A Prospective Randomized Controlled Study of Two

Original Article

A Prospective Randomized Controlled Study of Two Fluid Regimens in the Initial Management of Septic Shock in the Emergency Department Indumathy Santhanam, MD,* Shanthi Sangareddi, MD,y Shekhar Venkataraman, MD,z Niranjan Kissoon, MD,x Vaidehi Thiruvengadamudayan, MD,k and R. Kulandhai Kasthuri, MDy

Objectives: To compare the impact of 40 mL/kg of fluid over 15 minutes followed by dopamine and further titration of therapy to achieve therapeutic goals (study protocol) versus 20 mL/kg over 20 minutes up to a maximum of 60 mL/kg over 1 hour followed by dopamine (control protocol) in septic shock. Design and Setting: Prospective randomized controlled study in the emergency department of a public hospital in India. Patients: One hundred forty-seven children older than 1 month presenting with septic shock were enrolled into the study. Outcome Measures: Hospital mortality (primary outcome), 72-hour survival, achievement of therapeutic goals of shock resolution, incidence of hypoxia, hepatomegaly, intubation at 20, 40, and 60 minutes (secondary outcomes) were compared between the arms. Results: Seventy-four and 73 children were assigned to the study and control group, respectively. Overall mortality was 17.6%, 26 deaths with 13 in each arm. Mortality in the study cohort was lower than our historical mortality of 50% (P G 0.0001), 95% confidence interval (CI), 11.9Y24.8. Cumulative survival at 72 hours was 72.5% (95% CI, 58.9Y86.1) and 77.6% (95% CI, 66.0%Y89.2%) in the control and study groups, respectively. Resolution of shock in the emergency department was associated with survival odds ratio (OR) 9.2 (95% CI, 2.1Y40.8). Rapidity of achieving therapeutic goals was not significantly different between groups. Intubation rates were also the same (46.5% in the control group versus 55% in the study group; P = 0.28). At 20 minutes, 35.6% of the control group and 70% of the study group had hepatomegaly (P G 0.01). Conclusion: There was no difference in the overall mortality, rapidity of shock resolution, or incidence of complications between the groups. The occurrence of hepatomegaly at 20 minutes following 40 mL/kg is of concern in settings with limited access to post-resuscitation ventilator care.

*Pediatric Emergency Medicine, yPediatric Intensive Care Department, Institute of Child Health and Hospital for Children, Madras Medical College, Chennai, India; zDepartment of Critical Care Medicine and Pediatrics, University of Pittsburgh School of Medicine, Pittsburgh, PA; ‘Acute and Critical Care Programs, Department of Pediatrics, University of British Columbia, Vancouver, British Columbia, Canada; and kDepartment of Microbiology, Sundaram Medical Foundation, Rangarajan Memorial Hospital, Chennai, India. Address correspondence and reprint requests to Niranjan Kissoon, MD, CPE, University of British Columbia, British Columbia Children’s Hospital, Rm K4-105, 4480 Oak St, Vancouver, British Columbia V6H 3V4, Canada. E-mail: [email protected]. Copyright * 2008 by Lippincott Williams & Wilkins ISSN: 0749-5161/08/2410-0647

Key Words: septic shock, pulmonary edema, resuscitation, therapeutic goals, mortality

S

erious infections are a major cause of mortality in the developing world, and shock complicates many of these cases. Fifty percent of deaths occur within the first 24 hours of admission.1 Although it is well known that time-sensitive, goal-directed methods to correct shock in the first hour improve survival in severe sepsis,2Y4 provision of emergency advanced life support on arrival in these countries is handicapped by lack of timely laboratory services.5 Before 2002, the Pediatric Advanced Life Support (PALS) suggested an approach to correct shock.6 In 2002, the American College of Critical Care Medicine published a stepwise approach to normalized blood pressure (BP) and peripheral perfusion within the first hour of presentation in septic shock.7 However, there was little data in India to support the use of either protocol. The Institute of Child Health is a 537 bedded referral public hospital with 14 intensive care beds providing free care to 700,000 children who register at the outpatient department every year. Every year, 5000 to 6000 children require resuscitation at the emergency department (ED). A previous study from this institution showed that patients presenting late with shock and respiratory failure had high mortality rates.8 Equipped with resuscitation equipment and manned round-the-clock by a trained emergency team, our ED reflects the realities faced by a hospital catering to overwhelming volumes of critically ill children in the developing world where high-technology equipment and invasive monitoring is limited. Faced with these realities, we designed a protocol which relied entirely on clinical assessment, pulse oximetry, electrocardiogram, and urine output to guide therapy during the correction of shock as compared with the American College of Critical Care Medicine guidelines which suggest invasive monitoring. On the basis of a pilot study that showed an intubation rate of 80% when up to 60 mL/kg of Ringers Lactate (RL) was administered in 15 minutes to correct shock, the study protocol was modified to administer up to 40 mL/kg of RL in 15 minutes. The objective of our study was to determine whether the study group (40 mL/kg of RL over 15 minutes followed by dopamine) would achieve similar resolution of shock, less need for ventilation and similar outcomes as compared with the control protocol (up to 60 mL/kg over an hour, followed by dopamine). We hypothesized that there would be

Pediatric Emergency Care Volume 24, Number 10, October 2008

Copyright @ 2008 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

647


Santhanam et al

no differences in outcomes and less need for ventilation with slower fluid administration.

This study was approved by the institutional review board of Madras Medical College to which the Institute of Child Health is affiliated.

of metabolism, drug toxicity, trauma, burns, stridor, near fatal asthma, prehospital fluid resuscitation, grade 3 malnutrition, chronic systemic co-morbidities, genetic disorders, malignancies, immunocompromised conditions, human immunodeficiency virus, do not resuscitate orders, physician’s decision not to treat, and cardio pulmonary arrest before arrival or within the first hour of resuscitation.

Participants

Interventions

Septic shock was defined using the Sepsis Consensus Conference definition.9 Healthy children aged between 1 month and 12 years who were triaged as septic shock at the outpatient department (during the hours when the principal investigator [PI] was available) were enrolled in to the study. The criteria for exclusion were the following: age younger than 30 days, shock due to hypovolemia, hemorrhage, anaphylaxis, envenomation, diabetic ketoacidosis, inborn errors

Enrolled patients were resuscitated on arrival based on the PALS guidelines. A stop clock used for time was used to guide fluid therapy. The rapid cardiopulmonary assessment evaluated the following variables before and after each aliquot of 20 mL/kg: airway patency, respiratory rate, grunt, retractions, abdominal respiration, air-entry, adventitious sounds, color, heart rate, difference in pulse volume between femoral and dorsalis pedis arteries, core-peripheral temperature gradient, capillary refill

METHODS

FIGURE 1. Study protocol in the ED.

648

* 2008 Lippincott Williams & Wilkins



time (CRT), BP, liver span,10 mental status (Alert, Voice Response, Pain Response and Unresponsive Scale),6,11 pupillary response for inequality, and urine output by void or catheterization. The upper border of the liver was identified by percussion and the lower border by palpation and marked using a pen. The span was measured in the midclavicular line using a tape and was documented. Following each fluid bolus, the lower border was re-assessed and measured.

Therapeutic Goals For treatment purposes, the attainment of the following therapeutic goals of shock resolution is needed to be achieved for our study: normal systolic BP for age, warm peripheries, CRT shorter than 2 seconds and urine output greater than 1 mL/kg. Because drugs used for intubation caused tachycardia and unresponsiveness, alertness and heart rate could not be used for defining shock resolution at 60 minutes in all patients.

Fluid Regimens in the Initial Management of Septic Shock

Study Protocol Twenty milliliters per kilogram up to 40 mL/kg of RL was administered in 15 minutes using a 3-way stop cock and a rapid pull-push method. If therapeutic goals were not attained after 40 mL/kg, dopamine infusion was started and maintained at 10 2g/kg/min. A minimal volume of 20 mL to a maximal volume required to correct therapeutic goals was planned in the study protocol. If pulmonary edema (PE) or hepatomegaly were noted during fluid therapy, further fluids were stopped, and intubation was performed followed by manual ventilation. If shock persisted after intubation but PE and hepatomegaly had resolved, smaller aliquots were pushed until shock resolved or PE or hepatomegaly recurred (Fig. 1). At 45 minutes, if hypotensive shock had not resolved, catecholamine infusion was initiated. If shock persisted, but BP was high, dobutamine was started.

FIGURE 2. Control protocol in the ED. * 2008 Lippincott Williams & Wilkins


649

Santhanam et al


If the child remained hypotensive at 60 minutes, 2 mg/kg of hydrocortisone was administered.

Control Protocol Twenty milliliters per kilogram of RL was administered over 20 minutes over 20 minutes using a 3 way stopcock and a rapid pull push method. If therapeutic goals were not achieved, a further aliquot was administered up to 60 mL/kg at 1 hour. If shock persisted at 60 mL/kg, dopamine was started but further fluids were not administered in the emergency department. Intubation and ventilation using bag valve mask device was performed if signs of PE or hepatomegaly

were noted at 20, 40, or 60 mL/kg. If shock persisted after intubation and PE and hepatomegaly had resolved, 1 more aliquot of 20 mL/kg was administered (Fig. 2). During fluid administration, children were monitored for development of PE: airway instability, pink froth, cough, decreased or increased respiratory rates, grunt, retractions, abdominal respirations, new rales, drop in saturations, gallop, hepatomegaly, agitation, fighting the mask and combativeness. If any of these signs was noted, further fluids were stopped, dopamine infusion was initiated, and intubation was performed. Other indications for intubation were the following: respiratory failure on arrival, hypotension and seizures refractory to both lorazepam and phenytoin loading.

FIGURE 3. Consort diagram outlining flow of study participants.

650





TABLE 1. Distribution of Admission Variables Between the Two Groups Variable

Control Group (N = 73)

Study Group (N = 74)

65.8 57.5 83.6 80.8 27.4 15.1 27.4 56.2

78.6 54 83.8 85.1 13.5 13.5 13.5 70.3

Incessant cry Lethargy Unresponsiveness Posturing Seizures AWD Vomiting Respiratory distress

Percentile Difference

95% CI of Percentile Difference

P

(j2, +27) (j20, +13) (j12, +12) (j8, +16) (j26.7, +1) (j12.9, +9.8) (j26.7,1) (j1, +30)

0.28 0.74 1.00 0.52 0.04 0.81 0.73 0.09

12.8 j3.5 0.2 4.3 13.9 1.6 j13.9 14.1

Control group indicates standard fluid resuscitation; intervention group, aggressive fluid resuscitation; AWD, acute watery diarrhea; other variables on admission were also not significantly different between the 2 groups (data not shown). P (Fisher exact test).

Each assessment was performed independently, both by the PI and a resident to reduce observer bias. If disagreements occurred, consensus was sought after reassessment.

Ongoing Care for Both Groups Ketamine, atropine, and succinylcholine were used to facilitate intubation. Children presenting with seizures were intubated using drugs to avoid intracranial hypertension. Documented hypoglycemia and hypocalcaemia were corrected. Anticonvulsant drugs for seizures, bronchodilator nebulizations, and hydrocortisone for asthma exacerbations were given according to standard protocols. Blood was collected for a complete blood cell count and biochemistry. Blood and body fluids were collected separately for culture. Intravenous 150 to 200 mg/kg of cefotaxime and ampicillin were given in infants younger than 3 months, whereas Cefotaxime was given to older children on arrival to the wards or intensive care unit. Most children had already received prehospital parenteral antibiotics. Obvious foci of sepsis were drained by surgeons while resuscitation was in progress. All children were planned to be transferred out of the ED after 1 hour.

Post-ED Care All intubated children were manually ventilated by attendants until ventilators were available. Post-ED care was

classified as not needing mechanical ventilation, mechanical ventilator needed and available within 6 hours or extubated within 6 hours and mechanical ventilator not available within 6 hours. Once admitted, treatment of fluid refractory dopamine-resistant shock was similar in both groups.

Outcomes The primary outcome was hospital survival. The secondary outcomes were peripheral warmth, CRT shorter than 2 seconds, normal BP for age and urine output 91 mL/kg/h. Complications evaluated were hypoxia, hepatomegaly, and intubation at 20, 40, and 60 minutes for each regimen.

Sample Size Calculation Hospital mortality for septicemia between 2000 and 2003 was 50%.12 Assuming the latter as control, for a reduction by 25% (a 50% reduction) with an ! error of 0.05 and a power of 0.8, a sample of 60 were needed to be enrolled in each arm. To account for exclusion of patients after enrollment, 20% more patients were enrolled increasing the total to 80 patients in each arm.

Randomization Informed consent was obtained verbally, followed by written consent in Tamil from the parent. The children were randomly assigned to either goal-directed therapy (study) or to control therapy.

TABLE 2. Cumulative Fluids and Outcomes Between the Two Treatment Groups Control Group

Study Group

Variable

n

Median

25thY75th Quartile

n

Median

25thY75th Quartile

P

Total fluids (mL/kg) Hospital stay in hours Ventilator hours Inotrope duration in hours Albumin (g/dL)

73 73 32 72 54

60 54.5 10 9 3.1

60Y60 30.8Y136.5 5Y28 3.5Y17.5 2.6Y3.5

74 74 38 74 54

72.5 69.5 24 7 3.1

60Y90 43Y119.8 7Y40.8 1Y19 2.6Y3.4

0.00012 0.66 0.2 0.36 0.76

P (Mann-Whitney U rank sum test).



651


Santhanam et al

TABLE 3. Comparison of the Number of Children Who Achieved Therapeutic Goals at 20, 40, and 60 Minutes Between the Two Arms Clinical Parameters in Minutes

Study (N = 74)

Control (N = 73)

Shock resolution 20 9 9 40 38 39 60 62 61 Normalization of CRT 0 55 54 20 64 66 40 73 68 60 74 73 Normalization of core-peripheral temperature gap 0 53 58 20 49 51 40 66 68 60 Normalization of urine output 0 0 0 20 20 16 40 54 53 60 62 62 Normalization of systolic blood pressure 0 63 63 20 67 69 40 68 72 60 72 71

Significance 0.99

0.98

Statistical Methods Statistical Package for the Social Sciences for Windows (version 10) was used. Differences between the 2 groups were tested with Student t test, Pearson # 2 test, Fisher exact test and Wilcoxon rank sum test. A P G0.05 was considered significant. A multivariate analysis was performed using survival as the dependent variable and age, sex, treatment groups, total fluids administered, presenting symptoms, and signs and the level of post-ED care as independent variables. Variables that had a P less than 0.05 were included in the model. The adequacy of fit for the final multiple variable model was evaluated using the Hosmer-Lemeshow test.

RESULTS Flow of Participants: Consort Diagram 0.99

During the study period, 416 patients were eligible but 256 were excluded, leaving 160 to be randomized (Fig. 3).

Recruitment 0.84

This study was conducted between November 2003 and December 2004. Patients were followed up until discharge or death in the hospital.

Baseline Characteristics 0.99

Among the 160 patients enrolled in the study, 147 were included for the intention to treat analysis after exclusion of 13 (8.1%) (Fig. 3). There were no differences in the baseline demographic or clinical characteristics between 74/147 study group patients and 73/147 control group patients. One hundred twenty-two children, 63 in the study protocol and 59 in the standard protocol group: (P = 0.7) received antibiotics before arrival in the ED (Table 1).

Sequence Generation Random numbers were generated using randomization tables of blocks of eight. It was prepared separately for 2 strata of severity: compensated and de-compensated shock at arrival.

Allocation Concealment Sealed, opaque, randomly assorted envelopes were opened by a registered nurse who was not part of the study team. The A or B therapy was initiated by the nurses only after announcement of the study assignments.

Implementation The bio-statistician prepared the randomization.

Blinding Participants were unaware of the study group assignments. The nurse who administered fluids was aware of the study assignment but did not assess patients or influence therapeutic decisions. The PI was not blinded. The residents in the ED and the physicians in the wards were not aware that a study was in progress or the study-group assignments. The PI did not influence patient management after transfer. The epidemiologist who performed the analysis was also blinded.

652

TABLE 4. Comparison of the Number of Children Who Developed Desaturation, Tracheal Intubation, and Hepatomegaly at 20, 40, and 60 Minutes Between the Two Arms Clinical Parameters in Minutes Desaturation On arrival 20 40 60 Tracheal intubation 20 40 60 Hepatomegaly at arrival 20 40 60

Study (N = 74)

Control (N = 73)

17 22 14 3

24 12 13 1

0.18

1 17 37 28 52 43 24

0 10 28 23 26 32 25

0.61

Significance

0.42 0.001 0.18 0.1




Administered Therapy The volume of fluids administered was significantly higher in the study group than in the control group (P 9 0.01). Steroids were administered in 7 (9.6%) and 4 (5.4%) in the control and study group respectively. There was no difference in the intubation rate between the 2 groups (46.5% in the control group vs 55% in the treatment group; P = 0.28 by # 2 test). There were no differences in the duration of shock requiring inotropes, ventilation, and hospital stay between the 2 arms (Table 2).

Therapeutic Response Normalization of CRT, core-peripheral temperature gap, urine output, and BP at 20, 40, and 60 minutes were not significantly different between groups. At 20 minutes, only 9 (12.1%) patients in the study group and 9 (12.3%) in the control group had resolution of shock (P = 1.0 by Fisher exact test). At 40 minutes, 38 (51.4%) patients in the study group and 39 (53.4%) in the control group had resolution of shock (P = 1.0 by Fisher exact test). By 60 minutes, 62 (83.8%) patients in the study group and 63 (83.8%) patients in the control group had resolution of shock (1.0 by Fisher exact test) (Table 3).

Complications In the control group, 31% of the patients had hepatomegaly on arrival compared with 39.2% in the study group (P = 0.39 by Fisher exact test). At 20 minutes, 26/73 (35.6%) of the control group and 52/74 (70%) of the study group had hepatomegaly (P G 0.01). At 60 minutes, the incidence of hepatomegaly was 34.2% and 32.4% in the control and study groups, respectively. Likewise, the incidence of hypoxia and intubations at 20, 40, and 60 minutes was the same in both groups (Table 4).

Survival Analysis There were 26 deaths, with 13 deaths in each group (P = 1.0) and an overall mortality rate of 17.6% (95% CI -

FIGURE 4. Patient survival in relationship to need for respiratory support. A, No mechanical ventilation needed, n = 77. B, Ventilator available within 6 hours, n = 12. C, Manual ventilation for 6 hours or less; no ventilator, n = 21. D, Manual ventilation for longer than 6 hours, n = 37.


11.9Y24.8%). Compared with our historical mortality rate of 50%, the observed mortality of 17.6% was significantly lower (P G 0.0001 by Exact binomial test).

Kaplan-Meier Analysis for the First 72 Hours of Hospital Stay Cumulative survival at 72 hours was 72.5% (95% CI, 58.9Y86.1) and 77.6% (95% CI, 66.0Y89.2) in the control and study groups, respectively (P = 0.71 by log-rank test). The relative risk of death for the study group compared with the control group was 0.94 (95% CI, 0.77Y1.15) and the OR for death was 0.79 (95% CI, 0.38Y1.66). The hazards ratio for the study group compared with the control group was 0.81.

Survival Based on Post-ED Care The OR for survival if shock was corrected in the ED was 9.2 95% CI, 2.1Y40.8. Survival was 100% if mechanical ventilation was not needed and 62% if ventilation was needed (P = 0.0001 by Fisher exact test; OR, 92.3; 95% CI, 8.7Y492.3). Survival was 83.3% if a mechanical ventilator was available within 6 hours after intubation, 76.2% if extubated within 6 hours, and 48.6% if a ventilator was not available after the initial 6 hours (Fig. 4).

DISCUSSION We conducted this study comparing fluid regimens because of the need for clinical assessment to guide therapy and the lack of mechanical ventilators in resource limited environments. We found no difference between the 2 groups in the following outcome parameters: (1) shock resolution in the first hour, (2) hospital mortality, and (3) the incidence of intubation. However, the incidence of hepatomegaly was greater at 20 minutes in the study arm compared with the control arm. In 80% to 85% of children who had reversal of shock in the ED, there was a 9-fold increase in the odds for survival. Patients who had shock resolution without the need for ventilation had a 100% survival rate. Patients whose shock was not reversed in the ED and required ventilation had the highest mortality. Thus, the survival advantage conferred by reversing shock in severe sepsis in the ED cannot be understated in resource poor settings. However, overaggressive fluid administration needs to be avoided because of lack of support for the child who develops respiratory failure. Compared with historical controls, our overall mortality rate was much lower. The pretrial mortality was high because recognition of septic shock was late (usually hypotensive), time-sensitive fluid administration guided by specific therapeutic goals was not practiced, and intubations were performed late (virtually in cardio pulmonary arrest). The comparison with the historical mortality is used to highlight the fact that both regimes were similar to each other but different from our previous mortality rate. Compared with published mortality rates in septic shock from other Indian hospitals, Delhi 50%, Ludhiana 47%, and Chandigarh 54.6%,13Y15 our mortality rate appears lower. Western data have shown a steep decline in mortality rate of 0% to 10% in sepsis.16Y19 Improvement in outcomes in developed nations has been attributed to the cumulative



653


Santhanam et al

benefits of specialist retrieval systems and advanced intensive care facilities.19,20 In contrast, in our setting, children are brought late by parents with little prehospital resuscitation. Despite these challenges, our study has shown that a syndromic approach to the management of septic shock by trained front-line staff 5 with administration of 60 to 80 mL/kg fluid, inotrope infusion, correction of hypoglycemia, early intubation, and greater use of physical examination therapeutic end-points has had a favorable impact on the overall survival.21 The major disadvantage of administration of large volumes of fluids to correct shock is the development of PE. In our study, the overall need for intubation rate was 40% to 50% with no difference between the protocols. Although the overall change in liver size at the end of therapy was not significantly different in the 2 groups, the greater incidence of hepatomegaly at the end of 20 minutes for children who received 40 mL/kg was of concern in our setting where access to mechanical ventilators was limited. Indeed, this may be a greater problem than anticipated in developing countries where malnutrition may complicate sepsis and render children more sensitive to rapid fluid administration. The need for ventilation in 45% of children in the ED suggests that a severe form of the disease exists at arrival. Seventy-nine percent of children who did not have access to mechanical ventilation and were intubated for less than 6 hours survived compared with 83% who were intubated and had access to mechanical ventilation within 6 hours. Survival decreased to 49% in children who did not have access to mechanical ventilation even after 6 hours (Fig. 4). These data support the need for increased availability of ventilators to improve post resuscitation survival. There have been concerns whether assessment of bedside features of shock were reproducible between different observers.11 This concern was eliminated by ensuring that the PI was solely responsible for interpreting the cardiopulmonary assessment. Another study suggested that delayed CRT should not be taken as an isolated variable of shock.22 In our study, achievement of 4 therapeutic goals was taken as shock resolution. Clinical monitoring without high-technology facilities has shown a dramatic reduction in mortality in Vietnamese children presenting with moderate dengue shock syndrome.5,23 Our study has also demonstrated the impact of meticulous and serial cardiopulmonary assessment on mortality in septic shock. Our study has several limitations. First, we did not study interrater variability in a trial which depended so much on clinical assessment. The calculation of mortality using historical controls did not adjust for the presence of co-morbidities. We did not quantify for severity of illness using the Pediatric Risk of Mortality Scoring23 Besides, we did not control post-ED care or have access to serum lactate or central venous oxygen saturation estimations.

CONCLUSIONS Our study shows that resolution of shock or a reduction in mortality in severe sepsis can be achieved with either

654

regime. Though the overall complication rate was the same for both protocols, the greater occurrence of hepatomegaly following 40 mL/kg at 20 minutes was of concern in our setting where children had limited access to post-resuscitative ventilator support. ACKNOWLEDGMENTS The authors thank Dr Sarada Suresh MD, Professor of Pediatrics ICH, for mentoring, Dr Arjun Rajagopalan AB, Chief of Medical Staff, Sundaram Medical Foundation and Mr. PCM Santhanam for subsidizing cost of laboratory services, Dr Ramesh Dorairajan MD, Dr Suchitra Ranjit MD and Dr Ram Rajagopalan AB, J.A Carcillo MD for reviewing the manuscript, Ms Varshini R, Bragadeeshwaran for data entry, Dr Kumarasamy MD, Director of Immunology, Dr Venkatesan PHD, Director of Biostatistics and bioinformatics, Dr Fathima Beevi PHD, Deputy Director, Tuberculosis Research Center, Indian Council of Medical Research-Chennai and Dr Akila Bridgette, Epidemiologist, Sundaram Medical foundation for bio-statistical support and the residents and nurses in the emergency and intensive care department of the Institute of Child Health-Chennai during the study period. REFERENCES 1. Robertson MA, Molyneux EM. Description of serious illness and outcome in patients identified using ETAT guidelines in urban, Malawi. Arch Dis Child. 2001;85:214Y217. 2. Carcillo JA, Davis AL, Zaritsky A. Role of early fluid resuscitation in pediatric septic shock. JAMA. 1991;266:1242Y1245. 3. Rivers E, Nguyen B, Havstad S, et al. Early goal directed therapy collaborative group: early goal directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med. 2001;345:1368Y1377. 4. Han YY, Carcillo JA, Draggotta MA, et al. Early reversal of pediatricVneonatal septic shock by community physicians is associated with improved outcome. Pediatrics. 2003;112:793Y799. 5. Molyneux EM, Maitland K. Intravenous fluids-getting the balance right. N Engl J Med. 2005;353:941Y944. 6. Zaritsky AL, Nadkarni VM, Hickey RW et al, eds. Textbook of Pediatric Advanced Life Support. Dallas,TX: American Heart Association; 2002. 7. Carcillo JA, Fields A. Task force committee members: clinical parameters for hemodynamic support of pediatric and neonatal patients in septic shock. Crit Care Med. 2002;30:1365Y1377. 8. Santhanam I, Pai M, Kasthuri KR, et al. Mortality after admission in the pediatric emergency department: a prospective study from a referral children’s hospital in Southern India. Pediatr Crit Care Med. 2002;3: 358Y363. 9. Levy MM, Fink MP, Marshall JC, et al. 2001 SCCM/ESICM/ACCP/ ATS/ SIS International Sepsis Definitions Conference. Crit Care Med. 2003;31:1250Y1256. 10. Swash M. Hutchinson’s Clinical Methods. 21st Ed. Philadelphia, PA: W.B. Saunders; 2001:141Y145. 11. Otieno H, Were E, Ahmed I, et al. Are bedside features of shock reproducible between different observers? Arch Dis Child. 2004; 89:977Y979. 12. Annual Hospital Statistics: Institute of Child Health and Hospital for Children, Egmore, Chennai-600008; Compiled and published by the Medical Records Department, (Not for circulation) 2001:37;2002: 39;2003:61. 13. Sarthi M, Lodha R, Vivekananadan S, et al. Adrenal status in children with septic shock using low dose stimulation test. Pediatr Crit Care Med. 2007;8(1):23Y28. 14. Singh D, Chopra A, Pooni PA, et al. A clinical profile of shock in children in Punjab, India. Indian Pediatr. 2006;43:619Y623.



Pediatric Emergency Care Volume 24, Number 10, October 2008 15. Jayshree M, Gehlot S, Singhi S. Predictors of multiple organ system failure and death in children with severe sepsis. Pediatr Crit Care Med. 2007;8(suppl 3):A59. 16. Angus DC, Linde-Zwirble WT, Lidicker J, et al. Epidemiology of severe sepsis in the United States: analysis of incidence, outcomes and cost of care. Crit Care Med. 2001;29:1303Y1310. 17. Stoll BJ, Hollman RC, Schuchat A. Decline in sepsis-associated neonatal and infant deaths in the United States, 1979 through 1994. Pediatrics. 1998;102:18. 18. Kutko MC, Calarco MP, Flaherty MB, et al. Mortality rates in pediatric septic shock with and without multiple organ system failure. Pediatr Crit Care Med. 2003;4:333Y337. 19. Booey R, Habibi P, Nadel S, et al. Meningococcal research group:

20. 21. 22. 23.

Fluid Regimens in the Initial Management of Septic Shock reduction in case fatality rate from meningococcal disease associated with improved health care delivery. Arch Dis Child. 2001;85: 386Y390. Parker MM, Hazelet JA, Carcillo J. Pediatric consideration. Crit Care Med. 2004;32(suppl 11):S591YS594. Pamba A, Maitland K. Capillary refill: prognostic value in Kenyan children. Arch Dis Child. 2004;89:950Y955. Wills BA, Dung NM, Dong TH, et al. Comparison of three fluid solutions for resuscitation in dengue shock syndrome. N Engl J Med. 2005;353:877Y8894. Maulen-Radovan I, Gutie´rrez Casterello´n P, Zaldo R. PRISM score evaluation to predict outcome in pediatric patients on admission at an emergency department. Arch Med Res. 1996;27:553Y558.



655