratory ndings than those without AKI, most notably lower levels of lowest hematocrit, higher levels of serum DB, TB, ALT, AST, BUN, and Cr and prolonged APTT and PT (Table I). In multivariate analysis, DHF grade IV and obesity were identied as the major risk factors for AKI (Table II).

Non-oliguric AKI and Oliguric AKI Of the 25 patients with AKI, 24 had urine output data available. Their median urine output in the nadir phase was 0.06 mL/kg/h (IQR, 0.02 to 0.57). Of the 18 urine analyses available, all except 1 had abnormal urine ndings in which urine protein, blood, RBC, WBC, and cast were detected in 17, 17, 16, 15, and 9 cases, respectively. The median levels of urine protein, blood, RBC, WBC, and cast were 3 (IQR, 1 to 3), 3 (IQR, 3 to 3), 15 (IQR, 2.8 to 100), 2 (IQR, 1 to 3), and 0.5 (IQR, 0.0 to 2.2), respectively. The available fractional excretions of sodium (FENa) were high, at 28.1, 2.9, 25.8, 4.0, and 4.7. Oliguric and non-oliguric AKI were identied in 17 and 7 patients whose median urine outputs were 0.05 mL/kg/h (IQR, 0.00 to 0.17) and 0.78 mL/kg/h (IQR, 0.57 to 1.20), respectively. Patients who had oliguric AKI had higher

Table II. Risk Factors for DVI-Caused AKI

Univariate Factor DHF grade IV Obesity AKI (n = 25) n (%) 19 (76.0) 9 (42.9) (n = 21) No AKI (n = 50) n (%) 10 (20.0) 8 (16.0) OR 12.7 3.9 95% CI 4.0-40.0 1.2-12.4 OR 16.9 6.3 Multivariate 95% CI 4.2-68.5 1.4-28.8 P* <.01 <.01

*P value from likelihood ratio test in multivariate analysis.

Outcome of Dengue Hemorrhagic FeverCaused Acute Kidney Injury in Thai Children

305

THE JOURNAL OF PEDIATRICS

www.jpeds.com

Vol. 157, No. 2 which traces of complement component 3 (C3) with a granular pattern at some vessels were found, but no other irregularities (IgA, IgG, IgM, and C1q) were detected. Second-specimen IF revealed no immune complex deposition. Three specimens were examined via electromicroscopy, but only 1 specimen had enough kidney tissue to demonstrate interstitial nephritis and no viral particles were detected.

respiratory failure [16/17 (94.1%) vs 3/7 (42.9%), P = .01] and a higher mortality rate [14/17 (82.4%) vs 1/7 (14.3%), P < .01] than those with non-oliguric AKI. Urine ndings including WBC, RBC, protein and blood were not different between the 2 groups. Although BUN and serum Cr were higher in the oliguric AKI group, the differences were not signicant (data not shown). Causes and Risks of Death in Patients with DHF Who Had AKI Sixteen of the 25 patients (64.0%) who had AKI died; all except 1 had DHF grade IV. The other patient, who had DHF grade III, developed Pseudomonas stutzeri septicemia during the convalescence phase on the 4th day after admission, and although treated with appropriate antibiotics, he died of severe sepsis. The median duration from the rst day of fever until shock developed was 5 days (IQR, 4 to 6). The median duration from the onset of shock until death was a further 5 days (IQR, 1 to 7). Of the 6 patients who died within the rst 48 hours, none had dual or secondary infection, but of the 10 patients who died later than 48 hours after the onset of shock, infections were identied in 7 patients: 5 had pneumonia [Candida albicans (n = 3), Aspergillus (n = 1), and Acinetobacter baumannii (n = 1)] and one each Pseudomonas stutzeri septicemia and A baumannii pneumonia with peritoneal dialysis-related peritonitis from Candida parapsilosis. All fatal cases had respiratory failure, and 15 of 16 (93.8%) had massive bleeding requiring blood transfusions. The most common bleeding site was the gastrointestinal tract (n = 14), with 2 in the respiratory tract and one each intranasal and intracerebral (3 patients had 2 bleeding sites). Univariate analysis failed to indicate any difference between fatal and nonfatal groups in age, sex, WSDS, referral status, liver failure, bleeding disorder, nosocomial infection, platelet count, peak hematocrit, lowest hematocrit, DB, TB, BUN, and serum Cr levels. However, patients in the fatal group had lower median urine output, prolonged mean PT and APTT, higher mean levels of ALT and AST, and lower mean levels of albumin than patients in the survival group. Risk analysis indicated that patients with DHF grade IV, oliguric AKI, respiratory failure, or prolongation of PT or APTT more than twice the control levels had a signicantly increased risk of fatal DVI (Table III). Outcome of AKI in Patients with DHF Who Survived Of the 9 surviving patients, only 3 needed renal replacement therapy (RRT), and none developed chronic kidney disease. One patient had candidemia and responded well to amphotericin B. The median time before Cr levels returned to normal was 32 days (range, 1 to 48), and no other complications were recorded. Renal and liver biopsies were performed in 3 patients and 1 patient, respectively, in all cases with viral particles found in the hepatocyte but not in the renal tissue. In the kidney biopsies performed in 3 patients at 25, 11, and 44 days after AKI developed, interstitial nephritis was found in 2 and only a slightly increased mesangial matrix found in the other. Immunouorescent staining (IF) was done in 2 specimens, in
306

Discussion
In our cases, DVI-related AKI usually resulted from profound shock and was followed by multiple organ failure and death in 64.0% of cases. All patients who did not die fully recovered from the AKI within 1 month. Over the 19-year survey period, DVI-caused AKI was clinically estimated to be 0.9% of DVI patients, compared with 1.2% to 3.3% in earlier similar studies.9,17,18 The rate of DVI-caused AKI in our study and the others should not be considered as representative of the rate of DVI-caused AKI in the general population because our institution and the others are all major tertiary care and referral centers for large populations; thus, it is likely that only more severe cases were included in the studies. Lee et al9 found AKI at a rate of 3.3% (10/304) in adult patients hospitalized with DHF, with a mortality of 60.0%. The differing rates of AKI could be explained at least in part by noting that Lee et al studied adult patients, and in their group 8 of 10 patients who had AKI were older than 65 years and all had an underlying disease such as hypertension, diabetes mellitus, or chronic kidney disease, all of which increase the vulnerability to AKI when having renal hypoperfusion from DSS.9 In addition, concurrent serious bacterial infection, which can precipitate AKI, was found in one third of Lees group.9 In our study, only 2 children had an underlying disease (vesicoureteral reux grade III and G-6-PD deciency), which might have increased their vulnerability, and only 1 had bacteremia before developing AKI. DSS was the major cause of AKI in the Lee study as well as ours. We also found that obesity was another risk factor of DHF-caused AKI, as also noted in previous studies which have found that obese children were more vulnerable to serious DHF than children with normal body weight.19,20 FENa measured during prerenal AKI in DSS should be lower than 1. It is probable that the high FENa assays in our patients were found because their FENa levels were measured after intrinsic AKI developed and they were receiving loop diuretics, which increase FENa levels. In our study, none of the patients with DHF without AKI died, but 64.0% of those with AKI died. The high mortality rate in patients with AKI and DHF was primarily caused by the profound shock these patients developed, leading then to not only AKI but also to hepatic failure, respiratory failure, and massive bleeding, all of which are major causes of death in DHF.5-7,9 Our mortality rate in AKI was similar to the Lee study9 mentioned earlier (64.0% and 60.0%, respectively). We found that patients at increased risk of fatal DHF were more likely to have had DHF grade IV, oliguric AKI,
Laoprasopwattana et al

August 2010

ORIGINAL ARTICLES

Table III. Risk of death in patients with DVI-caused AKI

Factors Sex Male Female Obesity Obese Nonobese WSDS Median (IQR) Initial diagnosis Dengue fever DHF grade I-III DHF grade IV Final diagnosis DHF grade I-III DHF grade IV Urine output Oliguric Nonoliguric Median (IQR), mL/kg/h Respiratory failure Yes No Liver failure Yes No Bleeding disorder Yes No Nosocomial infection Yes No Highest PT >2 times controls #2 times controls Mean SD, seconds Highest APTT >2 times controls #2 times controls Mean SD, seconds Highest ALT >2000 U/L #2000 U/L Mean SD, units/L Highest AST >10,000 U/L #10,000 U/L Mean SD, units/L Lowest albumin <2.5 mg/dL #2.5 mg/dL Mean SD, mg/dL Died (n = 16) 9 7 4 9 1.4 (0.2-3.8) (n = 13) 2 0 14 1 15 14 1 0.04 (0.00-0.17) (n = 15) 16 0 16 0 15 1 7 9 15 0 51.7 28.1 (n = 15) 13 2 80.5 18.4 (n = 15) 11 5 2788 1221 9 7 12 293 6349 6 10 2.6 0.7 Survived (n = 9) 5 4 5 3 3.0 (0.9-6.2) (n = 8) 0 5 4 5 4 3 6 0.78 (0.57-1.20) 4 5 8 1 6 3 1 8 0 8 17.8 3.7 (n = 8) 1 7 46.3 15.8 (n = 8) 3 5 1632 1100 (n = 8) 1 7 6325 4921 (n = 8) 0 8 3.0 0.3 (n = 8) 1.00 (0.16-1) 0.00 (0-0.52) 0.78 (0.52-0.94) 0.17 (0.00-0.64) 0.79 (0.54-0.94) 0.82 (0.57-0.96) 0.14 (0.00-0.57) 0.80 (0.56-0.94) 0.00 (0.00-0.52) 0.67 (0.45-0.84) 0.00 (0-0.97*) 0.71 (0.48-0.89) 0.25 (0.01-0.80) 0.88 (0.47-1.00) 0.53 (0.28-0.77) 1.00 (0.78-1) 0.00 (0-0.37) 0.93 (0.66-1.00) 0.22 (0.03-0.60) 0.79 (0.49-0.95) 0.50 (0.19-0.81) 0.90 (0.55-0.99) 0.50 (0.23-0.77) 1.00 (0.54-1 ) 0.56 (0.31-0.78)

Risk (95% CI) 0.64 (0.35-0.87) 0.64 (0.31-0.89) 0.44 (0.14-0.79) 0.75 (0.43-0.95)

P* 1.00 .20 .35 <.01

.01 <.01 <.01 <.01 .36 .11 .18 <.01 <.01 <.01 <.01 0.20 .03 .08 .02 .07 <.05

WSDS, weight standard deviation score; DHF, dengue hemorrhagic fever; PT, prothrombin time; APTT, activated partial thromboplastin time; ALT, serum alanine aminotransferase; AST, serum aspartate aminotransferase. *Fisher exact test, t test, or Mann-Whitney U test, as appropriate. One-sided, 97.5% condence interval.

respiratory failure, and severe coagulopathy (especially those who had prolongation of PT and APTT more than twice the reference specimens). Another earlier study found a high mortality rate of 72.7% in patients with DHF who had acute respiratory failure, and of the 11 patients in this study who had respiratory failure, 7 (63.6%) also had AKI.7 Most of our patients with oliguric AKI had respiratory failure and a high mortality rate, indicating that volume overload after uid resuscitation in oliguric AKI also precipitated severe respiratory failure subsequently causing death. In addition, cardiomegaly in the chest radiographies was common in patients who had re-

spiratory failure, suggesting that oliguria, volume redistribution to the circulation after shock, and myocardial depression21,22 might precipitate congestive heart failure which contributes to intractable shock. Although several of our patients who died had no second serum sample to conrm the diagnosis of DVI, we believe there is a high probability that all of them had DSS since their clinical proles met the WHO criteria for DSS.15 We found, as in previous studies, that severe DVI is more commonly found in children with secondary DVI and that primary infection can cause severe DVI in infants.23,24
307

Outcome of Dengue Hemorrhagic FeverCaused Acute Kidney Injury in Thai Children

THE JOURNAL OF PEDIATRICS

www.jpeds.com

Vol. 157, No. 2

The authors thank Walailuk Jitpiboon for assistance with data analysis and David Patterson for help with English.
Submitted for publication Jun 3, 2009; last revision received Jan 20, 2010; accepted Feb 4, 2010. Reprint requests: Dr Kamolwish Laoprasopwattana, Department of Pediatrics, Faculty of Medicine, Prince of Songkla University, Hat Yai 90110, Thailand. E-mail: kamolwish@yahoo.com.

DVI-associated AKI without profound shock has a good prognosis with mortality being very rare.11,12,25-27 AKI itself does not increase the risk of fatality in DVI in patients who must have RRT, but profound shock subsequently causing respiratory failure and massive bleeding are the major causes of fatal DVI. More than half of our patients did not receive RRT because most patients had unstable vital signs and/or a severe bleeding disorder, and nearly one fourth of them died within 48 hours of admission. Although most of the AKI in our patients resulted from profound shock, we could not conrm whether the AKI in the patient with DHF grade II with status epilepticus was caused by rhabdomyolysis, nor could we conrm whether DSS or intravascular hemolysis caused AKI in the patient with G-6-PD deciency. Nosocomial-acquired infections were found in nearly one third of our DHF-caused AKI cases, with the most commonly found organisms fungus and A baumannii. Although we could not conrm whether C albicans and A baumannii identied from patients sputum samples were colonized or true organisms causing pneumonia, nonC albicans specimens from sterile sites in 2 patients and Aspergillus from the sputum of the patient with pneumonia indicated that fungal infections were the major nosocomial-acquired infections in these patients. There have been reported cases of patients with DHF who subsequently died from bacteremia,28 aspergillosis,29 and candidiasis,30 indicating that DSS patients who have clinical signs of superimposed infections and do not respond to antibiotics should be investigated for fungal infection, as prompt identication and treatment should reduce the mortality in this group of patients. Dual infection of leptospirosis and DVI has been reported and leptospirosis can cause severe hepatitis and AKI.31,32 However, no leptospirosis was found in any of our patients (although serology was only performed in 8 cases). One study in elderly patients found that AKI was the main risk factor for dual bacterial infection,33 unlike our study, which found nosocomial infections but not dual infections in DHF with AKI. Previous studies have hypothesized that an immune complex or direct DVI in the kidney might play a role in AKI8,13,34; however, of our 3 patients who survived and had a kidney biopsy, tuberointerstitial nephritis was found in 2 specimens, and the specimen from the patient with G-6-PD deciency showed only a slightly increased mesangial matrix. One case had a granular C3 pattern at the glomerular vessels, suggesting that the main cause of AKI in our patients might have been from both profound shock and deposition of immune complex in glomerular vessels. We do note that most of our patients did not have a kidney biopsy, so direct DVI in the kidney cannot be denitely excluded in our analysis. Profound shock and obesity were the major causes of DVIcaused AKI and led to a high mortality rate. Patients at increased risk for fatality were more likely to have had DHF grade IV, oliguric AKI, respiratory failure, or prolongation of PT or APTT more than twice that of the reference specimens. Among the patients who survived, none developed chronic kidney disease and the renal function returned to normal levels within 1 month. n
308

References
1. Vaughn DW, Green S, Kalayanarooj S, Innis BL, Nimmannitya S, Suntayakorn S, et al. Dengue viremia titer, antibody response pattern, and virus serotype correlate with disease severity. J Infect Dis 2000; 181:2-9. 2. Gibbons RV, Vaughn DW. Dengue: an escalating problem. BMJ 2002; 324:1563-6. 3. Anderson KB, Chunsuttiwat S, Nisalak A, Mammen MP, Libraty DH, Rothman AL, et al. Burden of symptomatic dengue infection in children at primary school in Thailand: a prospective study. Lancet 2007;369:1452-9. 4. Guzman MG, Kouri G, Bravo J, Valdes L, Vazquez S, Halstead SB. Effect of age on outcome of secondary dengue 2 infections. Int J Infect Dis 2002;6:118-24. 5. Ong A, Sandar M, Chen MI, Sin LY. Fatal dengue hemorrhagic fever in adults during a dengue epidemic in Singapore. Int J Infect Dis 2007;11: 263-7. 6. Lee IK, Liu JW, Yang KD. Clinical and laboratory characteristics and risk factors for fatality in elderly patients with dengue hemorrhagic fever. Am J Trop Med Hyg 2008;79:149-53. 7. Wang CC, Liu SF, Liao SC, Lee IK, Liu JW, Lin AS, et al. Acute respiratory failure in adult patients with dengue virus infection. Am J Trop Med Hyg 2007;77:151-8. 8. Futrakul P, Poshyachinda V, Mitrakul C, Kun-Anake C, Boonpucknavig V, Boompucknavig S, et al. Renal involvement and reticulo-endothelial-system clearance in dengue hemorrhagic fever. J Med Assoc Thai 1973;56:33-9. 9. Lee IK, Liu JW, Yang KD. Clinical characteristics, risk factors, and outcomes in adults experiencing dengue hemorrhagic fever complicated with acute renal failure. Am J Trop Med Hyg 2009;80:651-5. 10. Wiersinga WJ, Scheepstra CG, Kasanardjo JS, de Vries PJ, Zaaijer H, Geerlings SE. Dengue fever-induced hemolytic uremic syndrome. Clin Infect Dis 2006;43:800-1. 11. Lim M, Goh HK. Rhabdomyolysis following dengue virus infection. Singapore Med J 2005;46:645-6. 12. Davis JS, Bourke P. Rhabdomyolysis associated with dengue virus infection. Clin Infect Dis 2004;38:e109-11. 13. Jessie K, Fong MY, Devi S, Lam SK, Wong KT. Localization of dengue virus in naturally infected human tissues, by immunohistochemistry and in situ hybridization. J Infect Dis 2004;189:1411-8. 14. Vachvanichsanong P, Dissaneewate P, Lim A, McNeil E. Childhood acute renal failure: 22-year experience in a university hospital in southern Thailand. Pediatrics 2006;118:e786-91. 15. Anonymous. Dengue Hemorrhagic Fever, Diagnosis, Treatment and Control. Geneva, Switzerland: World Health Organization; 1997. 16. Chan JC, Williams DM, Roth KS. Kidney failure in infants and children. Pediatr Rev 2002;23:47-60. 17. Khan NA, Azhar EI, El-Fiky S, Madani HH, Abuljadial MA, Ashshi AM, et al. Clinical prole and outcome of hospitalized patients during rst outbreak of dengue in Makkah, Saudi Arabia. Acta Trop 2008;105:39-44. 18. Mendez A, Gonzalez G. Abnormal clinical manifestations of dengue hemorrhagic fever in children. Biomedica 2006;26:61-70. 19. Pichainarong N, Mongkalangoon N, Kalayanarooj S, Chaveepojnkamjorn W. Relationship between body size and severity of dengue hemorrhagic fever among children aged 0-14 years. Southeast Asian J Trop Med Public Health 2006;37:283-8. 20. Kalayanarooj S, Nimmannitya S. Is dengue severity related to nutritional status? Southeast Asian J Trop Med Public Health 2005;36:378-84.

Laoprasopwattana et al

August 2010
21. Khongphatthanayothin A, Lertsapcharoen P, Supachokchaiwattana P, La-Orkhun V, Khumtonvong A, Boonlarptaveechoke C, et al. Myocardial depression in dengue hemorrhagic fever: prevalence and clinical description. Pediatr Crit Care Med 2007;8:524-9. 22. Kabra SK, Juneja R, Jain Y, Madhulika, Singhal T, Dar L, et al. Myocardial dysfunction in children with dengue haemorrhagic fever. Natl Med J India 1998;11:59-61. 23. Guzman MG, Alvarez M, Rodriguez R, Rosario D, Vazquez S, Vald s L, et al. Fatal dengue hemorrhagic fever in Cuba, 1997. Int J Infect Dis 1999; 3:130-5. 24. Wichmann O, Hongsiriwon S, Bowonwatanuwong C, Chotivanich K, Sukthana Y, Pukrittayakamee S. Risk factors and clinical features associated with severe dengue infection in adults and children during the 2001 epidemic in Chonburi, Thailand. Trop Med Int Health 2004;9:1022-9. 25. Gunasekera HH, Adikaram AV, Herath CA, Samarasinghe HH. Myoglobinuric acute renal failure following dengue viral infection. Ceylon Med J 2000;45:181. 26. Tanphaichitr VS, Chonlasin R, Suwantol L, Pung-Amritt P, Tachavanich K, Yogsan S, et al. Effect of red blood cell glucose-6phosphate dehydrogenase deciency in patients with dengue hemorrhagic fever. J Med Assoc Thai 2002;85(Suppl 2):S522-9. 27. Lima EQ, Gorayeb FS, Zanon JR, Nogueira ML, Ramalho HJ, Burdmann EA. Dengue haemorrhagic fever-induced acute kidney injury

ORIGINAL ARTICLES
without hypotension, haemolysis or rhabdomyolysis. Nephrol Dial Transplant 2007;22:3322-6. Lahiri M, Fisher D, Tambyah PA. Dengue mortality: reassessing the risks in transition countries. Trans R Soc Trop Med Hyg 2008;102: 1011-6. Kohli U, Sahu J, Lodha R, Agarwal N, Ray R. Invasive nosocomial aspergillosis associated with heart failure and complete heart block following recovery from dengue shock syndrome. Pediatr Crit Care Med 2007;8: 389-91. Suzuki S, Kitazawa T, Ota Y, Okugawa S, Tsukada K, Nukui Y, et al. Dengue hemorrhagic shock and disseminated candidiasis. Intern Med 2007; 46:1043-6. Bruce MG, Sanders EJ, Leake JA, Zaidel O, Bragg SL, Aye T, et al. Leptospirosis among patients presenting with dengue-like illness in Puerto Rico. Acta Trop 2005;96:36-46. Guron G, Holmdahl J, Dotevall L. Acute renal failure after a holiday in the tropics. Clin Nephrol 2006;66:468-71. Lee IK, Liu JW, Yang KD. Clinical characteristics and risk factors for concurrent bacteremia in adults with dengue hemorrhagic fever. Am J Trop Med Hyg 2005;72:221-6. Boonpucknavig V, Bhamarapravati N, Boonpucknavig S, Futrakul P, Tanpaichitr P. Glomerular changes in dengue hemorrhagic fever. Arch Pathol Lab Med 1976;100:206-12.

28.

29.

30.

31.

32. 33.

34.

Outcome of Dengue Hemorrhagic FeverCaused Acute Kidney Injury in Thai Children

309