This observational-analytical prospective study was carried out in the Medical ICU of a tertiary care teaching hospital from January 2008 to June 2009. The study included 84 patients admitted to the ICU with severe sepsis and septic shock.

Patients admitted in Medical ICU meeting the diagnostic criteria for severe sepsis and septic shock during the first 24 h of admission.

The patients were grouped according to the aetiology of their illness e.g. tropical infection, respiratory infection, genito-urinary infection etc. A thorough clinical examination was carried out in all patients. Total leucocyte count, Packed Cell Volume (PCV), serum creatinine, serum bilirubin, serum sodium/potassium/bicarbonate and arterial pH was done in each patient.

For calculating APACHE-IV score, following data was collected-

1. Acute Physiology Points-Both worst and best value in the first 24 h of admission to ICU was collected for a) Temperature b) Pulse c) Blood Pressure d) Respiratory Rate e) Haematocrit f) Total Leucocyte Count g) Serum Blood Urea Nitrogen (BUN) h) Serum creatinine i) Serum sodium.

Single worst value in first 24 h was collected for a) FiO₂ b) PaO₂ c) PCO₂ d) pH e) Urine output f) Serum albumin g) Serum bilirubin h) Glasgow Coma Scale.

2. Age of the patient

3. Chronic Health Evaluation-Was done to know if the patient had severe organ system insufficiency or was immunocompromised. Conditions included are a) Chronic Liver Disease b) Leukemia/Lymphoma c) Acquired Immune Deficiency Syndrome (AIDS) d) Cancer metastasis e) Chronic Renal Insufficiency.

4. ICU admission information-Whether patient was admitted to ICU directly from Emergency Room or was transferred from other ward or some other hospital to ICU was noted.

5. Pre-ICU length of stay-Number of days for which patient was admitted in ward before being transferred to ICU.

6. Emergency Surgery in ICU

7. Re admission to ICU

8. Requirement of ventilatory support

9. Thrombolytic therapy-If the patient was diagnosed to have Acute Myocardial Infarction.

10. Primary admission diagnosis.

The data collected from patients was converted to APACHE-IV Score and Predicted Mortality Rate by readymade software programmed by Mazen Kherallah. (Site-http:/www/icumedicus.com/icu_scores/apache IV.php)

SAPS-II Scoring System was calculated using the following variables-

1. Acute Physiology Points-Both best and worst value in first 24 h of admission to ICU was collected for a) Pulse b) Temperature c) Systolic BP d) Total Leucocyte Count e) Serum BUN f) Serum sodium g) Serum potassium h) Serum bicarbonate.

Single worst value in first 24 h was collected for a) FiO₂ b) PaO₂ c) pH d) Urine output e) Serum bilirubin f) Serum albumin g) Glasgow Coma Scale.

2. Age of patient

3. Type of admission

4. Mechanical ventilation

5. Chronic Health Evaluation-Conditions included were a) Hematologic malignancy b) AIDS c) Cancer metastasis.

The data collected from patients was converted to SAPS-II Score and Predicted Mortality Rate by software programmed by Mazen Kherallah. (Site-http:/www/icumedicus.com/icu_scores/saps.php).

Comparison of both scoring systems was done in terms of Predicted and Actual Mortality Rates. The Standardised Mortality Rate was calculated by dividing Actual Mortality Rate by Predicted Mortality Rate for both the systems.

Data were expressed as mean, standard deviation and percentages. Unpaired student’s t-test and chi-square test were used to analyse the data. Statistical software SPSS was used for statistical analysis. p-value less than or equal to 0.05 was considered to be statistically significant for all statistical comparisons.

Eighty four patients fulfilling the inclusion criteria were recruited as study subjects.

This study showed that majority of subjects were in the age group of 21-50 y (64%). Mean age of the study subjects was 40.60 ± 17.24 yrs [Table/Fig-1].

Out of 84 study patients, 50 were males and 34 were females. Majority of study patients were admitted directly to ICU through the Emergency Room while others were transferred from other wards of the hospital to ICU or were referred to the ICU from other hospitals.

Majority of the study patients had Respiratory Sepsis that was followed by tropical infections [Table/Fig-2]. Out of 84 study patients, 12 had some chronic disease (chronic renal failure, cirrhosis of liver, AIDS, metastatic cancer, multiple myeloma).

The Actual Mortality Rate (AMR) in this study was 60.71% since 51 out of the 84 study patients had expired. Mean of Predicted Mortality Rate (PMR) for APACHE-IV was 37.85% and for SAPS-II, it was 72.36% which shows that APACHE-IV had under-predicted overall mortality while SAPS-II had over-predicted overall mortality. Standardised Mortality Rate (SMR) was calculated by dividing AMR by PMR [Table/Fig-3].

The highest and lowest readings of temperature and heart rate in the first 24 h of ICU stay were found to be higher in non-survivors as compared to the survivors. Likewise, the lowest value of SBP in the first 24 h along with urine output and Glasgow Coma Scale (GCS) rating was significantly low in non-survivors [Table/Fig-4].

Non-survivors had significantly high leucocyte count and FiO when compared with survivors; while survivors had significantly higher serum bicarbonate, albumin and pH as compared to non-survivors [Table/Fig-5].

Most common organ failure in the study patients was cardio-vascular failure followed by respiratory failure [Table/Fig-6].

As the total score of the APACHE-IV Scoring System increased, both PMR and AMR increased. AMR was higher than PMR of almost all APACHE-IV Scores [Table/Fig-7]. SAPS-II Score also had a direct relationship with AMR and PMR. However, for each score, PMR was higher than AMR for SAPS-II [Table/Fig-8].

Sepsis is an infection-induced syndrome characterized by a number of symptoms and clinical signs including fever or hypothermia, leucocytosis or leucopenia, tachycardia and tachypnea [7]. If organ-system failure is associated with the condition, sepsis is considered to be severe.

The mean score of APACHE-IV and SAPS-II of the subjects in this study was 97.42 ± 19.51 and 68.20 ± 19.63 respectively. Mean Predicted Mortality Rate for APACHE-IV and SAPS-II were 37.85% and 72.36% respectively. But the Actual Mortality rate was 60.71%. Thus, APACHE-IV under-predicted mortality while SAPS-II over-predicted mortality of the study subjects, giving a Standardized Mortality Rate of 1.60 and 0.83 respectively.

The measures of the severity of illness (scores) are used as an important tool after admission to predict the outcome of patients. APACHE and SAPS scoring systems are commonly used in judging disease severity and organ failure in critically ill patients.

A multicentric, prospective observational study done in India, to determine the incidence and outcome of severe sepsis among adult patients found the median APACHE II score to be 22. The Standardised Mortality Ratio was found to be 1.40 in the study [8].

A European study was carried out in a single-center Intensive Care Unit (ICU) to validate the SAPS II and APACHE II scores. In this study, the observed hospital mortality of patients with risk of death higher than 60% was overpredicted by SAPS II and underpredicted by APACHE II. This study validated both SAPS II and APACHE II scores in the ICU population which comprised mainly of surgical patients [9].

A study done in a tertiary care hospital in Saudi Arabia showed that predicted mortality by SAPS II and APACHE II systems was not significantly different from the actual mortality. The Standardised Mortality Ratio for APACHE II was 1.00 and that of SAPS II was 1.09 in this study [10].

Another study was conducted in a tertiary care medical/surgical Intensive Care Unit in Saudi Arabia to assess the validity of mortality prediction systems in patients admitted to the ICU with severe sepsis and septic shock. APACHE II and SAPS II scores were included in the study. Predicted and actual mortality rates along with standardised mortality ratio were calculated. Calibration, as tested by C-statistics, was poor for both APACHE II and SAPS II scores. The study concluded that though general ICU mortality system models had accurate mortality prediction, they had poor calibration. However, customization of SAPS II improved calibration. The customized model may be a useful tool when evaluating outcomes in patients with sepsis [11].

As new and effective therapeutic agents are becoming available, it is important that epidemiologic data be updated in order to better understand the incidence and pathophysiology of the disease and to plan the rational treatment of patients.

The manifestations of response to sepsis are usually super-imposed on the symptoms and signs of the patient are underlying illness and primary infection. However, there are individual variations in the clinical presentation. Hyper-ventilation is an early sign of sepsis. Oliguria in sepsis occurs secondary to shock that causes diminished renal perfusion leading to Acute Renal Failure. Insulin resistance and hyperglycemia are invariably present in sepsis. Hyperglycemia needs to be corrected since high blood glucose levels increase the risk of infection, delays wound healing, impairs neutrophil function and stimulates blood coagulation [12]. Anaemia is common in sepsis and results from depression of erythropoiesis.Hematocrit values of less than 30% have been used as an indication of blood transfusion in patients with sepsis with good results.

The Institute for Healthcare Improvement (IHI) has highlighted sepsis as an area of focus and has identified several deficiencies that may cause sub-optimal care of patients with severe sepsis. These deficiencies include inconsistency in the early diagnosis of severe sepsis and septic shock, frequent inadequate volume resuscitation without defined endpoints, late or inadequate use of antibiotics, frequent failure to support the cardiac output when depressed, frequent failure to control hyperglycemia adequately, frequent failure to use low tidal volumes and pressures in Acute Lung Injury (ALI) and frequent failure to treat adrenal insufficiency in refractory shock. Although the mortality rate for severe sepsis is 30 to 50%, this increases to 80 to 90% for septic shock with multiple organ dysfunction [13].

The aims of clinical evaluation in cases of sepsis are-

The measures of the severity of illness (Scores) are used as an important tool after admission to predict the outcome of the patient. Mortality risk stratification in severe sepsis and septic shock is commonly used in clinical trials and in practice, which helps to improve accuracy when evaluating new therapies [14]. By facilitating comparison of the actual with predicted mortalities, the use of such Scoring Systems can provide valuable information about the performance of individual ICUs in treating patients with sepsis.

Severity of illness Scoring Systems are relatively simple to use and widely available. But, there is no ideal system for patients with sepsis. Most of the systems were developed for ICU patients in general and when applied to a particular group of patients, such as those with sepsis, their accuracy declines.

The advantages of an internationally valid mortality prediction system for patients with severe sepsis and septic shock are-

This study has used APACHE-IV and SAPS-II Scoring Systems for risk stratification as outcome prediction models and gives some insight into the issue of the outcome of septic patients from a tertiary care perspective. Customization of these systems to predict the outcome of sepsis is an attractive option.

The present study shows that neither APACHE-IV nor SAPS-II Scoring System performed well in predicting mortality in patients with severe sepsis and septic shock. However, Discrimination (ability to distinguish between survivors and non-survivors) of SAPS-II was better than that of APACHE-IV for patients with severe sepsis and septic shock.